Table of contents

Hydropower is one of leading sources of energy used to meet the primary, secondary and territory ancillary services to balance the energy demand. Hydraulic turbine is essential component of a hydro power plant to generate electricity. The IAHR symposium on hydraulic machinery and systems provides unique opportunity to the academic and industrial research teams to exchange the stat-of-the-art knowledge and ideas on the development of the next-generation of hydropower technology. 31^st Symposium on Hydraulic Machinery and Systems was organized by Norwegian University of Science and Technology, Trondheim, Norway, during 26 June – 01 July 2022. This preface describes the overall summary of the symposium, including the parallel sessions, scientific and cultural tours. For more specific detail, please visit the symposium website www.ntnu.edu/iahr2022. Scientific manuscripts reviewed by the experts, accepted and presented during the symposium are published in a proceeding at IOP Science EES conference series.

All papers published in this volume have been reviewed through processes administered by the Editors. Reviews were conducted by expert referees to the professional and scientific standards expected of a proceedings journal published by IOP Publishing Publishing.

• Type of peer review: Single Anonymous

• Conference submission management system: Conference organizer's internal system - 'eventair'

• Number of submissions received: 123

• Number of submissions sent for review: 123

• Number of submissions accepted: 121

• Acceptance Rate (Submissions Accepted / Submissions Received × 100): 98.4

• Average number of reviews per paper: 2

• Total number of reviewers involved: 56

• Contact person for queries:

Name: Chirag Trivedi

Email: chirag.trivedi@ntnu.no

Affiliation: NTNU

The present paper is an extension of the authors' previous work ^[1], and the goal is to analyze and reveal the complex flow patterns in the pump intake with a low water level (lower than the critical water level) through PIV Technology (Particle Image Velocimetry). The results show that: the lower the water level (a) the geometric scale of the free-surface vortices increases, which potentially leads to pulling more air into the pipe system and the pump units of the pump station; (b) the sidewall-attached vortices have a larger geometric scale; (c) more backwall- and floor-attached vortices are captured, and their geometric scale is large. In brief, the experimental data of the paper indicate that the free-surface, backwall-, and floor-attached vortices are the main characters of the pump intake with a low water level. In engineering, the vortex suppression/eliminating devices in the vicinity of these locations should be emphasized.

The underground pumping station is widely used in many important fields. However, the stagnating flow caused by the sharp geometric gradient at the interface between the primary pump intake and the main pipeline not only causes the performance deteriorations of the whole pump station but also leads to safety and operational stability challenges. In this paper, the numerical model of the related hydraulic phenomena is presented, moreover, the effectiveness and accuracy of the numerical model are verified by comparing it with the experimental results. The simulation results indicate that the hydraulic geometry optimization scheme at the inlet of the main pipe with 0.10 D (inner diameter of the main pipe) local rounding contributes the best effect on the hydraulic efficiency and operation stability of underground pump stations. The results may give a reference for the design of underground pump stations and the optimization of hydraulic geometry in related fields.

Water hammer shortens hydraulic passage lifespan and may cause sudden failure. The primary goal is to use a hierarchical approach to assess the main parameters associated with water hammer. This will help investigate their influence and assist in decision making. Analytical calculation results and a numerical model are compared against experimental data. Our investigations examine water hammer overpressure loading induced by transient regimes. We used data from experimental campaigns carried out within the Hydro-Québec fleet that cover different types of hydraulic turbines and hydraulic passage configurations as experimental dataset. Guide vane closing rate was the main parameter controlled during the overpressure experiments and a general trend was identified for overpressure. This empirical trend is compared to model estimates in order to validate the hypothesis taken into account for calculations. An in-depth understanding of the water hammer phenomenon helps to select the appropriate theoretical model and recommend the optimal operating parameters to extend lifetime and to avoid catastrophic failures. Our study case suggests that available experimental data can be used along with gradually increasing analysis complexity to identify the optimal methodology for a given configuration.

This paper aims to investigate the hydropower station with downstream channel (HSDC) and reveal the effect of fluctuation from downstream channel on its stability and dynamic characteristics under load disturbance. Firstly, a novel mathematical model of HSDC considering the fluctuation of downstream channel water level is established. Then, the stability and dynamic characteristics of HSDC are analysed by Hopf bifurcation. Finally, it is revealed that the effect of the fluctuation from downstream channel water level on the stability and dynamic characteristics of hydropower station. The results indicate that the HSDC is a supercritical system, and the stable domain is on the lower left to the bifurcation line in K_p-K_i plane. Considering the fluctuation of downstream water level, the stable domain of HSDC is wider than the reference where downstream channel water level is constant. The fluctuation of downstream water level has two-sided effect on the dynamic characteristics of HSDC, which makes the attenuation rate lager but setting time longer. Moreover, the more obviously the downstream water level changes with flow, the stronger the influence on stability and dynamic characteristics of HSDC are. Therefore, it is safer to design hydropower station and select the parameters of governor according to constant water level in engineering.

Over past experience on reduced scale physical models of Pelton turbines, it has been noticed that a few models exhibited disturbed performance curves over a given domain of speed factor and discharge factor under specific test head values, while the same performance curves were smooth at other test heads. Cross-checking of different model component combinations helped identify that those disturbances occur with specific nozzles independently from other model components. The experienced disturbances have been understood to be consequences of instabilities attributed to the nozzle.

Attempts to understand root causes and predict such instabilities with physics-based approaches first proved unsuccessful. The proposed research presents the analysis performed on an available model test results database to predict the occurrence of instability thanks to neural network classifiers, the so-called AI-based approach.

A model test campaign including a nozzle known to be subject to instabilities but not part of the database used to train the classifier has been designed and conducted. The agreement between the operating domain with predicted disturbance and the operating domain with measured disturbance is finally discussed.

Accurate models are required for obtaining accurate simulation results. High fidelity numerical models exist, but for the simulations of large hydropower conduit systems one-dimensional (1D) models are still required. For the operation and characteristics of hydraulic machinery the state variables in 1D pipe computation are not necessarily providing sufficient level of detail on what happens inside the machine to accurately predict component and system behaviour. This work is looking into the physics of the spatial distribution of the flow at the outlet of a Francis turbine runner in order to include 2D effects into a 1D analysis. The key finding is a differential equation describing the distribution of the relative velocity W between the flow and the runner outlet, which enables numerical integrating for finding the distribution of W itself, and subsequently the circumferential and meridional components needed to execute the integrals turning the 2D results into functions of the 1D state variables. The presented work is reflecting condition at the best efficiency point, but ongoing work is expanding this to map the entire region of a turbine runner. The approach should be very relevant for future implementations into digitalization schemes such as digital twins

The present work examines the effects of the radial protrusion of four cylindrical rods at different lengths within the flow field of a down-scaled turbine draft tube under part-load operating conditions. Four rods were placed on the same plane 90 degrees apart. The protrusion length was varied from zero to approximately 90 % of the draft tube radius. Time-resolved pressure measurements were performed to quantify the effect of the rod protrusion, using two pressure sensors at the same vertical level 180 degrees apart. Such sensor configuration enabled the decomposition of the signals into rotating and plunging components of the rotating vortex rope (RVR). The results show that different levels of mitigation are achieved for the rotating and plunging components depending on the protrusion length. The effects on the plunging component differ from the ones on the rotating component. The RVR plunging pressure pulsations slightly increase with the initial rod protrusion and then significantly drop after a certain length. On the contrary, the rotating component of the pressure pulsation amplitudes immediately decreases with the onset of rod protrusion. However, an optimum length is obtained in both cases where the highest mitigation occurs before reaching the maximum protrusion. This observation falls in line with the previous investigations conducted for oscillatory rod protrusions, further approving the point that a closed-loop controller should accompany the mitigation technique to achieve optimum mitigation.

Hydro-abrasive erosion in hydraulic turbines is critical and one of the prominent issue due to its association with maintenance costs and production losses in the hydropower plant. IEC 62364:2019* standard guide focuses mainly on hydroelectric powerplant equipment and provides the standard on particle abrasion rates on several combinations like operating conditions, component materials properties, water quality among many factors. With the consideration of different critical parameters, a theoretical model of abrasion rate on hydraulic turbines is proposed by IEC 62364:2019. Present study is conducted to elucidate the several terms used in the theoretical model of abrasion rate for Francis turbine as per the guidelines. The work has taken account into run-off river (RoR) hydropower plant consisting of Francis runner operating in sediment laden rivers in the Himalayan area. Theoretical expected erosion depth for runner inlet, runner outlet, guide vanes facing plates and labyrinth seals is calculated. Characteristic velocities of runner (W_run) and guide vanes (W_gv) were estimated to be 32.26 m/s and 35.05 m/s respectively. Particle load was calculated based upon the sampling data available from the site. Measurement data from field observation during overhauling was used for comparison with the data calculated from empirical relation. For 229 hours operation of turbine, observed abrasion depth varies from 8.1 mm in guide vanes to 1.5 mm in labyrinth ring corresponding to calculated values of 7.53 mm and 1.89 mm for same components. Results shows good correlation among calculated values from IEC and measured values from the site. An optimized solution can thus be devised based on the evaluation of hydro-abrasive erosion along with energy production and maintenance expenses.

*IEC 62364:2019. Hydraulic machines - Guidelines for dealing with hydro-abrasive erosion in Kaplan, Francis, and Pelton turbines. Edition 2.0 (March 2019) International Electrotechnical Commission, Geneva, Switzerland [1]

The spiral casing of a Francis turbine distributes the water from the penstock to the stay and guide vanes circumferentially and uniformly, which is important for achieving the required flow conditions at the runner entrance. Moreover, the spiral casing is supposed to provide the required inlet velocity in front of the stay vanes and create minimal hydraulic losses. The dimensions and shape of the spiral casing depend on the hydraulic and energy parameters of the turbine. A calculation methodology for a Francis spiral casing hydraulic design is presented in this paper. The methodology developed is based on the main condition to achieve a uniform water flow rate into the stay vanes system and the wicket gate over the entire perimeter. The free vortex flow theory is implemented in this research, where the design is based on the law of constant velocity moment. The parametric definition of the spiral casing makes the geometries generated for certain input combinations suitable for numerical analysis using commercially available Computational Fluid Dynamics (CFD) software. The calculation procedure can be used for any set of energy and geometry turbine parameters, such as water discharge, angle of streamline departing from the spiral casing, and stay ring diameter and height. The automated approach integrating MATLAB and ANSYS Workbench capabilities is presented as a spiral casing design tool. The product is a design solution proposed on basis of the turbine parameters. The spiral casing geometry generation is followed by a CFD analysis. Considering input parameters of different existing Francis turbines, the spiral casing is redesigned accordingly. One of the obtained spiral casings geometry is numerically tested. The results show that the uniform discharge distribution is achieved. The automation of the design procedure allows further optimization based on chosen input parameters.

This paper compares the average flow topology in the draft tube cone of a high head Francis turbine operated at full-gate opening no-load (runaway) and speed-no-load (SNL). The comparison is based on the swirl level in the turbine quantified with the angular momentum parameter (RC_u11) and the Swirl number. This study shows that RC_u11 only depends on the flow angle at the guide vane outlet, the distributor height and the runner outlet diameter. The Swirl number has strong limitations in characterizing the flow at runaway and SNL and is unsuitable for no-load conditions.

Sediments flowing with water in the Himalayan Rivers of Nepal erodes the components of the turbine heavily. The design of hydraulic turbines prioritizes the hydraulic performance of the turbine, neglecting erosion challenges. The necessity of turbine design and its model testing has been felt in the past two decades in Nepal. An attempt is made to fill such conditions by developing a test rig for testing the 92 kW model Francis turbine at Turbine Testing Lab, Kathmandu University, Nepal. The model turbine is a scaled-down model of a 4.1 MW Francis turbine of a Jhimruk Hydropower Plant in Nepal severely affected by sediment erosion. The design of the prototype turbine was carried out considering the hydraulic performance as well as erosion resistivity. The prototype turbine was scaled-down utilising model similarity conditions to meet the lab's specifications. Each component of the rig has been optimized using CFD to match the lab's specifications. Comparing the results from CFD and experiment on velocity, pressure, torque, and RPM measurements were comparable.

Pumped-storage power stations which can be flexibly adjusted for power generation and energy storage, are widely constructed worldwide. Pumped-storage units work under different operating conditions according to the demand of the power grid, and vortex ropes in the draft tube will appear and show different shapes under the partial load conditions of the unit in generation mode. The unstable pressure pulsation caused by the eccentric vortex ropes inside the draft tube may cause the vibration of the unit during operation. So it is important to study the characteristics of draft tube vortex ropes for flow and vibration analysis. In this paper, a three-dimensional model of the unit is fully established, and the unsteady calculations of 60 %, 86.7%, and 100 % load conditions in turbine mode are carried out by numerical calculations to obtain the flow characteristics of the full flow field. The pressure and velocity distributions, shapes of vortex rope in the draft tube under different load conditions are extracted from the CFD analyses, respectively. The pressure pulsations of different monitoring points set up inside the draft tube are evaluated in the time domain and frequency domain. The results show that the vortex rope in the draft tube is more obvious under small flow discharge, and the vortex rope itself is not sensitive to the selection of Q criterion threshold, but the vortex on the draft tube wall is greatly affected by the selection of the threshold. The influence of draft tube vortex ropes on the vibration of the pumped-storage unit is also discussed.

Slurry erosion phenomena are inevitable and commonly occurring processes in all types of hydro turbines and components that are exposed to sediment particles. The design of these components and the choice of the materials to resist erosion require laboratory testing using various erosion testing apparatus. These test apparatuses are designed specifically to meet the particular study requirements and to simulate the actual flow conditions in turbines. However, the design of these test apparatus must be flexible and versatile to assess erosion resistance under different combinations of operating conditions. The existing test rigs mostly comprise of rotary type with re-circulation of the slurry. This study proposes a new type of test setup that describes the sediment particles injection method into the system as well as immediate rejection of those particles striking the test specimen. The components of test rig and the test methodology are described. An accelerated tests have been performed on Cross-flow turbine to verify the suitability of the rig for erosion testing in laboratory environment. The design principle of this setup is to eliminate the limitations of the existing test rigs such as slurry ageing problem, use of acceleration tube to maintain the particle velocity and constant operating parameters (discharge velocity and pressure head). The non-recirculation of the slurry is maintained by the use of hydro cyclone as the particle separation device, which is found to have an efficiency of 97%. This test facility provides flexibility in the performance testing of miniature model of different types of hydro turbines with limited adjustments.

The BEM method is extensively used for analyzing the aerodynamic performance of wind turbines and marine propellers. It is computationally fast and is easily implemented while it can give fairly accurate results. Application of the BEM method to predict the forces acting on rotor blades for a model scale axial shaft-driven Counter-Rotating Pump-Turbine (CRPT) is investigated. Some modifications have been proposed to adopt the classical BEM method for CRPT machine and the results are validated against results from Computational Fluid Dynamics (CFD). The results display that the proposed modifications can improve the loading predicted by BEM. However, the improvements are more pronounced in pump mode rather than turbine mode.

