The design and testing of a high reaction turbine blade profile

In order to improve the efficiency of high pressure cylinder of 50∼350MW steam turbine and reduce the manufacturing cost, a steam turbine blade profiles are designed. The numerical simulation and test study of the new blade profile are carried out, and the performance of the new blade profile with variable inlet angle and mach number is discussed, and the performance of the new blade profile is improved when applied to stage environment. The results show that the performance of variable inlet angle and variable mach number of the new blade profile is better than that of the original profile. The aerodynamic efficiency of the new blade profile is obviously better than that of the original profile. The blade loss of the new design profile is lower than the original profile, and the aerodynamic efficiency of the new design profile is obviously better than the original profile. The stage test results show that the efficiency of the new design profile increases by about 0.5% compared with that of the original profile.


Introduction
As people's demand for electricity increases day by day, so does the requirement for the efficiency of steam turbines.As the main equipment of thermal power generation, steam turbine has become a common concern of major power plants and manufacturers for its high efficiency.As the main component of the steam turbine, the blade plays a vital role in the efficiency of the steam turbine.In a typical high-pressure cylinder of a steam turbine, the sum of blade profile loss and secondary flow loss accounts for about 60% of the total loss [1].Therefore, it is one of the most effective measures to improve the efficiency of steam turbines by developing a new blade profile to reduce blade profile loss and secondary flow loss.
The flow of steam in the cascade is very complex and has a high degree of three-dimensionality.There are steam leakage at the root of the stator and the shroud area at the top of the rotor, which will lead to steam leakage from the steam seal.Steam leakage at the seal will mix with the main flow, deteriorating the secondary flow at the end and creating a high loss area.And in the area where the steam leakage sucks the main flow, the suction of the steam leakage reduces the action area of the secondary flow and reduces the high loss area.Therefore, during the design of the blade, it is necessary to change the distribution of the fluid flow along the height of the blade.The load distribution along the blade height is finely designed, the fluid in the high loss area is transferred to the low loss area, the proportion of fluid in the high loss area is reduced, and the flow efficiency is improved.
Many scholars have conducted a large number of theoretical and test studies on the aerodynamic performance and loss mechanism of blades.Ainley [2] found through experimental research that there is 2 still a special flow in the cascade, that is, "secondary flow", and according to the existing research results, the flow loss in the cascade is divided into blade profile loss, secondary flow loss and end loss.Havakechian [3] found through research that, from the perspective of secondary flow, the aft-loaded profile is better, because most of the load of the aft-loaded profile is located in the rear area of the profile, which can delay the development of cross flow.In terms of profile loss, it is crucial to control the spread of flow on the suction side because 80% of the profile loss of most blade profiles occurs on the suction side.Xue Zhiheng [4] analyzed the flow characteristics of turbine vanes at different angles of attack through numerical simulation, and found that the change of the inlet angle of attack has a great influence on the flow in the cascade channel.With the increase of the angle of attack, flow separation occurs on the cascade pressure face.At large positive and negative angles of attack, the secondary flow losses transform into frictional losses in the boundary layer.Zhong Zhuhai et al. [5] conducted a numerical study on the transition characteristics of front-loaded and aft-loaded blades on incoming Reynolds number and inlet turbulence intensity, and the results showed that the influence trends were basically the same.Increased turbulence intensity will move the separation line at the lower end of the suction surface forward.
The current article presents a design and optimization process for a reactionary blade profile.The two-dimensional profile is parameterized by using the profile parameterization method, and the parameterized profile is associated with CFD software to carry out the geometric design and aerodynamic performance analysis of the profile, thereby completing the design of the stator and rotor.Test is one of the most effective means to verify whether the blade design is reasonable or not.Therefore, this paper has carried out corresponding experimental research on the designed blades, including plane blade test and multi-stage air turbine test.First, the plane cascade air test was carried out on the original profile and the new design profile, and it was confirmed that the new design profile has less profile loss.Then, two four-stage air turbine stage tests were carried out on the original profile and the new design profile to verify the improvement of the aerodynamic performance of the new design profile under real multi-stage flow conditions.

