Table of contents

2021 4th International Conference on Mechatronics and Computer Technology Engineering (MCTE 2021) has been held successfully on October 15-17, 2021. Due to the COVID-19 pandemic around the world and with the strict travelling rules in China, it is still difficult to take international travel for our attendees abroad. Therefore, MCTE 2021 was held both in physical (Xi'an, China) and online (Zoom).

The conference mainly focuses on mechatronics, computer technology engineering, computer science and other research fields, aiming at providing a platform for experts, scholars, engineers and technical researchers engaged in mechatronics, computer technology engineering and computer science research to share scientific research achievements and cutting-edge technologies, understand academic development trends, broaden research ideas, strengthen academic research and discussion, and promote cooperation in industrialization of academic achievements. Experts, scholars, business people and other relevant personnel from universities and research institutions at home and abroad are cordially invited to attend and exchange.

During the conference, the conference model was divided into three sessions, including oral presentations, keynote speeches, and online Q&A discussion. In the first part, some scholars, whose submissions were selected as the excellent papers, were given about 5-10 minutes to perform their oral presentations one by one. Then in the second part, keynote speakers were each allocated 30-45 minutes to hold their speeches. There were 90 individuals who attended this hybrid conference.

List of Committee member is available in this pdf.

All papers published in this volume of Journal of Physics: Conference Series have been peer 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.

• Type of peer review: Double-blind

• Conference submission management system: AI Scholar System

• Number of submissions received: 156

• Number of submissions sent for review: 151

• Number of submissions accepted: 86

• Acceptance Rate (Number of Submissions Accepted/Number of Submissions Received X 100): 55.1%

• Average number of reviews per paper: 2

• Total number of reviewers involved: 50

• Any additional info on review process:

• Step 1. Each of selected paper will be reviewed by two/three professional experts in the related subject area.

• Step 2. Review Reports received from the experts will be judged by one of the editors either Review Reports are logical or not?

• Step 3. If not logical, then editor can assign new reviewer or can also judge at his/her own.

• Step 4. If logical, then Review Reports will be sent to authors to modify the manuscript accordingly.

• Step 5. Authors will be required to revise their papers according to the points raised.

• Step 6. Revised version will then be evaluated by the editor for the incorporation of the points raised by the reviewers.

• Step 7. Then the editor will send the revised manuscript to the reviewers again for re-evaluation.

• Step 8. If the reviewers approve the revise version of the manuscript, then will be accepted for publication.

• Contact person for queries:

Xuexia Ye

publication@keoaeic.org

AEIC Academic Exchange Information Centre

Very recently special attention has been paid to soft sensors for motion tracking. It is known from the work by Chen et al [2021, Comp Anim Virtual Worlds. 2021; e1993.] that a wearable motion tracking system was developed, in which five sensors were placed around the region of arm and shoulder. In this study, we explore the effect of different sensors on motion recognition and select the sensors with excellent differentiation for different movements.

The multi-robot path planning aims to explore a set of non-colliding paths with the shortest sum of lengths for multiple robots. The most popular approach is to artificially decompose the map into discrete small grids before applying heuristic algorithms. To solve the path planning in continuous environments, we propose a decentralized two-stage algorithm to solve the path-planning problem, where the obstacle and inter-robot collisions are both considered. In the first stage, an obstacle- avoidance path-planning problem is mathematically developed by minimizing the travel length of each robot. Specifically, the obstacle-avoidance trajectories are generated by approximating the obstacles as convex-concave constraints. In the second stage, with the given trajectories, we formulate a quadratic programming (QP) problem for velocity control using the control barrier and Lyapunov function (CBF-CLF). In this way, the multi-robot collision avoidance as well as time efficiency are satisfied by adapting the velocities of robots. In sharp contrast to the conventional heuristic methods, path length, smoothness and safety are fully considered by mathematically formulating the optimization problems in continuous environments. Extensive experiments as well as computer simulations are conducted to validate the effectiveness of the proposed path-planning algorithm.