Operating hydro turbines in off-design conditions increases the risk of cavitation occurrence, which in turn, leads to numerous problems such as performance degradations, structural vibrations, and most importantly, mechanical damage due to erosion. It is therefore crucial to develop a monitoring system that detects the occurrence and severity of cavitation in real time. For this purpose, a cavitation detection methodology has been developed that is based on the analysis of acoustic emissions of a turbine with machine learning algorithms. In this method, a conventional microphone is used to record the airborne noise emitted from a turbine under different working conditions, and then, a supervised learning algorithm is trained to classify the recorded noise signals into cavitating and non-cavitating categories. The detection system was developed based on laboratory tests and was validated in Ernen hydropower plant located in Canton of Wallis in southeast of Switzerland. This power plant consists of two identical double-flux Francis turbines each having a maximum power of 16 MW and a net head of 270 mWC. The preliminary results obtained from a two-day experimental campaign in the Ernen powerplant are very promising in terms of cavitation detection with a classification accuracy of more than 90 %. The system could be implemented either for real-time monitoring of cavitation occurrence allowing the operators to avoid such a condition or as a post processing tool to evaluate the number of hours a turbine has worked under severe conditions. Work is still ongoing to deploy more complex learning algorithms for this task to minimize expert intervention and/or interpretation during the setup process.

The development of turbine technology faces growing demands to maximize the survival of migratory fish entrained into turbine flows. In this context, two strategies have emerged for quantifying hazardous hydraulic conditions: computer-based evaluations at design stage and recordings with autonomous sensors deployed in prototypes. The former is a desktop evaluation with many modelling assumptions (idealization) and the latter is a field technique that introduces various unknown and uncontrollable factors (uncertainty.) The present work introduces and implements a third method based on test rig measurements and the corresponding computer-based predictions of conditions that negatively affect fish survivability. The experimental work was conducted in a five-bladed Kaplan turbine model in which miniaturized autonomous sensors (SF Mini, developed at the Pacific Northwest National Laboratory, U.S.) measured fish-relevant hydraulic features. The modelling work involved flow simulations according to industry practices and the representation of fish trajectories through the simulated flow conditions. We compared both the experimental measurements and CFD outcomes and discussed the challenges and advantages of the modelling strategies, as well as the benefits for turbine engineers in need of incorporating effective design concepts to mitigate fish mortality through turbines.

This paper presents an experimental and numerical investigation of the internal flow in a Francis turbine draft tube previously designed for minimizing pressure fluctuations and energy losses in off-design conditions. The design of the draft tube geometry is based on an original approach combining Design of Experiments and steady/unsteady Computational Fluid Dynamics (CFD) simulations of the draft tube internal flow. The proposed method provides new insight on the influence of the draft tube geometry on the flow dynamic behaviour on one hand and enables the determination of a geometry promoting flow stability and hydraulic performance on another hand. CFD simulations of the internal flow in the final geometry showed promising results in terms of flow stability compared with the initial geometry designed by conventional CFD-aided methods. A reduced-scale model of the prototype machine featuring the final draft tube geometry is finally installed and tested in laboratory. Tests include performance and pressure fluctuations measurements over the complete operating range. The analysis of the results shows that the draft tube flow remains globally stable over the complete part-load range with pressure fluctuations amplitude lower than 1% of the net head. It is also shown that the dominant pressure component at the runner outlet in the draft tube cone is of synchronous nature. The physical mechanisms of excitation are finally highlighted by analysis of unsteady CFD simulation results.

The main purpose of the paper is to investigate experimentally a new concept by using a free runner downstream of the main hydraulic runner turbine to increase the flexibility in operation. The free runner concept supposes that rotates at the runaway speed with vanishing mechanical torque. The main purpose is to redistribute between the shaft and the periphery the total pressure and the moment of momentum, such that the flux of total pressure and the moment of momentum are not altered. Moreover, the free runner does not modify the operating point of the main hydraulic turbine runner. The experimental investigation focuses on velocity profiles measurements with the LDV system and the unsteady pressure measurements downstream the free runner, in the draft tube cone.

Emergence of new renewable energies, such as solar or wind power, has introduced intermittency on the grid. Hydraulic turbines, initially designed to perform close to their best efficiency point, are more and more at off-design operating conditions, such as no-load regimes, when the runner rotates without energy extraction. These no-load regimes are characterized by complex flow phenomena inducing fluctuations of the blade loading and resulting in a reduced lifetime of the runner. Understanding the origins of pressure fluctuations in no-load conditions is consequently essential to prevent fatigue damage to the turbine. This paper presents the study, with experimental flow visualization, pressure measurements and numerical simulations, of the flow behavior at speed no-load in a medium head Francis turbine model. Evolution of the different flow structures along the no-load curve is then assessed. At speed-no-load, flow visualizations show the presence of a cavitating zone at the blade trailing edge. According to the results of the numerical simulations, this zone seems to correspond to the region where a backflow in the draft tube enters the runner. This interaction results in a backflow inside the runner and in a blockage of the upper part of the inter-blade channel. Sensitivity of this cavitating region to test parameters (discharge and rotation speed) is investigated for different no-load operating points with high-speed visualization at a constant cavitation number. Decrease of the energy coefficient tends to intensify the phenomena and increase the volume of cavitation.

In the present paper the prototypes of small axial hydro-kinetic turbines are designed, built and tested in laboratory conditions in order to demonstrate their operability. The hydrokinetic turbine prototypes have a 0.2 m tip diameter, three blades and are meant to be used in a free water stream flowing with 1 m/s velocity. Two geometries of the turbine runner are studied in order to determine the best correlation between the water flow velocity and the cross-sectional area of the turbine. The 3D designs of the turbine are used for the prototype printing in order to obtain a lightweight compact assembly of reduced dimensions. The experimental measurements are conducted in a closed-loop laboratory setup, specially designed for hydro-kinetic turbines prototypes testing. The flow velocity, runner speed and shaft torque are measured. The aim of the experimental testing is to determine the characteristic parameters for the tested turbines as power coefficient and mechanical power output.

Within the framework of the classical turbine design setup, we present a numerical investigation of unsteady cavitating flow in tandem cascades of guide vanes and rotor blades. The practical relevance of the present investigation is ensured by choosing real blade profiles and domain geometry for the large Kaplan turbines (9.5 m rotor diameter) from the Iron Gates I hydropower plant on the Danube River. The 2D tandem cascade setup is obtained with a conformal mapping of the circular guide vane cascade into a straight cascade, while the rotor blade cascade is by default a straight one for axial turbines. Both single-phase and two-phase cavitating flows results are presented and analysed. The single-phase liquid flow displays rather modest pressure and tangential force fluctuations, with the characteristic blade passing frequency, due to the interaction of the rotor blades with the wakes of guide vanes. The two-phase cavitating flow displays a completely different dynamics in comparison to the liquid flow, with periodic cavitation growth and collapse at a specific frequency lower than the blade-passing one. Moreover, the tangential force on the rotor blade displays severe intermittency and the force fluctuations are almost one order of magnitude larger than in the single-phase flow.

The natural frequencies of a turbine can be calculated from numerical methods. By comparing these natural frequencies with excitation sources, one can know the danger of a resonance and a possible failure in a component of the turbine. Therefore, it is often very important to have an accurate numerical model of the turbine to determine these natural frequencies. There are not many publications on the determination of the natural frequencies of reduced-scale models of Kaplan turbines. More papers exist for pump turbines or Francis turbines. For real Kaplan turbines, very few experiments can be found to determine mode shapes and natural frequencies. In this paper a Kaplan turbine of 37MW (maximum power), 12.5m (maximum head) and 50 m³/s (maximum flowrate) was tested. The turbine was equipped to determine the natural frequencies of the runner in air. For this purpose, one accelerometer in each blade of the runner was installed and a total of 16 impacts were done in each blade. Frequencies and mode shapes were obtained. In parallel, a numerical model was obtained. Numerical and experimental results were compared and an accurate numerical model is presented. With this numerical model the natural frequencies of the runner in water were calculated.

This paper describes flow behavior which causes extraordinary transient stress on runner blade that occurs during start-up of a Francis turbine. Firstly, transient computational fluid dynamics (CFD) was conducted to evaluate hydraulic similarity for the condition of extremely low rotation speed and small guide vane opening. CFD results show that the pressure is below saturation vapor pressure all over the inside of the runner at extremely low rotation speed conditions. Secondly, model test was carried out to measure and evaluate pressure and strain on the blade. The test results revealed that there is a tendency that the phenomenon becomes more severe when rotation speed is lower, when guide vane opening is larger, and when suction pressure is lower. It is concluded that the extraordinary phenomena are not 'extraordinary' but commonly arisen as the transient hydraulic behavior and the impact load causing extraordinary stress on runner blades is related to the instantaneous condensation of vapor bubbles such as steam hammer accompanied by the explosive pressure rise.

This paper describes the mechanism of extraordinary transient stress that occurs during start-up of a Francis turbine by investigating the detailed analysis with measured stress of prototype runner and transient response calculation. According to the dynamic stress measurement with high sampling frequency, it has been clear that a unique phenomenon occurs during start-up especially at the low rotation speed. From the macro perspective, impulsive stress occurred at blade outlet instantaneously and sporadically. Time interval of the event can be considered close to random, and there is no synchronization with the rotation speed. From the micro perspective, the strain occurs owing to the entire vane vibration, and the impulsive loads bump blade outlet from the side of the pressure surface. A Blade shows dumped free vibration in its natural mode in water. As the conclusion, the phenomenon that runner blades are struck by impulsive load on the pressure surface during start-up is compared to a drum beaten by drumsticks randomly in short time, and we named the unique phenomenon "Hydro Drum"

The powerplants in the Himalayan region are generally subjected to hydro-abrasive erosion. The deterioration of components due to erosion causes reduction in performance, changes the fluid flow pattern and even the breakdown of the turbine. In the present work, the Bhilangana-III powerplant Francis turbine is considered a reference turbine for numerical investigation using the computational fluid dynamics (CFD) tool. The study presented an evaluation of hydro-abrasive erosion of high head Francis turbine for the different operating conditions and sediment concentrations. The modified Grant and Tabakoff erosion model is used for the erosion rate calculation. To investigate the effect of various operating conditions, the study is conducted at part load, best efficiency point (BEP) and high load. Also, to examine the impact of suspended sediment on the erosion of turbine components, the solid concentration varies from 500, 1200, 2000, 3000 and 4000 ppm. The observed erosion pattern is similar at all operating conditions for the stay vane, guide vane and runner, but the density of erosion is different. The observed deterioration density due to erosion is higher for part and full load than in the BEP condition. The guide vanes and runner vanes are highly susceptible to erosion due to higher absolute velocity and relative velocity at guide vane and runner vane, respectively. The observed erosion rate variation is qualitatively in-line with the on-field turbine erosion condition. It is also found that the increase in erosion rate with the sediment concentration is almost linear but more than 90% and 79% higher for the runner than stay vane and guide vane, respectively.

At high load, Francis Turbines may experience self-sustained pressure surge leading to significant power swing and pressure fluctuations along the waterway. The physical mechanism initiating this instability phenomenon has been the subject of much research. The development of the axisymmetric cavitating vortex rope at the runner outlet modifies the hydroacoustic properties of the draft tube waterway. Very low wave speed due to high cavitation volume combined with a high swirling number initiates the unstable axial pulsations of the cavitating vortex rope which frequency corresponds to a penstock's eigenfrequency. The 15 MW power plant of Monceaux-la-Virole in France, composed of two units fed by a single penstock, experiences such full-load surge. On-site tests have been carried out to analyze the envelope of pressure fluctuations along the penstock once instability occurs. Combined with a 1D SIMSEN model of the power plant, these measurements have allowed to enhance the understanding of this instability phenomenon. To achieve this, an advanced draft tube modelling taking into account distributed wave speed, convective terms and divergent geometry is used and frequency analysis is carried out. Unstable draft tube eigenmodes and stable penstock eigenmodes are predicted. The key draft tube model parameters such as wave speed and second viscosity are calibrated to set the draft tube eigenmode frequency to the unstable measured frequency for different operating points. This frequency analysis concludes that high load instability occurs when a matching between the draft tube and the penstock eigenfrequencies is experienced. Moreover, it is shown that the unstable draft tube eigenmode is able to interact with different order penstock eigenmodes as function of the operating point of the unit.

The present work investigates the effect of the axial water jet nozzle size on the unsteady pressure fluctuations arising from RVR formation in the draft tube of a Francis turbine operating at part load. The results compare two different nozzle diameter jets for their effect on the pressure pulsations and pressure recovery in the draft tube. Simulations were carried out at four jet injections varying from 2% to 10% of the main discharge. The study also presents the changes in the net effective area and average swirl in the draft tube with the introduction of the water jet.

The draft tube of the hydraulic turbine serves to recover the kinetic energy of the water flow that leaves the runner. The amount of this energy depends on the turbine type, its specific speed and flow capacity. For high specific speed turbines, influence of the draft tube on the turbine efficiency is more essential. The height of the elbow-type draft tube has a considerable influence on its performance. Statistics show that increasing the height of the draft tube leads to a considerable increase of the turbine efficiency, but it also increases the scope and cost of civil works. The present paper presents the solution of the multi-objective optimization problem directed to minimization of the height of the toroidal elbow-type draft tube and maximization of the overall efficiency of the turbine. During optimization, 11-19 parameters that determine the shape and dimensions of the draft tube were subject to variation. In optimization, efficiency was maximized in two operating points: best efficiency point and full-load point. Cavitation qualities were defined as constraints. In order to correctly assess the turbine efficiency and cavitation, the computational domain included one channel of the distributor, one channel of the runner and draft tube; whereas losses in the spiral case and stay ring were determined using empiric formulae. Within the optimization the flow analysis was performed using 3D steady-state Reynolds averaged Navier Stokes equations closed by the k-ε turbulence model. On solid walls the method of wall functions was used. The turbine head, being the difference of specific energies in the inlet and outlet cross sections, was pre-set as a constant value, while the discharge was determined in the course of solution of the problem. A multi-objective genetic algorithm was used to solve the optimization problem. For the draft tube shapes obtained in the course of optimization, their performance characteristics and flow character were compared by means of CFD. It was found that it is possible to significantly reduce the height of the classical toroidal draft tubes without considerably affecting their performance characteristics.