Blade profile design
The design of the stator and rotor is completed by the method of profile parameterization, and both the suction surface and the pressure surface are composed of Bezier curves.As shown in Figure 1, there are 8 and 5 control points on the suction surface and pressure surface respectively.The blade profile is optimized by adjusting the control points to reduce blade loss.Associate the parametric profile with the CFD software, set the basic parameters, and perform the original profile calculation and result analysis.Based on the original blade profile and according to the profile design objective, the optimization variables, variable range and variation range are set.After cyclic iterations, the optimal blade profile is finally determined.In the process of profile optimization, it is necessary to comprehensively consider the aerodynamic performance and manufacturing cost of the blade.On the premise of ensuring the aerodynamic performance of the blades, the number of blades is reduced as much as possible, and the difficulty of assembling the blades and the manufacturing cost are reduced.

Design of stator
Both the original stator profile and the newly designed stator profile are designed for axial steam inlet, and the exit Mach number is less than 1, which is a subsonic design.The original stator is a aft-loaded profile, and most of the blade load is located in the middle and rear of the profile.The newly designed stator is still a aft-loaded profile, but the load distribution of the blade is optimized.By controlling the diffusion on the suction surface, the newly designed profile has better afterload characteristics and a smaller degree of diffusion, which is beneficial to reduce secondary flow loss and profile loss.
Figure 2 shows the schematic diagram of the middle section of the original stator profile and the new stator profile.During the design process, keep the outlet angles of each section the same.Figure 3 shows the relative pitch (pitch/chord) distribution of the original stator and the newly designed stator.It can be seen from the figure that the relative pitch distribution of the original stator profile gradually decreases from the root to the top, while the new stator profile is designed with equal relative pitch.The main purpose of adopting this design is to reduce the blade loss and improve the performance of the unit by adjusting the load distribution of the blade.

Design of rotor
On the basis of the original rotor, the newly designed rotor increases the relative grid pitch at the root, reduces the axial width at the root, and makes the blade profile tend to the characteristic of equal axial width.Figure 4 shows the comparative schematic diagrams of the hub, middle and shroud characteristic cross-sections of the original rotor and the new rotor.Figure 5 shows the relative pitch (pitch/chord) distribution of the original rotor and the newly designed rotor.It can be seen from the figure that the relative pitch distribution of the original rotor increases approximately linearly from the hub to the shroud, and the relative pitch of the newly designed rotor has a small variation.In the high-pressure cylinder of a steam turbine, the secondary flow loss dominates.One of the motivations for this is to increase the blade aspect ratio and reduce the secondary flow loss.Another motivation is to reduce the sensitivity of the blade profile to incoming flow, ie to improve the profile angle of attack adaptation, especially in the hub region of the blade.Because the flow in the end wall region is very complex and highly three-dimensional.

Test research
Test is one of the most effective measures to verify the aerodynamic performance of the newly designed cascade.Therefore, the planar cascade test and the multi-stage turbine test were carried out on the newly designed profile and the original profile respectively on the turbine test bench of DTC.

Planar cascade test
The purpose of the planar cascade test is to test the profile loss characteristics and variable working condition performance of the newly designed profiles.For comparative analysis, the planar cascade tests of the newly designed stator and rotor, and the original stator and rotor were completed respectively.Figure 6 shows the photos of the plane cascade test bench and the cascade test piece.Table 1 shows the relevant parameters of the plane cascade test.Figure 7 shows the changes of the stator and rotor losses with the relative inflow angle of the original profile and the new design profile (middle section).Relative inflow angle equal to actual inflow angle minus design inflow angle.It can be seen from the figure that at the design point, that is, when the relative inflow angle is 0°, the energy loss coefficient of the newly designed stator section is about 0.6% lower than that of the original stator.And the energy loss coefficient of the newly designed rotor section is about 1.1% lower than that of the original rotor.In the whole test range near the design point, the energy loss of the stator and rotor section of the newly designed profile is lower than that of the original profile, and the newly designed stator and rotor have strong adaptability to the change of the inflow angle.Figure 8 shows the change of the stator and rotor profile loss with the exit Mach number of the original profile and the new design profile (middle section) respectively.It can be seen that the loss of the newly designed stator and rotor is lower than that of the original scheme at different Mach numbers.At the same time, with the change of Mach number, the change trend of the new design profile loss is more gentle, which means that the new design profile has better performance under variable conditions.