In order to improve the dynamic walking performance of quadruped robot, a control method of optimizing feet forces distribution based on virtual model is proposed. In the supporting phase, the virtual model control method is applied to solve the virtual force of the torso. Combined with the gravity of the center of mass (CoM), the distribution problem between the virtual force of the CoM and the feet forces of the supporting legs is transformed into a quadratic programming (QP) problem, which is solved by Gurobi to realize the optimal distribution of the feet forces. Similarly, the virtual force of the swinging leg is solved by using the virtual model, and the joints torques of the robot's swinging legs are obtained by combining the inverse dynamics feedforward of the swinging legs. Through the simulation of quadruped robot's trot gait walking by webots and vs2019, it is verified that this method can stabilize the robot's attitude angles and body speeds near the target values. Compared with the feet forces distribution method that abandons the lateral force control, the application of this method makes the fluctuation range of the attitude angles of the robot and the ground reaction forces (GRFs) of the supporting legs smaller. It is proved that this control method can effectively improve the walking stability and robustness of the quadruped robot.

To better apply the phase-sensitive optical time-domain reflectometer (Φ-OTDR) in projects with complex environments, given the problem of the low signal-to-noise ratio of the Φ-OTDR signal, a variational modal decomposition (VMD) is proposed. Signal processing algorithm combined with singular spectrum analysis. First, the equalization optimizer (EO) optimizes the VMD. The VMD decomposes the preprocessed signal into a multi-layer eigenmode function (IMF), selects the IMF component according to the correlation coefficient, and after the SSA noise reduction process, the signal Reconstruction to realize the noise reduction processing of the Φ-OTDR signal. Through experiments, it is proved that the relative mean square error (RMSE) of the method in this paper is lower than that of EO-VMD, and the noise reduction performance is better.

In the process of conventional rope core drilling, the connection and disassembly of waterway occupy a lot of auxiliary time, which seriously affects the drilling efficiency, and even causes drilling or safety accidents. The waterway control assembly and application method for the rope coring rig developed by the author's team can control the waterway flow direction. It does not need to disassemble the water pipe frequently to control the raising speed of the inner pipe and avoid damaging the drilling tool too fast.Introduction.

Array eddy current detection (ECAT) technology has the advantages of fast scanning speed and high detection efficiency, and has a wide range of application prospects. However, the traditional array eddy current sensor has a poor coupling effect with the inner wall of the pipeline and is not suitable for in-pipe inspection. Based on the basic principles of ECAT, a flexible array eddy current sensor made by flexible printed circuit board (FPCB) technology is designed and developed, which can realize 360° detection of defects on the inner wall of natural gas pipelines. The paper uses ANSYS finite element software to establish a simulation model of multi-parameter effects, study the influence of sensor size parameters and detection parameters on eddy current signals, carry out detection experiments on blind hole defects in the inner wall of steel pipes, analyze the response characteristics of defect signals, and verify the feasibility of using the FPCB array eddy current sensor for internal inspection of natural gas pipeline.

The fault characteristic signal energy for early gear tooth breakage is relatively weak and easily drowned by other signals, which is not conducive to the study of the fault development stage. A multi-order modulated sideband RMS (Root Mean Square) trend analysis method is proposed to analyse the development trend of the broken gear fault characteristics. By using this method to analyse gear breakage faults, the multi-order modulated sideband RMS trend analysis method can effectively determine the fault deterioration and fault stabilisation stages.

Comparators play a significant role in the semiconductor industry and have become indispensable in the design of ADC. The delay and energy consumption are two important indicators of the comparator. Many designs have been made to reduce the delay and energy consumption, such as separated gata-biasing cross-coupled transistors for a new latching stage, and pMOS is used to replace nMOS in comparators. This paper analyzes the working principle of the proposed comparators designed for different needs reported on different papers. It compares their simulation results about key data such as delay and energy consumption.

Most of Chinese aquacultural products rely on artificial nets to catch in 2020. Fish suction pump can replace artificial net, is an important mechanical equipment of fish breeding industry. The development of fish suction pump is relatively backward and the level of automation is still to be improved, which is not conducive to the large-scale and rapid development of aquaculture in various countries. This paper designs a single tank vacuum suction fish pump, the suction fish pump mainly uses the tank and the air pressure difference, through the automatic control system on the suction fish pump and each stage of the control, relying on the gravity of the mixture of fish and water to release fish. In this paper, fluent was used to conduct simulation analysis on the flow diagram of fish-water mixture inhaled at work, and the flow diagram and velocity vector diagram of the internal flow field were obtained. The results show that the vacuum fish suction pump can effectively reduce the fish collision damage. The vacuum fish suction pump designed in this paper can effectively complete the fishing work of cultured fish, with the characteristics of high efficiency and low damage, which can provide theoretical reference for the design of vacuum fish suction pump in the future.