Sediment erosion is one of the most notorious phenomenon damaging the turbines, in the Himalayan region. Francis turbine, which is one of the most used type of hydro turbine in the world as well as in Nepal, is affected quite severely, more so when operated in off design conditions. This study has taken a reference design of a Francis turbine from a hydropower in Nepal to optimize it for operation in variable speeds. Minimization of sediment erosion and maximization of efficiency are taken as the objective functions of the optimization, for which blade angles at trailing edge and blade angle distribution are taken as the design variables. The design space of the runner are constrained such that the optimized design could replace the existing runner in the turbine. Latin Hypercube Sampling technique is used to populate the design space such that the design variables are divided randomly to create required number of designs in the design space. Computational Fluid Dynamic analysis are performed on simplified numerical models of the samples to predict their performance and Sediment Erosion Rate Density (SERD), under various operating conditions. The results of output parameters, obtained from CFD, along with the design variables, are used to develop an approximation model relating the objective functions with the design parameters. NSGA-II optimization technique is used to search for the optimum design. The paper presents the comparison of the sediment erosion and efficiency of the reference runner and optimized runners under various operating conditions. It also presents an outline of the process used to optimize the runner for variable speed operation with minimum sediment erosion.

Accurate numerical models for hydraulic turbine applications are highly coveted. They must be able to correctly capture the swirling flow found at off-design operating conditions in the turbine draft tube. The GEKO model is a relatively fast and flexible eddy viscosity turbulence model with adjustable coefficients to tune the model to different flow scenarios. In this study, the GEKO model is tested on a swirling flow inside a diffusor similar to the flow conditions found at part-load operation of a propeller turbine. The diffusor investigated corresponds to the Porjus U9 draft tube cone section, including the runner cone. Results showed that the near-wall coefficient, with a value of 2, increased the wall shear stress and moved the separation point from the runner cone further downstream. Moreover, with a value of 0.7, the separation coefficient increased the eddy viscosity, which also moved the separation point from the runner cone further downstream. Both coefficients gave velocity profiles closer to experimental values and increased the swirl number at the outlet of the diffusor by up to 36.9 % compared to the GEKO default model. Overall, the near-wall coefficient with a value of 2 gave the best results. The GEKO model provides an opportunity to tweak numerical models to swirling flow.

A small-scale test rig is developed to study the flow phenomena during off-design operations of a Francis turbine. The test rig can be operated in open and closed loops for a maximum head and discharge of 8 meters and 0.055 m³/s. The runner outlet diameter is 200 mm and rotates at 500 rpm. However, the test rig is not a model turbine test rig as per IEC 600193. Therefore, the repeatability of parameters is the key to establishing reliability in measurement and performance estimation. This paper presents the repeatability results of the turbine unit performed at a constant head. A control system is developed in the LABVIEW program with PID (Proportional, Integral, and Derivative) to run the turbine at the constant head with variable discharge and constant discharge at the variable head.

The pumped-storage power station needs to adjust the load to meet the needs of the power grid at any time and cannot always operate under the best working conditions. In practical operation, vibration phenomenon with complex inducements occasionally occurs. In the stable operation of the Baishan energy storage power station, the headcover vibrates obviously. To find the cause of vibration, this paper modeled the unit flow area, used the numerical simulation method for steady and unsteady calculation, extracted the pressure fluctuation information in the flow field, and then analyzed the possibility of strong vibration caused by the instability of the flow field. The results show that there is a low frequency component of pressure fluctuation caused by the spiral motion of the fluid in the sealing groove in the upper canopy clearance flow, which may be the inducer of the vibration of the headcover.

Pump-turbines (RPT) nowadays represents the most common mechanical equipment adopted in the new generation of storage hydro plant. In order to balance the frequent changes in electricity production and consumption caused by unpredictable renewable energy sources, RPT are forced to rapidly switch between the pumping and generating mode also extending their operation under off-design conditions in unstable operating areas. Because of the design criterion adopted for the development of a RPT, an unstable behavior represented by a typical S-shaped profile with a positive slope in the machine's characteristic can occur near to the runaway condition.

With the purpose of evaluating the evolution of the fluid field near to the no-load condition, an in-depth CFD analysis of the RPT model test of the Norwegian Hydropower Center is performed by retracing the machine's characteristic curve running through the flow-speed characteristic curve up to the turbine brake region for fixed guide vanes opening. To validate the numerical results, a comparison with the experimental results in terms of characteristic curves and pressure signals is performed.

The results allow to capture the 3D characteristics of the unsteady phenomena, progressively evolving in an organized rotating stall, highlighting also the influence of the flow rate change from partial loads to the turbine brake operation on their development. In order to characterize the pulsating nature of the instability phenomena developing in the runner and in the rotor-stator interaction, a time-frequency analysis is performed on the numeric pressure and torque signals. The combination between fluid-dynamic and time-frequency analysis makes it possible to identify and characterize three evolution phases: inception, growth and consolidation.

A larger part of the electricity is today from intermittent renewable sources of energy. However, the energy production from such sources varies in time. Energy storage is one solution to compensate for this variation. Today pumped hydro storage (PHS) is the most common form of energy storage. Usually, it requires a large head, which limits where it can be built. In the EU project ALPHEUS, PHS technologies for low- to ultra-low heads are explored. One of the concepts is a contra-rotating pump-turbine (CRPT). The behaviour of this design at time-varying load conditions is today scarce. In the present work, the impact of the startup time for a CRPT is analysed through computational fluid dynamics (CFD) simulations. The analysis includes a comparison between a coarse and a fine CFD model. The coarse model produces acceptable results and is 50 times cheaper, this model is thus used to assess the startup time. It is found that longer startup times generate lesser loads and peak values. A startup time of 10 s may be a sufficient alternative as the peak loads are heavily reduced compared to faster startups. Furthermore, there is not much difference between a startup time of 20–30 s.

The national grid of Nepal faces several hurdles during its operation, especially in remote areas of hilly and mountain region. Citizens face blackouts of days to weeks, due disturbances caused in the grid in those regions. Micro hydro can be a very strategic backup for the national grid in those regions. However, due to lack of subsidy, which used to be provided in the past, for development of micro hydro, micro developers are unable to afford the construction cost of the micro hydro. Thus using Pumps that are readily available in the market at a much cheaper price, compared to custom built hydro turbines, as turbines can be a very economical alternative for such cases. But the concept of pump as turbine has not been utilized in Nepalese market. In this study, a centrifugal pump abundantly available in the Nepalese market has been used to evaluate its performance in turbine mode as well as in pump mode. A characteristic performance curve of the pump operated in turbine mode is obtained from the experimental data. The Best efficiency point and the operating regime is also determined based on the results obtained from experiments. The head conversion factor and discharge conversion factor calculated based on the data measured from the experiment, at various operating conditions, are compared to previous works done by other researchers. The performance obtained from the experimental analysis is comparable to the performance of the turbines installed in the micro hydro power projects. PAT can be a game changer in the micro hydro sector, by providing a very economical and technically viable option of expensive hydro turbines.

The dramatic changes in the internal flow and the corresponding structural behavior during the turbine start-up transient process of pump-turbines are extremely complex. The clearances in the upper crown chamber and bottom ring chamber affect the results of the flow field and structural field of the pump-turbine runner. Most of the previous studies ignored the effects of the clearance flow field to simplify the numerical simulations. In this study, numerical calculations were performed on the entire flow passage of a high-head prototype pump-turbine during the start-up in turbine mode, and the model with and without the clearance are analysed respectively. The causes of the flow field characteristic difference and external characteristic difference caused by the existence of intermediate clearance in the model are studied in detail. The results show that the clearance flow field has a great influence on the axial forces on the runner, which is mainly due to the higher pressure of the clearance flow field compared with the flow field in the runner; At the same time, because the existence of clearance flow field only has a small effect on the distribution of internal flow field, the hydraulic torque of runner considering clearance effects is basically the same as the one without clearance.

Despite the sidewall gaps of centrifugal pumps are tiny narrow spaces, the flow inside them can have a great impact on the flow field in the whole machine. Regardless of its influence, there is no full understanding of the flow in those regions. A theory exists for the simplified cylindrical cavity with a rotating disk, whose application to real hydraulic machines is questionable. To investigate the flow in the sidewall gaps of centrifugal pumps, a test rig including a real impeller has been built up. It enables observation of the flow in the back-sidewall gap by optical methods and evaluation of axial thrust and torque. The measurements were performed for a range of rotational speeds in order to map possible flow regimes and patterns. It was found out, that the basic theory and map of the flow regimes obtained for a simple rotating cavity is inapplicable for the back-sidewall gaps of real hydraulic machines. The flow field is significantly influenced by the impeller blades, and consequently, the map of the regimes is shifted compared to the one derived for the simple cylindrical cavity with a rotating disk. The discrepancies were described and a new map of flow regimes was introduced.

Hydropower is a major renewable clean energy and is widely used worldwide. The reversible pump-turbine unit of the pumped-storage power station is able to work in two main operating modes as required by the power grid: turbine-mode for power generation and pump-mode for power storage. In order to absorb unstable energies such as wind and solar energy and improve the quality of the electricity, reversible pump-turbines need to frequently change operating conditions, and experience more start-stops under different operating modes in a short period. The unstable flow during these transient processes will lead to high-level stresses on the structural components of the pump-turbine units. Therefore, it is of great engineering and academic significance to study the flow characteristics and structural characteristics of the unit during the transient processes. This paper has established a numerical calculation model for a prototype reversible pump-turbine unit, has carried out the CFD calculations of the pump-turbine fluid domains during the pump shutdown transient process, and has analysed the corresponding structural dynamic characteristics of the stationary components of the unit with the fluid-structure coupling method. The pressure variation trend of the spiral case outlet during pump shutdown has the same trend as that of the spiral case domain, and the guide vane flow domain. The maximum flow-induced deformation and stress of the stationary structures have a strong correlation with the axial thrust values of the head cover. The maximum deformation occurs at the inner edge of the head cover, and the maximum stress appears in the fillet of the stay vane leading edge. An increase in the number of shutdowns will result in a higher real risk of fatigue damage to the stay vanes. The conclusions obtained are of great value for safe operation, field condition monitoring, fault diagnosis, and predictive maintenance of the pump-turbine units.

The complex flow field caused by the dynamic stall can affect the operational stability of hydrodynamic machinery. In this paper, the NACA0009 blunt trailing edge hydrofoil is used as the object of study, and the dynamic stall characteristics of the hydrofoil are investigated by using the transition model and the dynamic mesh method. It is found that the hydrofoil deep stall calculated by the transition model is delayed compared to that calculated without the transition model. The hydrofoil dynamic stall can be divided into four stages, initial stage, development stage, stall inception stage and deep stall stage. In the initial stage and the development stage, the lift and drag characteristics are influenced by the shedding vortex. In the stall inception stage and the deep stall stage, the lift and drag characteristics are influenced by the leading edge separation vortex and the trailing edge vortex. The increase of angular velocity and Reynolds number of the dynamic hydrofoil delay the onset of the deep stall while accelerating the boundary layer transition. The research in this paper has a certain guiding effect for the safe and stable operation of hydrodynamic machinery.

High pressure side, as the runner outlet at pump mode and runner inlet at turbine mode, playing an important role in controlling hydraulic loss and flow characteristics in vaneless region, affecting both steady characteristics and unsteady characteristics. While when it comes to three-dimensional inverse design of a pump-turbine blade, scholars are forces on the design method in runner flow channel, ignoring the geometry profile of high-pressure side. Hence, in present paper, concepts "swept", "bowed (lean)" and "twisted" are innovatively introduced, eight new parameters are proposed to control the profile of blade high pressure side. And then a multi-objective optimization design system consisting of geometry generation, computational fluid dynamics, design of experiment, approximation model, multi-objective genetic algorithm, and self-organization map is built. Efficiency at both pump mode and turbine mode and S margin at turbine mode are selected as optimization objectives in the first optimization step. Then, the hump margin is regard as the objective in the second optimization step. Based on the two-step optimization, a runner with optimized high-pressure side is obtained and CFD results show that compared with the original runner, the efficiency at rated points and the margin of unsteady characteristics can be increased. The parametric design method on high-pressure side presented in our paper can be regard as an essential supplement to the traditional design method in hydraulic machinery.

A new storage pump replaced 3 multistage pumps to transport water between the Kreuzeck and the Reisseck power plant sides. Located in the valley of the two reservoirs, it provides impressive performance data for single-stage operation at n = 3000 rpm. Hydraulics already utilised for significantly lower heads are the basis of this pump. The pump was adapted to the new location, and CFD-based development was applied. Within the scope of a model test, the performance data (Q-H-Eta and Q-H-Sigma) were to be confirmed and measured, including the inflow and outflow situation. In addition, the pressure pulsations had to be verified. Finally, installing air vessels should reduce pressure pulsations on the suction and pressure sides. Furthermore, modifications in the bladed areas of the impeller and the guide vane section should enhance this reduction.

The diversity of energy conversion sources in the current energy market increases the demand for stabilizing the electrical grid through ancillary services, frequency and power regulation from the facilities. Pumped-storage units with reversible pump-turbines or ternary sets constitute the currently most advanced solution for providing these services. This implies that pumped-storage units are required to operate in larger head and power ranges than before. In order to increase their flexibility in pump mode, variable speed units have been applied more frequently with additional challenges for the operating range in pump mode.

The extension of the operating flexibility in pump mode makes use of a large portion of the model pump characteristic curve, especially in the case of variable speed units. Important challenges to the hydraulic development are among others the cavitation behaviour, hydraulic stability and pressure pulsation level. This study concentrates on the pressure pulsations in pump mode, which are of great relevance for the machine smooth operation.

The pressure pulsations in a regulated radial reversible pump-turbine with low specific speed are numerically simulated with computational fluid dynamics (CFD) for several points along the complete operating range of the pump characteristic curve. The finite volume model includes the complete hydraulic machine from draft tube to spiral case and hybrid turbulence models were used, in this case scale adaptive simulation (SAS). The numerical simulation offers the possibility to assess different flow quantities at any point of the finite volume model, providing additional data to the experimental model test results.

The integral quantities, e.g. head, flow and efficiency, were compared to the model test results to validate the numerical model. The simulated pressure pulsation amplitude and frequency were also compared to the measured pressure pulsations at the model at the available measuring locations in the spiral case, vaneless space and draft tube cone. After the validation, the computed flow fields were used to derive the pressure pulsation amplitude at other machine locations and components, e.g. at the runner.

The mesh significantly influences the quality of the numerical results in computational fluid dynamics (CFD) problems due to its correlation with turbulence models. In the present paper a structured mesh was generated using Gmsh and two semi-structured meshes were generated using snappyHexMesh to determine the suitable mesh distribution for simulating unsteady cavitation around a plane convex hydrofoil. The numerical simulation was conducted by using the software OpenFOAM with the k-ω SST SAS turbulence model and the Zwart-Gerber-Belamri (ZGB) cavitation model. Besides, the experimental results obtained by the Laboratory for Hydraulic Machines of École Polytechnique Fédérale de Lausanne were used to validate the numerical results. The results showed that both structured and semi-structured meshes predicted the cavitation pattern and the maximum cavity length. The semi-structured mesh with suitable refinement reproduced in detail the dynamic behavior of unsteady cavitation, while the structured mesh efficiently reproduced the phenomenon.