Multi-stage turbine test
Four-stage turbine tests were carried out on the original and newly designed stators and rotors on a multi-stage turbine test rig.In order to accurately evaluate the impact of blade profile changes on efficiency, during the test, the rotor root diameter, blade height, stationary blade retaining ring, steam seal structure and installation clearance corresponding to the original profile scheme and the new design profile scheme are all the same.Table 2 shows the main geometric parameters of the stator and rotor in the four-stage turbine test.At least one repeatability test was performed for each test to determine the validity of the test measurement data.During the test, flow probes were set up upstream of the firststage stationary blade and downstream of the last-stage moving blade, and the overall efficiency was tested using a hydraulic dynamometer and a flow meter.Figure 9 is a schematic diagram of the test bench and the test rotor.See equation ( 1) and ( 2) for the definition of efficiency in the test.
Among them: Nu is the amount of work done by the turbine.G is the mass-flow.
* in T is the total temperature at the inlet.Figure 10 illustrates the results obtained from multistage turbine tests.The optimum velocity ratio point of the new design profile is basically the same as the original profile.The efficiency of the newly designed profile is higher than that of the original profile in the entire ratio range.At the best velocity ratio point, the efficiency of the newly designed profile is about 0.5% higher than that of the original profile.

Numerical simulation
In order to compare with the test results, the same CFD calculation model was established according to the test data and corresponding calculations were carried out.FINE/IGG/AUTOGRID was used to build the model and the grid.Use ANSYS CFX to pre-process, solve and post-process the current model.Numerical calculations use the SST turbulence model.See formula (3) for the definition of efficiency.Figure 11 shows the CFD calculation model of the original profile, including the stator and rotor, the main flow area and the seal.Figure 12 shows the entropy distribution and streamline comparison of the suction surface of the second-stage stator, and it can be seen from the figure that the entropy of the suction surface of the newly designed stator is smaller than that of the original stator.At the same time, the streamline of the suction surface of the newly designed stator has a smaller secondary flow area, because the newly designed stator has better afterload characteristics and a smaller degree of diffusion, which is beneficial to reduce the secondary flow loss.Figure 13 shows the entropy distribution and streamline comparison of the suction surface of the second stage rotor.It can be seen from the streamline distribution of the suction surface of the stator and rotor that the newly designed profile has better secondary flow characteristics.Figure 14

Conclusions
The current article designs a reaction blade profile, and completes the corresponding planar cascade tests and multi-stage turbine tests, and mainly draws the following conclusions: 1) The newly designed stator has better aft-loaded characteristics and smaller diffusion degree, which is beneficial to reduce secondary flow loss and eddy current loss.
2) The newly designed rotor blade profile tends to have the characteristic of equal axial width.
Compared with the original rotor blade, the newly designed rotor blade has better secondary flow characteristics, less sensitivity to incoming flow angle, and better performance in variable working conditions.
3) The results of the planar cascade test show that the energy loss of the newly designed stator and rotor is lower than that of the original design.And the newly designed scheme has strong adaptability to the change of inflow angle and Mach number.4) The results of multi-stage turbine test show that the optimal speed ratio point of the new design profile is slightly larger than that of the original profile, and the efficiency of the new design scheme is about 0.5% higher than that of the original profile near the optimal speed ratio point.

Figure 2 .
Figure 2. Schematic diagram of the middle section of the stator.

Figure 3 .
Figure 3. Schematic diagram of the relative pitch distribution of the stator along the blade height.

Figure 4 .
Figure 4. Schematic diagram of the comparison of different cross-sections of the rotor.

Figure 5 .
Figure 5. Schematic diagram of the relative pitch distribution of the rotor along the blade height.

Figure 6 .
Figure 6.Plane cascade test device and cascade model.

outP
is the outlet pressure.* in P is the total pressure at the inlet.R is the gas constant of air.k is the (ideal) adiabatic index of air.

Figure 9 .
Figure 9. Schematic diagram of the multistage turbine test rig and rotor model.Figure10illustrates the results obtained from multistage turbine tests.The optimum velocity ratio point of the new design profile is basically the same as the original profile.The efficiency of the newly designed profile is higher than that of the original profile in the entire ratio range.At the best velocity ratio point, the efficiency of the newly designed profile is about 0.5% higher than that of the original profile.

Figure 10 .
Figure 10.Results of turbine model tests.
shows the comparison of CFD calculation results and experimental results of the original profiles and the newly designed profiles.It can be seen from the figure that the CFD calculation results are in good agreement with the test results.

T
is the total temperature at the inlet.* in P is the total pressure at the inlet.* out T is the total temperature at the outlet.out P is the pressure at the outlet.

Figure 14 .
Figure 14.Comparison of CFD calculation results with test results.

Table 1 .
The relevant parameters of the planar cascade test.

Table 2 .
The main geometric parameters of the stator and rotor of the four-stage turbine model.