With the continuous increase of web resources, information overload is becoming more and more serious. Getting the information that meets the needs of a large amount of data has become an urgent problem. One of the popular studies in click-through rate prediction is to construct feature interactions to improve prediction accuracy. Traditional models with low-order interactions cannot adequately represent the intersection information, and deep models with undifferentiated feature intersections lack relevance. In this paper, we propose a model called SELNN, which improves the accuracy of click-through rate prediction through attention mechanism and logarithmic conversion. On the one hand, improving attention mechanisms learns the importance of each feature itself. On the other hand, combining logarithmic conversion structure with feedforward neural networks to learn different orders of feature interactions. The experimental results indicated that the AUC values of SELNN on the Criteo and Movielens-1M datasets were 81.46% and 87.36%, respectively. SELNN effectively improves the prediction accuracy and reduces the number of parameters and computational effort.

Estimation of vehicle longitudinal acceleration is very important in vehicle active safety control system. In this paper, two driving conditions of a 4WD off-road vehicle are divided by vehicle signals such as steering angle. Under different working conditions, different estimation algorithms are adopted. In the straight driving condition, the longitudinal speed was estimated by adjusting the variance weight of acceleration Kalman observation noise based on kinematics method. For steering conditions, in order to obtain the longitudinal velocity at the center of mass, by dynamic method, a lateral state estimator was designed and tire sideslip dynamics was modeled. The CarSim-Simulink co-simulation results show that the proposed algorithm has high accuracy and strong practicability.

UAV base stations (UAVBS's) have been proposed as a revolution for the new architecture of 5G networks. The UAVBS's can be deployed as access points to provide wireless services to users in emergency scenarios. However, it is challenging to solve the highly coupled problem for UAVBS deployment and power allocation. In the meanwhile, the hybrid analog and digital beamforming is leverage to reduce the hardware cost for beamforming in 5G networks. In this work, we first use k-means algorithm to solve the 3D placement of UAVBS's by exploiting the optimal coverage altitude. Next, power allocation problem is resolved using the difference-of-two-convex functions (D.C.) programming algorithm. Furthermore, the quality of service (QoS) for each user is guaranteed by adjusting the transmitted power. Finally, extensive experiments are conducted to demonstrate the feasibility of the proposed algorithm.

The smoothed particle hydrodynamics (SPH) is widely used in numerical simulations as a kind of meshless method. The particles are used to represent object, which is a fully Lagrangian modeling scheme without the need to define a spatial mesh so it can simulate large deformation. The classic post process method cannot visualize the simulation results smoothly, so a novel post processing method is proposed in this paper. By developing a program, the coordinates and field variables of particles were collected to a file. Using the kernel function, the field variable on a specified slice plane was interpolated. By developing a program, the simulation results have been rendered smoothly. This method is illustrated by a penetration example in this paper.

Rotating machines are common equipment in industrial production, which may cause failure for a long time. Because of its convenient use and non-destructive to itself, acoustic detection method is suitable for fault diagnosis of rotating machinery. The convolution neural network model is used to identify several typical rotating machine faults. The repeatability experiments and different training sets show that the method has good universality. A visual fault identification system is built, and the effect of the system is verified by experiments.

With the rapid development of new energy generation, the intermittence and randomicity of its power output will have a significant impact on the transmission capacity of DC motor. Therefore, a virtual DC motor stability control method considering the fluctuation of new energy generation is proposed. The natural frequencies and modes of the virtual DC motor shafting rotor are analyzed by means of a steady sinusoidal excitation at zero speed. Considering the transient dynamic response of the shafting rotor of virtual DC motor under the fluctuation of new energy generation, Taylor series and transfer acceleration matrix method are used to calculate the transient dynamic response of shafting rotor under the fluctuation of new energy generation, and the parameters of virtual DC motor are identified and estimated. Based on this, a proportional resonance controller is designed to realize the stability control of virtual DC motor. Experimental results show that the interactive power curve between virtual DC motor and regional distribution network is smoother after optimal control, and this method can effectively improve the power balance ability of virtual DC motor.