Hydropower plants can play an important role to stabilize the electrical grid when the share of volatile renewable energies increases. This goes hand in hand with a more flexible operation of the turbines and longer operation in off-design operating points. The increased pressure fluctuations that occur at off-design conditions may reduce the lifetime of the turbines. Consequently, this should be considered within the design process to develop efficient but also robust turbines. To derive suitable simulation setups that can be used in the design process, it is crucial to have a deep understanding of specific flow phenomena. As pressure fluctuations are of main importance in terms of fatigue, it is important that simulations can capture them well. Within this study, a deep part load operating point of a Francis turbine at model scale is investigated. By using three different meshes the influence of mesh refinement on the predicted pressure fluctuations is analyzed. Furthermore, by means of single- and two-phase simulations it is investigated how the pressure fluctuations are affected by the occurrence of cavitation. The results show only minor differences between the different meshes, whereas significant differences are observed between a single- and two-phase treatment.

The paper presents a new two-phase flow approach to simulate unsteady cavitating flows. The study applied the in-house code neptune_cfd, which allows two-fluid modelling of liquid-vapor flows [1]. The code solves the ensemble-averaged equations of mass, momentum and energy conservation for each phase (liquid and vapour) for a total of six conservation equations. The equations system requires closure laws for the interfacial terms that represent the mass, momentum and energy transfers occurring between the liquid and vapor phases. An original approach is implemented by using an energy-source term based on the difference between the local enthalpy and its saturation value. 2D and 3D simulations are carried out for a NACA 65−012 hydrofoil with an angle of attack of 6°, a cavity length of 40% of the hydrofoil chord and different flow velocities. The methodology to obtain the cavity length and the main frequency of its periodic behavior is presented. Then the numerical results are compared with those previously obtained by a homogeneous approach [2] and with the available experimental data [3]. The quantitative predictions of the hydrodynamic characteristics (i.e. cavitation sigma number, cavity length, shedding frequency and cavity shape) obtained from two-phase flow modeling appear in better accordance with experimental data.

The cavitation regime has a substantial influence on the damage potential, thus it has to be considered in any specific investigation. For this purpose, we set up a test rig at the Technische Universität Darmstadt using a Circular Leading Edge hydrofoil (CLE) to analyse the damage potential of sheet and cloud cavitation. Exceeding a critical Reynolds number Re_c, the cavitation regime transitions from harmless sheet cavitation to aggressive cloud cavitation. High-speed recordings of the cavitation regime are correlated with high frequency pressure data from a wall-mounted piezoelectric pressure transducer. Spatial and temporal content of the cavitating flow are captured applying proper orthogonal decomposition (POD) to the high-speed recordings. In order to determine the damage potential of the cavitation regime we apply a copper foil on the hydrofoil surface, on which plastic, crater-shaped deformations due to bubble collapses occur. Images of the surface are recorded before and after each run via two-dimensional Pit-Count microscopy. We correlate spatial modes from the cavitating flow field with the eroded surface rate from pitting tests leading to the result that cloud cavitation associated with increasing cloud size is more aggressive. A power law is identified where pitting rate increases with fourteenth power of the Reynolds number.

In the long-distance water conveyance channels or pipelines of sediment laden flow, due to the change of the flow discharge, there may be different levels of siltation condition, and regular dredging work needs to be carried out. If the water conveyance channels are an open system, it is easier to dredge, but if it is a pressurized pipeline system, dredging is more difficult. Some water delivery irrigation projects of sediment laden flow could not be promoted due to the unresolved problem of siltation. This paper takes a long-distance gravity flow pipeline system as the research object. The distribution branch pipe on the main pipeline is connected to the downstream reservoir for field irrigation. According to the pipe data, the one-dimensional full pipeline numerical model and the three-dimensional pipe calculation model were established respectively. The flow velocity calculation of main pipeline under the condition of simultaneous irrigation of different numbers of distribution branch pipes, and the distribution of different velocities in the pipe was obtained. At the same time, the three-dimensional unsteady flow analysis was carried out to analyse the flow characteristics and siltation patterns in the main pipeline under different flow discharge. It is helpful to predict the silt phenomenon that may exist in the water pipeline in advance by calculation, and give technical suggestions for the pipeline design.

The present study introduces the concept of mitigating pressure pulsations in a hydro-turbine draft tube. The concept refers to using an adjustable guide vane system in the draft tube. The adjustability relates to its ability to rotate around an axis. The test rig for the experimental study is a high-head Francis model turbine. Three sets of guide vanes are distributed evenly circumferentially in the draft tube. Each guide vanes consists of two hydrofoils. The upper hydrofoil can move around an axis. The lower hydrofoil is fixed. The turbine operating head for the experiments was 12 m. The operating condition considered is at part load, for Q/Q_BEP = 0.71. The results indicate that using the guide vanes in the draft tube, the plunging mode of the rotating vortex rope becomes insignificant for nearly all upper hydrofoil configurations considered. The reduction in the rotating mode of the vortex rope is between 50% and 80%. The vortex rope frequency shifts from 0.307·f₀ and varies between 0.33·f₀ to 0.617·f₀, which is a function of upper hydrofoil angles

The paper presents the numerical simulations of the flow inside the draft tube of Francis-99 turbine at the part load (PL) operating condition. The rotating vortex rope (RVR) is a phenomenon that occurs during the PL operating regime inside the draft tube of hydraulic turbines. To reduce the computational cost, the numerical simulations are carried out in two steps. Firstly, steady state numerical simulations are performed in a reduced geometry of the runner which is made of a runner passage and part of the draft tube. The velocity profiles from the steady state simulation are used as a boundary condition for unsteady numerical simulation on the inlet of the full draft tube geometry. The velocities from the numerical simulations are time-averaged over a period of 5 RVR rotations and validated with the experimental velocities averaged over the same period. Further, a two-dimensional (2D) linear global stability analysis is performed on a plane extracted from the cone of the draft tube using the time-averaged flow. The frequency of three-dimensional (3D) flow simulation and of the 2D stability analysis are found to be in good agreement with the experimental frequency.

Hydropower plants equipped with Francis turbines may enter unstable conditions at high turbine discharge [13], with strong pulsations of power output and internal pressure. The cause are oscillations of a vapor-filled cavity in the vortex flow downstream of the turbine runner, interacting with the draft tube and penstock pressure. The paper describes the structure and simulation results of a 1D distributed-parameter model for full-load surge, parametrized from results of a high-resolution unsteady two-phase CFD simulation study (Wack [14]) of high-load pulsations in the reduced-scale model of a medium-specific speed Francis turbine. In earlier research on the same model turbine, Müller [12] had suggested that runner blade cavitation could cause instability. To clarify this, variation of turbine intake flow was sup-pressed, and the influence of pressure variation via runner blade cavitation on the relative runner exit flow angle and angular-momentum flux was simulated. Due to the upstream boundary condition, the influence of mass-flow gain χ is replaced by a cavitation gain factor ψ, likewise delayed by the limited speed of swirl propagation. It is found that gain factors obtained from steady-state simulation cannot be directly used with a lumped-parameter model because the swirl effect is much reduced due to its phase changes along the vortex; using the steady-state gain factors would thus exaggerate the influence of swirl on cavity volume. Simulation in frequency domain with realistic swirl transport delay confirms that instability can occur because in some parts of the cavitation zone the swirl provides oscillation power.

In this experimental study, a passive flow control concept is provided to extend the stable part load operation regime by shifting the rise of the precessing vortex core (PVC), also known as vortex rope, toward lower mass flows respectively deeper part loads. A parametrized runner crown design working as a passive flow control device is derived. This control device aims for shifting the bifurcation point of the PVC to lower flow rates. To determine the most influential design parameters and derive an optimized runner crown design, a design of experiments (DoE) approach is used. This DoE approach is based on data obtained from differential pressure sensors inside the draft tube wall of a generic hydro turbine test rig using air as working fluid. By means of stochastic modeling, the growth rate of the PVC mode is derived from the statistics of the measured pressure signals. The growth rate is used to estimate the bifurcation point of the PVC characterized by a certain normalized flow rate. It is shown that the stable part load operation regime is extended by up to 25% due to the passive impact of the modified runner crown. Moreover, the operational range featuring considerable PVC-induced pressure oscillations is diminished and the pressure recovery of the draft tube is improved compared to the baseline case.

Stochastic modeling and local linear stability analysis (LSA) is employed to predict the onset of the precessing vortex core (PVC) in the hydro turbine model. The method of the stochastic modeling based on the pressure fluctuation signals correctly predicts the instability of the azimuthal mode m = 1 at flow rates below 0.7Q_c. This is in line with local LSA that shows that the azimuthal modes m = 1 and m = 2 are absolutely unstable below the flow rate of 0.7Q_c. The absolute instability of mode m = 2 is a new observation in the part load regimes of hydro turbines and plays a significant role in the dynamics of the PVC. As demonstrated in this paper, local LSA and stochastic modelling are both methods to uncover the driver of the PVC using sparse experimental data stemming from either spatially resolved but non-timeresolved PIV snapshots or single-point time-resolved wall pressure recordings, respectively. This makes these methods suitable to be applied to configurations of industrial relevance.

The precessing vortex core (PVC), also known as vortex rope, in a draft tube of a Francis-99 hydro turbine is investigated. The goal is to increase our comprehension of the root of the PVC in order to attenuate or suppress the PVC, thus extending the stable operational range below the best efficiency point at part load conditions. Unsteady Reynolds-averaged Navier– Stokes simulations are conducted and used as a basis for all the analyses performed in this work. The discrete Fourier transform (DFT) and the spectral proper orthogonal decomposition (SPOD) as data-driven methods and the linear stability analysis (LSA) as a physics-based, operator-driven method are used to examine the PVC in detail. With the DFT and SPOD, two dominant modes are found inside the draft tube. Likewise, the LSA reveals two distinct linear instabilities of single-helical and double-helical shape, which agree with the findings of the SPOD in terms of spatial shape and temporal frequency. A particular focus is laid upon the region upstream of the draft tube. An adjoint-based sensitivity analysis reveals that both instability modes are highly sensitive to mean flow modifications inside the transitional segment between runner and draft tube, such as induced by passive control devices. The knowledge of these sensitivities will guide to an optimized runner and draft tube design for controlling the PVC and the double-helical mode.

The paper revisits the theory originally published by Hall that explains the presence of adverse pressure gradient at the axis of a diffuser with swirling flow. It presents the fields of the axial pressure derivative obtained by CFD simulations of flow in a swirl generator, which support the theory. The last part brings cautionary results regarding the suppression of the backflow caused by this adverse pressure gradient by water jet injection. On a case with a strong swirl intensity leading to a strong vortex rope, it is shown that the injected jet may be deflected to the outer walls and rotate with the outer swirling flow, eventually leading to stronger pressure pulsations than in the original state. Further investigation shows that high enough diameter of the jet is needed to ensure its stability.

Radially protruded solid rods and their interaction with the rotating vortex rope at part load condition are investigated numerically on an axial model turbine.

The commercially available software ANSYS CFX was used to perform the simulation, and the test case was the Porjus U9 Kaplan turbine model operating at a fixed runner blade angle at part load condition. Four rods, with a rod diameter equal to 15% of the runner diameter were evenly distributed in a horizontal plane in the draft tube cone and protruded to a length set to intercept the RVR in its unperturbed trajectory. It is shown that the RVR plunging (synchronous) mode is completely mitigated upstream and downstream of the protruded rods. The RVR rotating (asynchronous) mode is reduced by 47% and 63% at the two monitor positions located upstream of the protruding rods, while only a minor reduction occurs to the first RVR harmonic at the monitor positions located downstream of the protruded rods. The perturbed RVR experiences an increased angular velocity due to the flow area decrease caused by the protruding rods, thus increasing the RVR frequency by approximately 53% compared to the unperturbed value. Investigation of the swirling flow indicates a locally increased swirl in the center of the draft tube downstream of the protruded rods which could explain the reduction of the RVR pressure amplitude. The overall turbine efficiency with solid rods protruded causes a marginally efficiency reduction of 0.85%. However, as the RVR pressure pulsations are reduced significantly, a more comprehensive investigation of the rods impact on the turbine performance and life time should be performed to elucidate the suitability of using solid rod protrusion for RVR mitigation.

At present, the measurement of internal characteristics of hydraulic machinery runner (blade pressure pulsation, pressure field and dynamic stress) and blade torque is difficult to be realized by the existing wired measurement methods due to the measurement object is a high-speed rotating part. This paper will introduce the system composition and application in hydraulic machinery of the wireless measurement technology one by one. In the measurement of model hydraulic machinery, it is mainly used to measure the rotating part. The sensors work in water, including measuring the internal characteristics of the runner (blade pressure pulsation, pressure field and dynamic stress) and blade torque. The wireless measurement technology that is by laying the corresponding measurement unit at the measured part, the signal sensed by the sensor is collected and stored by the node, and then wirelessly transmitted through the node, and the data is received through the wireless gateway and enters the computer acquisition system.

Overload instability is a self-excited phenomenon that occurs in Francis turbines working over the Best Efficiency Point. It provokes huge power swings and pressure fluctuations in the hydraulic circuit. One particular issue is that this phenomenon appears suddenly and just before its onset the machine can operate in a very stable manner. In this study, we show that artificial intelligence techniques such as Neural Networks can be used to evaluate the risk of overload instability several seconds before its appearance. Experimental data, acquired during several overload instability tests in a huge prototype, has been used. The techniques proposed in this paper could be used in advanced condition monitoring systems and could permit a safer operation of the turbine working at high loads.

Extreme values of strain signals are important for hydroelectric turbine runners fatigue assessment. However, due to measurements limitations such as short data length and limited subset of measured operation conditions, the extreme values are rarely fully captured. Our study aims to estimate the extreme values of runner strain at non-measured operating conditions by interpolating the extreme components from the measured ones. The method is based on the peaks over threshold technique and the kriging interpolation. A case study with two similar Francis turbines (same design and power plant) is presented. The comparison is made between the use of independent interpolation models and a combined model for the two turbines. This helps assess the assumption that similar turbines share similar fatigue loading. If that is the case, a common model for the whole fleet of similar turbines design in a given facility could be considered, and thus contribute to reduce the uncertainties related to unmeasured loadings without the need for in-situ measurements on every runner.

In the frame of the AFC4Hydro H2020 research project, an extensive measurement campaign has been carried out on a reduced scale Kaplan turbine model at the Vattenfall Research and Development facility in Älvkarleby, Sweden. The objective of the tests has been to monitor and characterize the dynamic response of the machine when it operates in propeller mode with a fixed blade angle corresponding to the best efficiency point. A series of measurements have been taken at steady state operating conditions, such as speed no load, part load and best efficiency point, as well as during transient conditions. The turbine has been instrumented with sensors to measure vibrations, displacements, strains and pressures both off-board and on-board. The excitation due to a rotating vortex rope at part load and the structural response induced by it have been clearly measured by the system. A sub-synchronous frequency with a maximum amplitude has been identified for a given part load condition. The evolution of this frequency with the discharge level and the presence of cavitation has been evaluated. Moreover, the response induced by the speed no load condition has been compared against the best efficiency point. Finally, a turbine start-up sequence has been analyzed in the time-frequency domain.

The present paper aims to improve the applicability of the pressure-time method for flow rate evaluation by eliminating the necessity to cut off the machine discharge completely and taking into account the compressibility of the fluid. As a result, the flow rate determination is possible for any load changes, increase or decrease. The method uses the water hammer equations with the differential pressure measured between two cross-sections. Furthermore, the instantaneous flow rate is analytically linked to the pressure variation and the losses. The method is presented and validated against numerical results.

Hammering method is a common solution to the measurement on natural frequency of solid components. Due to the difficulty to directly measure the natural frequency of a rotating runner in the water, numerical simulation is widely applied to predict the runner behaviours. To verify the influence coefficient of the natural frequency in the water than that in the air, specific measurements are taken on a model runner both in the water and air and a prototype runner by hammering method. Test results show that the ratio of natural frequency in the water to in the air lies in 0.692 ∼ 0.753 on the model runner, and 0.746 ∼ 0.812 on the prototype runner. Practice in this paper shows that hammering method is a helpful way to predict the natural frequency of runner blade in the water, then the operation risk of resonance may be avoided.

The presence of mineral sediments in water led to significant erosion damage to a power plant in the Chilean Andes. The plant is operated by Enel and is equipped with three vertical Francis power units, with approximately 25 MW each operating under a net head of around 120 m. The overall machine condition with reported significant damage on the turbine caused by erosion led to a modernization project with runner replacement and output and efficiency increase. In order to ensure a proper sediment measurement and quantification during the turbine operation, Voith developed, supplied and installed a sediment monitoring system in power unit #01. The monitoring system, developed using the OnCare.Health Hydro platform, applies optical sensors installed directly in the spiral case inlet section to measure the water turbidity. Other operational parameters are recorded, processed and stored together in the same database. Two optical sensors were installed in the spiral case inlet section and provide turbidity readings. The first one is positioned directly flush to the flow. The second one is located in a sampling pipe that collects water from four different points of the same cross section. An automatic water sampler installed in the sampling pipe collects water periodically, which is sent to a laboratory for particle analysis. The laboratory results allow an identification of important parameters such as sediments concentration, particle sizes and materials that are used for correlation with the indicated turbidities. The main data analysis shows that the power unit operated most of the time close to the rated condition. A comparison between turbidities indicated by both sensors shows an offset, probably associated to the particles distribution along the cross section. A system calibration process successfully correlated suspended sediment concentration and turbidity values to allow online indications of the sediment concentration. The system has been operating for almost two years in a close cooperation with the engineering teams of the plant owner and system provider. Main results and system findings for this application will be presented and discussed in this paper.

Hydropower plants operating in fragile mountainous regions face severe sediment erosion of hydraulic turbines, resulting in decreased efficiency, frequent power interruptions, and downtime for maintenance. The sediment size and shape are detrimental to the hydraulic turbines and influence sediment erosion. Sediment management is a crucial aspect of tackling sediment erosion in hydraulic turbines. The traditional method of measuring particle size and shape is time-consuming, labour-intensive, and usually requires statistical analysis. The present study applies the recent technique of dynamic imaging analysis to determine the particle size distribution and shape of sediment. Five different silt-sand mixtures have been used to check the ability of dynamic imaging analysis. Further, the repeatability and reliability of the results were checked by carrying out five repeated measurements of each sample under the same condition in a short period of time. This study will help the researchers and sedimentologists to adopt the dynamic imaging analysis technique for measuring particle size and shape and help the instrument developer in designing a better future version of the instrument.

Condition monitoring is the process of detecting faults in machine components with the help of measurement & analysis of pre identified parameters in a hydro power plant. This paper presents the different problems identified in the selected hydropower plants owned by Nepal Electricity Authority. The power plants have been selected on the basis of type of turbines and total power output. They include all types of turbines that has been in use in Nepal i.e., Pelton, Francis and Bulb turbines. Similarly, they have been selected in accordance to power output i.e., from 1 MW to 144 MW. The questionnaire has been developed and surveyed for the identification of problems in these hydropower plants as well as status of measurement devices. The paper presents the status of different measurement processes, instruments and sensors being used in the selected hydropower plants. The paper further presents possibly of condition monitoring and fault detection process that can be implied in these power plants.

In this study, the flow in the conical section of the draft tube of a propeller turbine has been investigated at the best efficiency point and part-load operating conditions using 2D and stereoscopic 3D particle image velocimetry. Since the flow in the turbine is periodic, it is necessary to study the mean flow field rather than the instantaneous one to identify the flow characteristics from a statistical standpoint. However, the statistical convergence of the obtained mean velocity is questionable. Thus, the current work proposes a methodology for investigating the convergence of mean velocity profiles based on the central limit theorem. The methodology is applied to the best efficiency point and part-load results. The results show that 3D PIV results have lower uncertainty than 2D PIV results because measuring the tangential velocity component affects uncertainty, only measured in 3D PIV. The uncertainty difference is more significant, especially in part-load operation, due to the presence of the rotating vortex rope, and therefore a more accurate measurement is necessary to produce a reliable mean flow field. Furthermore, the convergence of the mean velocity profile is faster, with lower uncertainty for best efficiency point results since, at the part-load condition, the tangential velocity component of the flow is higher. In addition, the converged mean velocity profiles show a backflow region with minor rotation in the center, surrounded by a high rotational axial flow during the part-load operation of the turbine.

As an important index which can be used to evaluate the economic operation result for hydropower station, the water consumption rate of hydropower unit can be acquired with hill chart curve of hydraulic turbine, the relationship curve between flow discharge of hydropower station and water level of reservoir. However, considering that the hill chart curve of turbine is calculated and converted by the model test, there are certain errors when used on the prototype machine, which also brings limitations to the applicability. According to the definition of water consumption rate, it can be determined by the unit's net head and efficiency. The static water head of the unit can be obtained by measuring pressure for spiral case inlet and draft tube outlet. Therefore, the key issue for the measurement of unit water consumption rate is the measurement of turbine flow discharge. Therefore, according to the operating characteristics of high-head turbines, this article adopts thermodynamics to measure the flow discharge of turbine, studies the installation and implementation of high- and low-pressure section test instruments with the site condition of a hydropower project, and analyses of the water consumption rate of units under different conditions such as the multiple-unit's operation, various water heads, etc. The analysis results show the relationship and trend for the water consumption rate and unit operation condition, which can provide references for the optimal dispatch operation of multi-units' hydropower stations, achieving the purpose of reducing the water consumption rate of power generation and improving the efficiency of power generation.

Strain measurements on turbine blades are difficult and costly tasks. Such measurements, when carried out, generally only happen during the runner commissioning. This gives rise to two problems. The first is that some of the sensors often stop functioning properly during the measurement campaign, which leads to distorted data, and the second is that runner blade strains are not available for long-term monitoring after the measurement campaign. To alleviate the consequences of distorted or missing values, we propose the use of neural networks to automate the imputations of missing values in measurement campaign data using virtual sensors. Three types of network architecture are proposed: Long Short-Term Memory (LSTM) in different multi-stage/multi-layer configurations in Nonlinear Auto-Regressive Neural Networks with exogenous input (NARXNN), injector multi-scale attention network (Injector MA-Net), and a combined architecture using both. The performance of these architectures will be compared in four situations: the loss of strain gauge rosette branches; the loss of a complete strain gauge rosette; the loss of data on a complete blade; and the absence of strain data, which is related to the problem of identifying which sensors could be used for long-term monitoring. The performance of the proposed algorithms will be evaluated on real case scenarios from a measurement campaign during a recent unit commissioning.

Swirling flow is a dominant feature in a significant number of technical applications. Hydraulic turbines at part-load are strongly affected by the related vortex rope phenomenon. Its dynamic behavior has a negative impact on the operating performance and durability of the machine.

CFD can be used to get additional insight in this complex phenomenon but requires a valid simulation model able to capture the relevant flow physics, which is driven by highly anisotropic turbulent structures. The simulation results are therefore strongly affected by the turbulence modeling. A swirl apparatus (AC6-14), for which extensive experimental data is available, is used in this work for the assessment and validation of different turbulence models. The state-of-the-art SST k-ω model, with and without curvature correction, is compared to a coupled full Reynolds stress model. All models are integrated into a pressure-based coupled flow solver.

The investigation revealed that both, SST k-ω with curvature correction and the full Reynolds stress model better predict the time-averaged velocity profiles in the diffuser compared to standard SST k-ω. The swirl component is thereby best captured with the Reynolds stress model. All models deliver a reasonable frequency spectrum for the dynamic behavior of the vortex rope. However, flow visualization shows that standard SST k-ω is not capable of predicting the shape and size of the vortex rope accordingly. Both, SST k-ω with curvature correction and the full Reynolds stress model, can be used in the future for more detailed flow investigations, which include also the assessment of flow control measures.

The need to decarbonise our energy system push us toward the use of different clean renewable energy sources, such as solar, wind and hydroelectric energy. It is well known that some of these renewable energy sources suffer of intermittence, which requires the conventional dispatchable power plants to become more flexible to respond to any power change to ensure a stable electrical grid. For this specific purpose, the European project XFLEX HYDRO is focusing on the hydropower sector, with seven demonstrators in three different countries to demonstrate the possibility to enhance the flexibility, one option being to use the variable-speed technology combined with smart digital control. Among these seven demonstrators, one is the Alto Lindoso hydropower plant owned by EDP and located in North-West of Portugal. It consists of two fixed-speed vertical Francis turbines with a high head of approximately 275 m and a rated power of 317 MW. The objective of this demonstrator is to assess the potential of a variable speed using the Doubly Fed Induction Machine (DFIM) technology to increase the flexibility of the power and the lifetime of the turbine. Therefore, this paper explores numerically the hydraulic behaviour of the model runner when operated at a selected part load conditions (60% BEP) with the rotational speed changed by ±10%. Second, a FEM analysis is carried out to evaluate the runner blades damage rate. This preliminary study showed that in term of wear and tear, the variable speed technology would bring in an advantage.

Thinner structures, lead to higher efficiencies, but also to higher mechanical stresses. The increase of the hydraulic performances of Pelton runners is hence limited by mechanical stress constraints. The design process of Pelton runner therefore includes stress and hydraulic assessment during profile development phase. This assessment is based on the combination of Finite Element Method (FEM) for the mechanical stress and computational fluid dynamics (CFD) for the hydraulic behavior of the machine. This combined approach enables the generation of hydraulic profiles with high efficiency while respecting defined limits for the stress. The validation of the combined simulation is achieved by comparison with prototype measurements. For several projects the runners have been equipped with strain gauges in the root and close to the cut-out to record the deformation for different operating points. The comparison with numerical simulation results shows some limitations of the approach but concludes with a satisfying agreement between numerical simulation and measurement.

The very low specific speed pump (nq = 8.9) operated in turbine mode was analyzed. The experimental and numerical studies were carried out in order to show effect of different blade layout on pump-as-turbine (PaT) performance. In total three different impellers were analyzed. One impeller consisting of four main blades and two impellers consisting of four main blades and different arrangement of splitter blades. Either single splitter blade or two splitter blades are placed between each of main blades. While measurements pointed out the main PaT performance, the simulations enable to analyze internal flow fields and point out the mechanisms of performance variation using different impellers. The main aim of this study is to clarify usability of very low specific speed pump for energy recovery in terms of pump as turbine operation or for storage capability in terms of pump-turbine.

The mode split on disc like structures rotating in a dense fluid leads to a deviation of eigenfrequencies at high rotational speeds compared to their values in still water. Predicting eigenfrequencies correctly is essential to avoid fatigue cracks on prototype turbine runners. Analytical models for simple geometric configurations and complex numerical models using fully coupled fluid structure interaction to predict the mode split on arbitrary geometries exist. We are presenting a complementary approach of intermediate complexity applicable to arbitrary geometries. Mode shapes and modal parameters are computed by finite element analysis in still water. These mode shapes are imposed with a harmonic variation in time during an unsteady computational fluid dynamics computation. From the interaction between the flow and the modal motion, the modal force and the modal work can be computed. These can be converted into added modal mass and hydrodynamic damping and further into the shift of the eigenfrequency under rotation due to the fluid for a given mode. The tendencies of the frequencies with rotation compare reasonably well with experimental data. The numerical method can be applied to disc rotation speeds far beyond the range of experimental data revealing interesting tendencies and a phenomenological interpretation of the cause of the mode split.

Recently more start-stops per day of the pump-turbine units are required by the power grid for load regulation. During the turbine start-up and shut-down transient processes, the flow and pressure of the prototype pump-turbine change dramatically at different guide vane opening angles. The extremely unsteady pressure fluctuation in the flow passages can induce large deformation, stress concentration, and strong vibration of the pump-turbine runner. Therefore, it is significantly important to study the unsteady flow characteristics and the corresponding flow-Induced vibration of the runner during turbine start-up and shut-down transient process. In this investigation, a 3D model of the pump-turbine unit including the flow passages of the spiral casing, stay vane, guide vane, runner, and draft tube, crown chamber, band chamber, labyrinth seals, and balance pipes are constructed. The flow characteristics of the unit during the turbine start-up and shut-down process were analysed by coupling the 1D pipeline calculation and 3D turbine flow domain calculation. By mapping the pressure distribution of the fluid domain to the runner structure domain, the flow-induced dynamic behavior of the pump-turbine runner is performed, and the large deformation and stress concentration of the runner are investigated in detail. The flow-induced vibration results achieved are able to provide meaningful suggestions for safe operation and for improving the pump-turbine runner design.

The analysis of unsteady flows is becoming more and more important in the field of hydraulic turbomachinery applications, both for stable and especially for unstable operating conditions and other transient phenomena. Unsteady Reynolds-averaged Navier-Stokes (RANS) methods for incompressible flows, implemented via finite volume or finite element approaches, reach their limit in both accuracy and acceptable computational time. The accuracy mainly suffers due to the strong dependence of the solution on turbulence model variants. Furthermore, incompressibility of the flow and the linked implicit pressure equation makes large eddy simulations (LES) of Navier-Stokes-based methods for such flows prohibitively expensive. An alternative to Navier-Stokes fluid solvers are methods based on the solution of the discrete Boltzmann equation, so-called Lattice Boltzmann Methods (LBM). The purpose of this work is to present first results of an ongoing project involving the simulation of unsteady hydraulic turbomachines at high Reynolds numbers. For this purpose, the ERCOFTAC pump impeller was selected as a test case. Results obtained with the current implementation of the in-house code suggest, that the present methodology could provide a viable alternative for the analysis of unsteady flow phenomena in hydromachines.

The flow rate is a challenging hydrodynamic parameter to measure in order to determine turbine efficiency. The pressure-time method is a cost-effective alternative to estimate the flow rate. Its principle is based on the transformation of momentum into pressure during the deceleration of the liquid mass. Numerical simulations give valuable information to develop the method. One of the challenges in the numerical study of the pressure-time method is modelling the valve movement. In previous numerical studies, the dynamic mesh has been used for valve closure modelling. The dynamic mesh may lead to divergence, and re-meshing makes this method more time-consuming.

In this paper, the valve closure is modelled considering immersed solid method. The results' sensitivity is studied by different time-step, grid, and boundary conditions at the inlet. It is shown that the result is independent of the time-step size smaller than 0.1 ms. Furthermore, it is observed that only the opening boundary condition at the inlet can predict the oscillation of variables during water hammer phenomenon. Then, the numerical results are validated with experimental data. Finally, the pressure-time method is applied to the data, and the flow rate is estimated.

The increased interest for further exploitation of hydropower and pumped hydro storage sites of lower head and capacity, and the tightening on the other hand of the environmental terms and restrictions, would require the design of corresponding hydro turbines and reversible machines with improved environmental performance. This work presents a numerical methodology to optimize the design of a Deriaz turbine in order to achieve high energy efficiency and improved fish friendly behaviour in both pump and turbine operation modes. Various numerical simulation and optimization software and tools are used, while the geometry of the machine is fully parameterized to allow for wide design modifications and corresponding numerical tests. At first, the most important design parameters and their variation range limits are identified by a sensitivity study. Next, several multi-objective optimization procedures are carried out, using general quantitative targets for a comparative evaluation of the various machine runner designs. The results show a complex interdependence or competitiveness of the machine efficiency and the various fish impact performance indices that requires careful analysis. The developed and applied methodology can produce one or more candidate designs that satisfactorily meet all the desired objectives.

A fully automated multi-disciplinary design optimization procedure for a Francis turbine runner has been developed in a previous task of the Horizon2020-HydroFlex project. The design optimization was limited to blade design, with the goal of improving the hydraulic efficiency and torque, and reducing the harmonic structural stresses. This is to ensure that the turbine is less prone to fatigue, but still performs well hydraulically. Results from the numerical optimization are presented in this paper. From the design optimization, two runner designs are highlighted. One that performs significantly better than a reference design, and one that performs significantly worse. It is observed that small, but significant improvements can be obtained in both torque and efficiency, while at the same time reducing the structural stresses drastically. This shows that there might be previously unknown areas in the design space that can be explored, especially on the structural side.

Within the XFLEX HYDRO project, hydraulic short circuit operation (HSC) is being investigated as one of the possibilities to enhance the flexibility of pumped storage powerplants. Therefore, the structural behaviour of a headwater piping bifurcation of a pumped-storage power plant is investigated. Based on unsteady CFD simulations, time history mechanical simulations are carried out to assess the dynamic integrity of the structure in pump (PP), turbine (TT) and hydraulic short circuit operation. Since the plant at hand was initially designed for PP and TT operation, simulating these load conditions delivers a base of assessment for the admissibility of HSC loading. Herein, the focus of the assessment lays on the welded joints. First, a modal analysis of the rock-supported structure is conducted to identify the required numerical analysis method to be employed for the time-history structural simulations. The CFD pressure fluctuations are then applied to the structure within time history FE analyses. The dynamic structural simulations deliver low stress fluctuations which go hand in hand with the low pressure fluctuations previously determined using CFD. Besides, the structure exhibits comparable stress oscillations for all three loading conditions. The latter stress oscillations lie below the fatigue limit of the welds. The study herewith presents promising findings towards safe HSC operation ensuring more flexibility of the power plant at hand.

Variable speed hydropower units offer a large spectrum of grid regulation services and may therefore contribute to the stability of future power supply systems. Full Size Frequency Converters (FSFC) already found real world application in Pumped Storage Hydropower Plants up to a rated power of 100 MW and are even considered scalable up to a few hundred MW. Apart from the extension of the power range and grid regulation capacities, the FSFC technology also provides new control possibilities during transient operations such as start-up in generating mode. Thus, harsh conditions with damaging impact on the hydromechanical components may be avoided by tuning the operating point trajectory in the start-up phase. In this paper, runner fatigue damage during start-up in generating mode of a 5 MW variable speed Francis pump-turbine prototype equipped with a FSFC is numerically analyzed. The fixed speed solution is compared to a variable speed solution following a BEP tracking control strategy. 1D hydraulic transient simulations provide boundary conditions for detailed 3-D CFD/FEA simulations. Full and reduced numerical domains are used and compared. The overall outcome of the present numerical study indicates an important reduction of partial damages using variable speed drives for turbine start-up manoeuvres.

The pumped-storage power stations (PSPSs) with variable speed units (VSUs) have been emerging in recent years, and the research on the transient processes of those PSPSs is of great significance. In this paper, the transient processes of a PSPS with four units, i.e. three fixed speed units (FSUs) and one VSU, sharing one tunnel are numerically simulated under the conditions of large fluctuation and small fluctuation. Firstly, combining the calculation method of hydraulic transient processes and the basic equations of pump turbine, doubly-fed induction motor (DFIM) and governor, a hydraulic-mechanical-electrical coupling model of PSPS is established. Secondly, the data of an actual PSPS is substituted into the model and the corresponding parameters are set. Then, the transient processes of typical conditions of large fluctuation and small fluctuation are simulated respectively. Finally, according to the numerical simulation results, the characteristics of the transient processes of PSPS with three FSUs and one VSU are analyzed. The results show that, under large fluctuation conditions, whether the initial operating states of FSUs and VSU are consistent or not, the transient processes are similar. Under small fluctuation conditions, the VSU can adjust the active power to meet the grid demand more quickly than the FSUs.

High vibrations due to auto-oscillation effects can occur at power units with Kaplan type runner during load rejection. These vibrations may be present in the turbine braking mode when the wicket gates are almost closed, and the rotating runner pushes the water towards the draft tube resulting in higher pressure below the runner blades. The upward leakage flow between runner blades and discharge ring is suspected to cause the vibration phenomenon.

Within this contribution, a fluid-structure-interaction (FSI) technique is applied to identify the mechanism of self-excitation. The investigation is carried out on a disk that represents the simplified geometry of a Kaplan runner. Computational fluid dynamics (CFD) simulations are used in order to obtain parameters in terms of stiffness, damping and inertia of the fluid to be applied in the following shaft line analysis. Damped modal analyses of the shaft line allow for evaluation of the stability of natural frequencies and corresponding mode shapes of the shaft line. A resulting negative damping indicates the occurrence of auto-oscillation behavior of the shaft line. For the operating condition under investigation, mode shapes with negative damping are obtained with this approach and a dynamic bending moment on the runner due to lateral movement in combination with a tilting at the runner in these mode shapes is identified as source of instability.

Based on these findings, an analytical approach for modeling the relevant effect is derived and compared to CFD result as well as results from acoustic finite element modelling. Finally, conclusions for avoidance of self-excited vibrations at Kaplan runners are proposed.

Different methods combining analytical guidelines, numerical simulations, and the manufacturer's experience are employed to design and optimise the distributor manifold of a Pelton turbine. All these methods have in common to assume undisturbed straight inflow to the manifold, thus neglecting the upstream flow conditions. Our numerical simulations executed with the open-source code OpenFOAM v2012 show that the insufficient consideration of the upstream flow situation may lead to inaccurate predictions of the manifold flow. Five turbulence models were tested in our simulations, and the inflow turbulence intensity was varied from 1% to 10%. The flow quality was assessed by evaluating the head loss coefficient from total pressure drop, the head loss coefficient from the entropy production, the secondary velocity ratio upstream the injectors and the mass flow imbalances in the injectors. The study of turbulence models revealed that the k-ω Shear Stress Transport model predicted the head loss and secondary flows with reasonable accuracy. Also, the computationally less expensive model of Spalart-Allmaras leads to similar results and therefore emerges as an appropriate model for optimisation. The simulations with varying inflow turbulence intensity levels indicate a flow pattern change, if the specified inflow turbulence intensity is less than 4%. Below this value of inflow turbulence intensity, a significant increase of the secondary flow upstream of the injectors was observed.

Guide vanes of hydraulic turbines can maximise turbine efficiency and directly influence on turbine operating characteristics. A hydraulic analysis is carried out on geometrically parameterised guide vane foils, to maximise efficiency and potentially observe which configuration gives operating range expansion. Variable-speed operated turbines in hydropower plants tend to run with improved performance at off-design operating points due to the speed variations. A combination of variable speed turbine operation and guide vane shape optimization gives a win-win situation in water to energy usage ratio. Human controlled developed designs and CFD calculations were previously done. The obtained results are compared with automated approach in ANSYS Workbench using Design of Experiments. The numerical tests were made on parametrically developed guide vane cascades, where the turbine efficiency is the evaluated parameter and it is observed in the range of several rotational speeds and calculated flow rates. The cascade design which produces less turbine efficiency deviations for the tested rotational speed range is later exported and hydraulically and geometrically compared with the developed design from the manual approach. The tested configurations of the guide vanes are developed on the basis of Francis 99 turbine model from the Waterpower Laboratory at the Norwegian University of Science and Technology in the existing geometric space for the guide vanes and blades axis positioning circumference.

In the present paper a sensitivity study of numerical and geometrical parameters related to structural mechanical analyses of a hydraulic machine is presented. The sensitivity study is carried out for the blades of two different hydraulic machines, a propeller blade of a submersible vehicle and a runner blade of a kinetic river turbine. The focus of this work is on static structural mechanical analyses, which are performed with the open source finite element program CalculiX. Parameters for the sensitivity study are numerical parameters like the global mesh size as well as the local mesh refinement at certain locations of the blade. Furthermore, the sensitivity of geometrical parameters such as the blade thickness at different locations or the shape of the fillet is investigated. The results of the sensitivity study show that the numerical parameters have a small influence compared to the geometrical parameters. The most dominant geometrical parameter is the blade thickness at the hub. In case of a high load due to the centrifugal force, the influence of the blade thickness at the tip of the blade increases.

In this paper, we investigate the formation of the Rotating Vortex Rope (RVR) using scale-resolving methods, SAS and Wall-Modeled LES (WMLES). We compare the results from these simulation methods with the experimental data of the Francis-99 workshop. This comparison shows that the general features of the RVR can be captured with both methods. However, using WMLES methods would lead to a better quantitative agreement between the velocity profiles in the draft tube in the simulation and the experiment. The reasons for this better agreement are discussed in detail. A comparison of the pressure fluctuations in the draft tube captured in the simulations and the experiment is also presented. This comparison shows that all simulations under-predict the Root Mean Square (RMS) of these pressure fluctuations, although the RMS values predicted by the WMLES simulation are closer to the experimental values.

The operation of pumped storage power plants at off-design conditions requires an understanding of the flow field and the resulting damage to ensure a safe and economic operation. Within this study, CFD simulations are carried out for an existing pump-turbine at deep part load operation to enable calculating the damage in later structural analyses. Furthermore, this study includes the comparison between the SST and SBES turbulence model at different mesh sizes. The simulation results are compared with model test data regarding head, torque and pressure fluctuations at a bottom ring and draft tube cone sensor. Whereas all simulation setups are within a reasonable range to the available model test data, the pressure fluctuations on the runner blades as well as the resolved vortices within the runner blade channels show a qualitative difference between both models. The SBES turbulence model resolves more vortices and thus different pressure fluctuations. Subsequent structural analyses show higher accuracy of the SBES turbulence model compared with the SST model. A mesh refinement from 25 million nodes to 80 million nodes with the SBES model reveals a better resolution of vortices, but only local differences in pressure fluctuations.

The paper focuses on implementing the wall model developed by Manhart, in Reynolds Averaged Navier - Stokes (RANS) turbulence models used in the field of Computational Fluid Dynamics (CFD). This wall model considers the influence of the streamwise pressure gradient in addition to the existing wall models used in the usual CFD codes. In the present work, two RANS numerical simulations are carried out using the k-ω Shear Stress Transport (SST) turbulence model on an asymmetric diffuser geometry. One numerical simulation is carried out using the implementation of the Manhart wall model in the k-ω SST turbulence model, and the other numerical simulation is performed using the standard formulation of the k-ω SST turbulence model. The numerical simulations carried out using the Manhart wall model and the standard formulation of the k-ω SST are compared with experimental measurements made on the asymmetric diffuser experimental installation. The numerical simulations are carried out using a free, open-source CFD tool, Code_Saturne. The comparisons between numerical simulations and the experimental data are in good agreement in the boundary layer of the flow inside the diffuser. The Manhart wall model had a faster convergence resulting in a shorter simulation time.

One of the most interesting operating points of Francis turbines is the high part load point, which is characterized by synchronous pressure pulsations of large amplitude. Physical mechanisms of these pulsations are not well understood. Experimental observations show that the frequency of the pulsations is 1.5 to 4 times higher than the runner rotation frequency and there exists a 180 degree phase shift of pressure signal measured in the spiral casing and in the draft tube. Numerical modeling this operating point is challenging as it has to account several complex phenomena: cavitation, rotating vortex rope and interaction of the turbine flow with the whole water system of the test rig. In the present paper we carried out a series of 1D-3D calculations of a model turbine in an attempt to simulate the high part load pulsations. 1D hydro-acoustic equations were used for the upstream waterways, while 3D Reynolds averaged Navier-Stokes model of two-phase mixture was used for the turbine. Although the main factors were taken into account, these computations were not able to capture the high part load pulsations. Namely, the frequency and amplitude of the pressure pulsations were significantly lower than those in the experiment, and there was no phase shift in turbine domain. In order to investigate the shortcomings of the present model, we considered a problem of propagation of disturbances in two-phase "liquid-gas" flows. Several 2D test cases were numerically investigated using both incompressible and compressible gaseous phase model. It was shown that in the absence of phase transfer the compressible model correctly represents the speed of sound in both homogeneous and stratified flow regimes. From the other hand, in case of phase transfer the model damps out propagation of pressure disturbances in upstream direction. This behavior explains the failure of the present 1D-3D model to correctly describe the acoustic properties of the two-phase flow in the draft tube, playing a crucial role in the development of high part load pulsations.

Within the XFLEX HYDRO project, operating power plants in low load conditions is being investigated as one of the possibilities to enhance the flexibility of hydropower. Therefore, the structural dynamic behaviour of a pump-turbine runner is numerically analysed in deep partial load (DPL) operating points. Based on highly resolved unsteady CFD simulations performed prior to this work, the pressure fluctuations are extracted and mapped onto the structure for subsequent Finite Element (FE) analyses. Herein, an efficient hybrid methodology combining purely structural and coupled Fluid-Structure-Interaction (FSI) modelling is applied. The obtained simulation results are compared to experimental measurements. Herein, the influence of several factors such as the CFD mesh size, the turbulence model, the simulated time window width, and the structural simulation type is studied. As a result, the numerical outcome is validated with good accuracy. The importance of highly resolved CFD simulations with appropriate turbulence modelling is highlighted. Besides, the necessity of coupled acoustic-structural analyses to accurately resolve the dynamic response to excitation by rotor stator interaction (RSI) pressure shapes is demonstrated despite the chaotically dominated nature of the structural vibrations. The results of this work can be used to extend the operating range of the pump-turbine runner and, most importantly, to calibrate numerical models to predict the structural vibrations accurately and efficiently in DPL and finally ensure safe operation of these components.

Due to the volatile electricity received from solar and wind power plants, the energy market requires highly variable hydropower machines which are able to cope with fast changes of their operating regimes. With the need for flexible operation, Francis runners are exposed to various operating conditions outside the traditional operating range. Hence, the runners must be specifically designed for long time part load operation to meet hydraulic and structural requirements. Therefore, it is inevitable to have a reliable but also practical engineering approach in the design phase of the projects in order to ensure sufficient fatigue life of the Francis runners. Deep part load operation mainly consists of stochastic loads acting on the runner blades. Previous publications show that it is reasonable to normalize the Rainflow matrix of stress amplitudes to represent the characteristic behaviour at deep part load operation. Hence, it is feasible to utilize the normalized Rainflow matrix to predict the fatigue damage of Francis runner designs without prototype measurements. Within this paper, a scaling method to derive project specific fatigue results for deep part load is introduced and subsequently validated through an intercomparison of available strain gauge measurement data and prediction. Based on the results, the accuracy of the proposed engineering approach and possible impacts of different boundary conditions are discussed.

In order to satisfy the power demand in the electrical grid, hydraulic turbine units frequently work under off-design operation conditions and pass through transient events. These operation conditions can lead to high vibration amplitudes in the turbine runners, decreasing their useful life, and in some cases to premature failures. To determine and to understand the behaviour of the fluid damping is a relevant topic, because this parameter limits the maximum amplitude in resonance conditions. The runner of some types of turbines, such as reversible pump-turbine and high head Francis turbine, can be modelled as a disk-like structure, due to their similar mode shapes. Because of this, in this work, the fluid damping of a vibrating disk was studied. The disk was submerged in water and was put in a resonant state at different vibration amplitudes. Moreover, this structure was excited at different distances to a rigid surface, in order to analyse the effects of the distance between the runner and the casing. The main effects on the fluid damping were determined and characterized, showing a dependency of the fluid damping ratio on the different parameters.

Facing an increased demand for flexible operation of hydro power plants (HPPs), Francis runners are substantially subjected to operating conditions beyond the traditional operating ranges. In order to ensure a reliable service of major HPP components covering their complete service life, a fundamental understanding of both load and material fatigue behavior is essential. Due to the cyclic loading in water, reliable corrosion fatigue data is required for the runner fatigue assessments. In previous publications, a state-of-the-art corrosion fatigue testing campaign within a collaborative German research project was described and results for a common cast steel used for the manufacturing of hydro runners were presented. The material testing program for the 13 % Chromium and 4 % Nickel steel has advanced since then and important aspects beyond varying water qualities and different stress ratios have been investigated. Within this conference contribution, corrosion fatigue investigations that were carried out on cast, forged and welded nickel-martensitic steels are described and recently gained results are presented.

In this work, the complex structural response of the Francis-99 turbine runner is further investigated by decomposition of the output signal from previous Laser Doppler Vibrometry (LDV) measurements into the motion of each nodal diameter. During the structural measurements, the non-rotating runner was installed in the turbine pit, submerged in a non-flowing water, and excited with piezoelectric patches mounted on the hub. The patches were excited with phase shifted sinusoidal voltage to create overall excitation of the runner with a desired number of nodal diameters. The deflection of selected locations on the trailing edges were scanned with LDV, one point at a time, and the global movement was reconstructed by combining the data for all points.

The Francis-99 runner has its blades bolted to an over-dimensioned hub and shroud, where the hub is not fully axi-symmetrical and has several hollows in it. This, together with the fact that one patch was found to be non-functional, is believed to have excited other ND patterns in addition to the one that was intentionally excited, therefore contaminating the movement of the trailing edges with movements that does not belongs to the excited ND. To mitigate this and create a better representation of the movement of the trailing edge, which is not affected by the bleed from other ND, the LDV signal for each excited frequency of a particular ND is post-processed using discreet Fourier transformation to decompose the motion of each nodal diameter in the range ND0 to ND7. This unveils the contribution of each nodal diameter within the output signal where a spike is seen for the excited ND in all measurements. Influence from other nodal diameters were found, where the failed patched is believed to cause a ND1 like movement. In addition the clustering of multiple eigenmodes with differing nodal diameters previously found in narrow frequency bands were also found as interfering contribution when exciting at the relevant frequencies.

Dynamic models for hydropower generators treat the rotor part as a rigid body; however, many studies have shown the opposite. The electromagnetic force distribution of deformed rotors is uneven, creating Unbalance Magnetic Pull, causing high forces on generator components leading to a risk of fatigue, therefore shortening the life of machines. Unbalancing masses can worsen the asymmetries of the rotor, which would further increase the effect of the electromagnetic interactions. This paper evaluates the rotor response using different unbalancing masses at the rotor and at the poles to quantify their impact in displacements and exciting frequencies. The model employed in this paper is based on the equation of motion derived using Lagrange equations in both co-rotating and stationary frames of reference, considering the effects of Centrifugal loads, Coriolis, and magnetization of poles. Different unbalancing mass placements affect different variables; extra weights in the poles contribute predominantly to the deformation of the rim, while the unbalance in the shaft affects the position of the shaft; a combination of placements was also studied. The simulations were performed and compared with and without radial electromagnetic forces, showing how the presence of magnetized poles further deforms the shapes of the rotor.

Methods used to calculate the eigenmodes and eigenfrequencies of a Francis runner with the added mass from water are well established and have been verified by several experiments. The common explanation of the water's role is the added mass it brings to the runners structural eigenmodes and thereby reducing the eigenfrequencies. In this paper we calculate the first eigenfrequency of the runner from the acoustic properties of the runner channel and compare the result with full FEA calculations. Using this method we show that the local speed of sound for the water in the channel can be as low as 200m/s and that the density of the runner is not necessarily an important parameter for the first eigenfrequency with some Nodal Diameters of High Head Runners.

The operating modes of pumped storage power plants, such as that of the Forces Motrices Hongrin-Léman (FMHL) in Switzerland, have drastically changed over the past decades due to the emergence of new renewable energies. The number of starts and stops experienced by machines increased significantly leading to broader fatigue cycles on the different parts of the machine. Therefore, the development of predictive maintenance tools is paramount to increase the availability and reinforce operational safety of the power plant. In this context, vibration measurements were performed on a 60 MW Pelton turbine of one of the ternary units of the FMHL power plant during dynamic and steady state operations. Non intrusive instrumentation has been deployed including accelerometers on the bearing. Comparing the ramp-up and ramp-down of the Pelton turbine, a hysteresis of the vibration level for a similar power has been observed. A reduction of the vibration levels when an additional injector is engaged during the ramp-up is observed. This reduction correlates strongly with the load reduction on the buckets when the total flow rate is distributed over more injectors, leading to a smaller flow rate per injector.

This paper presents the design and evaluation of an active control system to reduce vibrations and strains of submerged structures. For that, a metallic disk mounted in a test rig has been equipped with on-board waterproof strain gauges and accelerometers, to measure its structural response, and with an on-board PZT actuator patch, to provide the required control force. This control force is computed in real-time by an optimal algorithm that determines the exact voltage to be supplied to the PZT patch in order to mitigate any increase of vibrations and/or strains detected by the sensors. In order to capture the dynamics of the structure when it is excited both in air and submerged in water, a plant model has been identified from the Frequency Response Functions (FRFs) obtained experimentally from the measured vibrations and numerically from simulated strains. With these plant models, two different control techniques based on a Linear Quadratic Gaussian (LQG) controller and on an H_∞ controller have been implemented and simulated under different types of excitations. The obtained results are satisfactory and support the need to continue the research to validate them experimentally.

The flow pattern in the draft tube is commonly considered when investigating rotor vibration problems, especially for Francis turbines. However, phenomenon in the vaneless space for Kaplan turbines have been shown to induce rotor vibrations when operating in non-standard conditions such as speed no load (SNL). These flow disturbances develop in the vaneless space at small guide vane openings and high swirls. Depending on the number of flow disturbances and runner blades they can excite transversal, axial or torsional vibrations. In the case of transversal excitation they can in some cases cause resonance problems if the frequency is close to a natural frequency of the rotating structure. Recently it was observed that the excitation causes severe resonance in the rotating structure on a refurbished turbine when operated at SNL.

The present paper presents frequency analysis of the flow-induced excitations coupled to the shaft vibrations on an old and a refurbished Kaplan turbine. To investigate the causes of the resonance problem, measurements have been performed on both the model and prototype turbine and on a twin unit with the old runner design. The measurements on the prototypes consist of pressure, shaft bending moment, and vibrations. Pressure and radial forces were measured on the model.

The result shows the dependence of the runner blade opening on the transversal excitation and the frequency of the flow disturbances. The rotordynamic analysis showed the same phenomena with forward and backward precession. The results agree well between the model and prototype when measured data are normalized by the runner rotational frequency. It is proposed, when refurbishing a unit, to use measured pressures, shaft bending moment and vibrations on the prototype before the refurbishment and the new model turbine to identify critical excitation frequencies for the new prototype turbine, i.e., use the information from the tests to set rotordynamic requirements for the new prototype unit.

After 30 years of operation, the runners of the four 100 MW high-head Francis turbines successively developed fatigue cracks starting from the hub-side trailing-edge fillets. Assessment of alternating stress in normal turbine operation revealed that stress levels were not nearly large enough to explain the damages. Consequently, a test campaign comprising start and stop manoeuvres in addition to normal turbine operation was conducted in 2017. Strain gauge signals obtained on-board of the runner blades were recorded together with casing vibration and acoustic signals. Intense high-frequency vibration centred around 611 Hz was detected in three transient conditions of operation: at speed-no-load (SNL), during coast-down and during back-to-back pump start. Common properties of these conditions were very low to no discharge, strong secondary flow in the runner and speed between 50 and 100 %. Due to the extremely narrow bandwidth, three possible ways of excitation were considered and checked: rotor-stator interaction (RSI), rotor-dynamic instability and Von Kármán excitation. RSI was excluded due to speed-independent frequency. Strong sidebands in the stationary-frame frequency spectra - obviously caused by runner shroud deformation shapes ND1 through ND5 - excluded any feedback from runner side chamber pressure; thus leaving Von Kármán excitation as the only possible diagnosis. Accordingly, a very safe profile was selected and implemented for the runner blade trailing edges. Subsequent field-testing at two modified units confirmed that the vibration problem at the harmful operation points had hereby been eliminated.

To accommodate renewable energy production and load demand variability, hydropower plant owners need to increase their operating range regardless of their units' original design load envelop. When this increased operating range is an issue for the fatigue life of the old runners, solutions need to be found with the design of a new runner to sustain those new challenging loads of the increased operating range. In recent years, many papers have been published to show the challenging loads on Francis runners at speed-no-load and deep part load conditions. Andritz demonstrated a good numerical prediction capability for stress levels at deep part load conditions for Francis runners. However, for axial units, very little has been published. Very recently, some papers showed good predictability by CFD of the flow behavior at deep part load including the vortices present at those conditions. This paper demonstrates the prediction capability of the numerical tools by comparing strain gage measurements on an axial runner to CFD-FEA stress predictions. The measurement campaign was conducted conjointly by the unit owner and the manufacturer for research purposes. In the deep part load operating zone under the effect of columnar vortices, frequency analysis of the measured vibrations and strain gage signals confirmed the flow behavior predicted by CFD, and the measured dynamic strain amplitudes were well predicted. Numerical prediction of dynamic stress in the complete range from 0-100% power of the measured unit as well as detection of high vibration zones was successful.

Downward migrating fish can be damaged when passing through turbines in the direction upstream to downstream. The overall influence on the fish population is not known yet. In recent decades, many publications with steady-state CFD calculations for the passage of fish have been published. However, transient CFD calculations show that the range in which fish pass the runner itself is not easily predictable but rather random, influenced by the current position of the runner, thus affecting the probability of survival. The pressures, which are also well represented by steady-state CFD calculations, are compared by applying "Live Fish" (typical types for the middle of Europe) as well as "Barotrauma Detections Sensors (BDS)" which were exposed to the area of the turbine intake at different heights and monitored during downstream migration.

Surge tanks are required to enable power and frequency governing of hydraulic machines in hydropower plants with pressure tunnels. To improve the hydraulic performance and reduce the construction costs, a new concept of semi-air-cushion surge tank (semi-ACST) is proposed. The semi-ACST can be particularly effective for upgrading and retrofitting of existing hydropower plants, when additional hydraulic machine capacity is added to an existing waterway. The key design element is a crown throttle, constructed as an inverted weir placed in the crown of the chamber. The purpose is to intentionally trap an air pocket during filling of the lower chamber. The air is released slowly through defined air pipes so that the air in practice contributes equivalent to added volume in the upper chamber. Thus, the semi-ACST improves the dampening of mass oscillation, without increasing the surge tank volume. The semi-ACST has been investigated and developed with multiphase 3D CFD simulations with RANS turbulence modelling. The design principles have been tested on the case-study Tonstad hydropower plant (960 MW). In the presented case study, the semi-ACST is proposed as an extension of one of the three exiting hydraulically coupled surge tanks.

Hydropower plants (HPPs) are acknowledged to be fundamental assets to provide generation flexibility to power systems hosting substantial amount of stochastic renewable energy sources. Within this context, providing this flexibility enforces HPP units to cross-over transient operating conditions or to operate in off-design conditions. The evaluation of the impact that such transients have on the hydraulic machines is a fundamental step to assess the operating costs in terms of lifetime reduction of the machine structural integrity.

When these assessments are performed on experimental test-rigs featuring a machine's reduced scale models, testing transient regimes may potentially trigger harmful pressure profiles along the piping system as well as resonant phenomena on the test-rig. Simulating how the several state variables evolve during the experimental tests allows for the prediction of damaging circumstances and, eventually, prevent performing experimental tests in detrimental conditions.

In this regard, this paper illustrates a numerical method able to reliably reproduce the experimental test-rig behavior while preserving the consistency of the scaled similarities with respect to a specific case study HPP: the Veytaux I pumped-storage hydropower plant located in Switzerland.

Moreover, in this paper, the developed 1D numerical model of the EPFL Technological Platform for Hydraulic Machines experimental test-rig is presented since some of the characteristic transient sequences of the studied HPP are simulated on this model and compared to the prototype scale 1D numerical results.

Variable speed pumped storage unit (VSPSU) is an innovative technology, and investigations regarding operating characteristics of VSPSU are significant. However, there are relatively few studies towards hydraulic disturbances of VSPSUs. The hydraulic disturbance means that pumped-storage units, which were originally in steady operation, are inevitably affected by suddenly operating changes of other units in the same waterway system. Normally, it will cause the power output of fixed-speed units to oscillate, and adversely affect the safety of generators and stability of power systems. Hence, investigations of VSPSUs' characteristics under hydraulic disturbances are necessary. In this paper, VSPSU using doubly fed induction machine (DFIM) is adopted as a study case, and fast power control strategy is used in the controlling of VSPSU. The simulation result shows that VSPSU has good anti-interference performance, and normally operating VSPSU can still maintain stable power output and stator current even under disturbance from other units. However, despite the good anti-interference performance of VSPSU, the mechanical power of VSPSU lags far behind the electromagnetic power, and the hydraulic-mechanical transient processes of VSPSU are similar to that of a fixed-speed unit. The findings in this paper could provide understanding and guidance for practical operation of VSPSUs.

With specific reference to the HPP start-up sequence, the conventional process requires to operate the runner in a particular operating regime called Speed-No-Load (SNL) that is detrimental for the runner. Thanks to last decade's improvements in MW-class power electronics, new generators' technologies have emerged providing an additional degree of freedom to HPPs control. In this respect, the control of the hydraulic turbine speed can be used to decrease the level of damage induced on the runner by choosing the appropriate trajectory for transient operations. Variable speed is a recent technology and its capability to improve the transient operation of hydraulic machines is not fully addressed in literature and, specifically, for start-up sequences. To fully take advantage of such an added degree of freedom, an in-depth experimental study has been performed at EPFL (PTMH) on a specific speed homologous reduced scale model of the unit 5 of Z'Mutt HPP equipped with a reversible pump-turbine. The paper presents a methodology to (i) investigate different transient sequences to start-up the hydraulic machine, which have been tested on reduced scale model, and to (ii) evaluate the equivalent damage impact on the runner. The results of this study highlight the advantages of leveraging the variable speed technology during the unit start-up.

In the framework of the XFLEX HYDRO H2020 European Project, the pumped-storage power plant of Grand'Maison (France), owned by Electricité De France, focuses on the implementation of the hydraulic short-circuit (HSC) operating mode. This mode increases the flexibility in pumping mode, which helps the integration of intermittent energies. Grand'Maison is divided into two power houses: the first features four Pelton turbine units and the second eight reversible pump-turbines units. A trifurcation splits the flow into three penstocks, each is then split into two branches that feed each power house. The HSC operating mode, which consists in operating the pumps and the Pelton turbines simultaneously, changes the flow paths in the junctions compared to the pump mode. The power plant was not designed to operate in HSC mode over a long duration, therefore an assessment of its feasibility is necessary. 151 computational fluid dynamic simulations are carried out for two bifurcations and one trifurcation. The numerical simulation results show that the local head losses in HSC mode represent less than 1% of the gross head. No flow instabilities are observed at the bifurcations contrary to the trifurcation. Additional analyses are required to better understand the flow in the trifurcation.

The objectives of the 2050 energy policy based on the decarbonization of the electric power networks generate drastic changes for grid balancing with a massive integration of non-dispatchable Renewable Energy Sources. Hydroelectric power plants already significantly support electricity power system flexibility with innovative solutions such as variable speed units, fast frequency control, fast generating to pumping modes transition, high ramping rate, inertia emulation, etc.

For pumped storage power plants (PSP), a quick solution to increase the flexibility without large investment is to operate the power plant in hydraulic short circuit (HSC) mode. This technological solution is simple to implement, but requires an in-depth study of various technical aspects among which the hydraulic transients of the new operating modes is of high importance from the installation's safety perspective. In the framework of XFLEX HYDRO H2020 European research project, the exploitation of this solution is under implementation at Grand-Maison PSP. Located in the French Alps, Grand-Maison PSP is equipped with 8 reversible multi-stage Francis pump-turbines and 4 Pelton turbines, for a total installed capacity of 1800MW, thus being the largest PSP in Europe and one of the major PSP in the world. The waterway includes a headrace tunnel, a headrace surge tank, 3 parallel penstocks feeding the 12 units operated under a maximum gross head of 955mWC.

In this paper, after a description of the general HSC considerations, the 1D model of the Grand Maison PSP and the related validation are presented. Finally, the most critical load cases in HSC operation are described to identify the potential hydraulic transient issues, such as extreme water levels in the upstream surge tank, maximum static pressure along the pressure shaft and minimum static pressure along the tunnels. The analysis performed for Grand Maison PSP is a contribution to the roadmap for the implementation of HSC operation in pumped storage power plant and will be made available as a public deliverable of the XFLEX HYDRO H2020 European research project.

In the current energy market, hydraulic turbines are increasingly demanded to work in Frequency containment reserve (FCR) mode to compensate the constant frequency fluctuations in the electrical grid. To do so, hydraulic turbines change their generating power continuously which implies to regulate the flow rate. Kaplan turbines are double regulated machines that change the position of both guide vanes and runner blades to regulate the flow rate maximizing their efficiency. Therefore, guide vanes and runner blades are continuously moving when they provide FCR, leading to high wear and tear of the regulation system components.

Within the frame of the European project XFLEX Hydro, a new technology to reduce the wear and tear of the regulation system in FRC have been implemented. This technology consists in the hybridization of the unit with a battery system. In that way, the battery is the one in charge of providing part of the power fluctuations to the grid, reducing the movements of guide vanes and runner blades of the turbine. The battery system was successfully installed in August 2021 in one unit of the Vogelgrun powerplant, in France. Since that moment, the unit has been working in hybrid mode.

A monitoring system was installed in the power plant in two different units, the one hybridized and another without hybrid system. Several sensors were installed and different parameters were measured simultaneously to calculate the wear and tear of the different components. In this paper, a comparison between the hybrid mode and the standard mode (non-hybrid) is performed in terms of mileage and wear and tear of the guide vanes and runner blades servomotors.

The stability of the electricity grid will be disrupted by the massive integration of new renewable energies. Hydropower plants have a major role to play in this transformation of the electricity market by increasing their operational flexibility and their ability to provide ancillary services. However, this flexibility may lead to an accelerated degradation of mechanical components. By changing the turbine operating point far from the best efficiency point or by increasing the number of transient manoeuvres such as start and stop sequences, unsteady flow phenomena, cavitation development and additional wear and tear stress the unit's components and impact its lifetime. The present work aims to provide preliminary insight on the optimization of the start-up sequence of a 5 MW reversible Francis pump-turbine equipped with a Full Size Frequency Converter (FSFC). The goal of the optimization approaches are to determine a start-up sequence which minimizes the runner damage, the penstock fatigue and the water losses. The objective functions are evaluated by 1D hydraulic transient simulations with the SIMSEN software and which it allow to compare the relative mitigation between the conventional fixed-speed start-up and the linear variable-speed start-up equipped with a Full Size Frequency Converter.

Hydropower is the backbone of energy transition thanks to its operational flexibility and ability to provide ancillary services. These grid support capabilities are called upon to play a major role to maintain the grid vulnerability at acceptable levels in view of the increasing penetration of stochastic and intermittent renewable energies. In this context, variable speed reversible pump-turbine technology is a key asset as it can extend the head operating range of a powerplant, enable power network control in pump mode and further increase the flexibility services to the electrical grid. Furthermore, by taking advantage of the so-called flywheel effect, variable speed enables the fast active power injection/absorption in pumping and generating mode. The additional degree of freedom offered by the variable speed opens the door to different control strategies for the hydroelectric plant, as the converter can be used for speed or power control. This can be exploited to maximize either the efficiency or the power reserve dedicated to grid support. In this paper, the strategies to maximize the ancillary service of Frades 2 pumped storage power plant (PSPP) are investigated. The plant features two high-head single-stage reversible variable speed units, coupled with 420 MVA doubly-fed induction motor-generators (DFIM). Using a 1D numerical model to simulate the power plant behavior, it is explored to which extent Frades 2 can deliver frequency containment reserve (FCR) power, while complying with the ENTSO grid code, and speed deviation constrains inherent to DFIM. By choosing a strategy which fully exploits the flywheel effect, it is established that the FCR power band can be equal to the whole operating range of the power plant. Moreover, the fast frequency response capacity (FFR) is also evaluated. It is found that by maximizing the energy stored in the rotating masses, each unit of Frades 2 can deliver up to 110 MW of active power in 1.3 seconds.

Variable speed operation technology has become an emerging development direction in hydropower industry. Variable-speed pumped storage plants (VSPSPs) constantly undergo various transient processes, and ensuring the amplitude of pressure pulsation within the control range is critical to the safe operation of variable-speed pumped storage units (VSPSUs). However, its extreme value is usually determined by rule of thumb in design stage, and few researches focus on the relationship between pressure pulsation and variable speed operation of VSPSUs. In this paper, characteristic curves of pressure pulsation are plotted by processing model test results with Gridfit function, taking a real VSPSP as an example. The total pressure during transient processes is predicted by simulation of a numerical model. Load rejection conditions (change initial speed) and large load regulation conditions (change speed command and power command) are simulated, and the influence of operating trajectory on pressure pulsation during variable speed operation is analyzed. The results show that with the increase of initial speed or speed command, the value of pressure pulsation increases; the trajectory passes through more high-amplitude pressure pulsation region and gradually shifts to the S-shaped region, causing the pressure oscillation. This study could provide theoretical reference for the operation of VSPSUs.

Refurbishment and upgrading of ageing hydropower plants contribute to increase of renewable energy share in modern electrical grid systems. The potential increase of discharge and flexibility of load variation may result in much higher dynamic loads on both refurbished and non-refurbished plant components during transient operating events. First, water hammer control strategies are outlined including operational scenarios, surge control devices, redesign of the pipeline components, or limitation of operating conditions. Water hammer models and solutions are briefly discussed in the light of their capability, availability and uncertainty. The core of the paper is devoted to investigations of water hammer effects in a high-head hydropower plant Piva, Montenegro which is currently in the final phase of refurbishment. The flow-passage system of the Piva HPP is comprised of the intake structure, followed by three parallel penstocks each with Francis type water turbine at the downstream end. The outlet part starts with three parallel draft tubes that are connected to a common lower orifice-type surge tank followed by tailrace tunnel and outlet structure. Computed and measured results for a selected emergency shut-down (ESD) of one turbine form full-load are compared and discussed. Then the validated computational model is used for simulation of ESDs for a wide operating range.

The grid-connected operating condition of hydropower station is a common operation mode to provide electric energy for the load side. This paper investigates the dynamic performance and sensitivity of grid-connected hydropower station (GCHS) under uncertain disturbance. Firstly, the nonlinear uncertain model of GCHS under uncertain disturbance is established. Then, the dynamic performance of GCHS is studied when the governor parameters change under certain step disturbance, periodic disturbance and uncertain random disturbance, respectively. Finally, based on the sensitivity index of the uncertain output obtained from the extended Fourier amplitude sensitivity test method, the sensitivity of the uncertain random disturbance at different input positions is studied. The results indicate that the GCHS under periodic disturbances or random disturbances have more complex dynamic performance than that under certain step disturbance. Under periodic disturbance, the forced oscillations and high frequency resonances are generated in dynamic response of GCHS. Under the uncertain random disturbance, the system of GCHS always presents random oscillation. The state variables q_H, z, qP, y, x_s, x_t, and δ of GCHS are the most sensitive to uncertain disturbances, which are introduced at the generator or surge tank. The uncertain disturbances have significant interaction on the dynamic response of GCHS.

The aim of this article is to present a new technical solution of the multifunctional water tower. The principle of the new designed a water tower is fundamentally based on self-sufficiency and independence of energy sources, either the energy of the sun or the wind is used. To the conversion of this energy into mechanical energy for pumping of water to storage tank is used a ball screw mechanism, which driving a high discharge displacement pump based on the piston principle.

Francis turbines operating in off-design conditions are subject to pressure fluctuations resulting from the development of hydrodynamic instabilities in the draft tube. Depending on the nature of the flow-induced pressure fluctuations (synchronous or convective), this may induce dynamic stresses on the runner blades, increasing fatigue and the risk of crack propagation. This paper proposes to identify the impact of draft tube flow instabilities on the dynamic stresses of Francis turbine runners. Measurements are conducted on a prototype Hydro-Québec Francis turbine from low-load to full-load, including pressure and strain measurements on the stationary and rotating components, respectively. It is first noted that the convective component of the part-load vortex is the main source of excitation for the runner blades. The amplitude of the corresponding dynamic stresses is however reduced at locations closer to the leading edge, for which the dominant fluctuations result from the propagation of synchronous pressure fluctuations. Finally, correlations between runner dynamic stresses and pressure fluctuations measured in water passages are tentatively established for flow instabilities observed at both deep part-load and part-load conditions. This aims to evaluate the feasibility of estimating runner blade dynamic stresses based on signals measured in the stationary components for further investigation.

In Nepal and the regions along with the Andes, sediment erosion is one of the major causes of the failure of hydro turbines. Every year lots of the components of turbines exposed to sediment-laden water got eroded resulting unexpected shutdown of the power plants. The cheapest method of repairing these eroded runners and other components of turbines are welding which may induce residual stress, embrittlement, and increase carbide content. Therefore, post welding heat treatment especially stress relieving annealing is always carried out to reduce the induced residual stresses, enhance strength and other mechanical properties as well as reduce the risk of cracking. As heating time is relative of low importance, great care should be taken during heat treatment to maintain the soaking time and cooling rate. Holding time for heat treatment will vary with the thickness of casting sections involved, but should be sufficiently long enough to heat all sections to a uniform temperature throughout. The cooling rate determines the material properties at room temperature and rapid cooling avoids the formation of chromium carbide. In this paper, the post-weld heat treatment used for repair and maintenance of turbines in Nepal is studied along with the investigation of heat treatment conditions on mechanical properties and microstructures of 13Cr4Ni stainless steel used as turbine materials to optimize the heat treatment itself. First of all relative analysis of five different heat treatment conditions with respect to 'no heat treatment' regarding mechanical properties i.e. toughness, hardness, and fatigue life was carried out. Heat treatments of the specimen were done in a muffle furnace. Hardness, toughness, and fatigue strength of the specimen using the Rockwell Hardness Test, Charpy Impact Test, and Fatigue Life Test were conducted respectively. Subsequently, microstructures of all the heat treatment conditions are closely examined using the metallurgical microscope. It was observed that holding time plays a great impact on improving mechanical properties.