In order to realize the nondestructive testing (NDT) of the internal leakage fault of hydraulic spool valves, the internal leakage rate must be predicted by AE (acoustic emission) technology. An AE experimental platform of internal leakage of hydraulic spool valves is built to study the characteristics of AE signals of internal leakage and the relationship between AE signals and leakage rates. The research results show the AE signals present a wideband characteristic. The main frequencies are concentrated in 30~50 kHz and the peak frequency is around 40 kHz. When the leakage rate is large, there are significant signal characteristics appearing in the high frequency band of 75~100 kHz. The exponent of the root mean square(RMS) of AE signals is positively correlated with the exponent of the leakage rate only if the leakage rate is greater than 2~3 mL/min. This find could be used to predict the internal leakage rate of hydraulic spool valves.

Aiming at the problem of low filtering accuracy and even divergence caused by model mismatch when using extended Kalman filter in ship GPS navigation and positioning state estimation, a positioning ship state estimation algorithm based on the fusion of improved unscented Kalman filter and particle filter is proposed. Compared with the traditional particle filtering algorithm, the algorithm has two improvements: first, the algorithm uses untraced Kalman as the main framework, and uses the optimal estimation of particle updating state by particle algorithm; Secondly, in the resampling process, a resampling algorithm based on weight optimization is proposed to increase the diversity of particles. The simulation results show that not only the particle degradation degree of the particle filter is reduced, but also the particle tracking accuracy is improved.

Aiming at the problems of strong non-linearity of gravimeter stabilisation platform system, poor robustness of linear PID control algorithm and non-adaptive control system. This paper designs a LADRC-based gravimetric stabilisation platform control system design and method based on the research of PID controller and ADRC control method, and gives the anti-saturation and anti-noise design applicable to it, and the simulation experiment shows that the method is feasible.

Along with the historical process of the industrial revolution, the automobile has undoubtedly made life more convenient. But at the same time, various problems also come along: Traffic jams are becoming more and more serious, and problems such as parking are becoming urgent problems. This design uses STC89C52 microcontroller, through the infrared sensor module to detect the information of the garage, then the nRF24L01 wireless module is used to carry out the wireless transmission of master and slave information. Finally, the LCD1602 LCD screen is used to establish the human-computer interaction page, and an intelligent parking space management system is designed. The owners can easily check the parking lot to understand the flow of the parking lot through the system, and judge whether to enter the parking lot, saving time. At the same time, it is easier to integrate with other intelligent systems, which provides a foundation for more advanced management and is the premise for making the parking lot a highly intelligent product in the future.

Aiming at the problem of detection and location of magnetic targets in water beach, the acoustic magnetic composite detection method is studied. After the sonar obtains the image of the suspicious object in the target area, the magnetic target recognition and location are realized by using the abnormal magnetic field distribution data near the target area measured by the shipborne magnetic sensor and the multi-sensor information fusion method. A target recognition and location method based on a priori information is proposed to solve the problem that the measurement results of magnetic sensor can not fully reflect the influence of ferromagnetic target on the surrounding magnetic field due to terrain constraints. In order to make up for this lack of information, taking the sonar measurement results as a priori information, the hypothesis test method is adopted to make full use of all the measurement results of different types of sensors to realize the recognition and positioning of magnetic targets.

In many cases, ships will encounter special conditions, such as bad sea conditions, and the effect of various disturbance factors, such as wind and waves, will cause low-frequency longitudinal and transverse rocking motion of ships. This would affect the rotor-bearing system of the power plant, which is directly attached to the hull of the ship, and would roll along with the hull in the same longitudinal and transverse motion. This will have a significant impact on the stability of the rotor bearing system of the ship power plant and the safety of the ship. In this paper, a rotor-bearing test rig is mounted on a multi-degree-of-freedom swing test rig simulating sea conditions. The multi-degree-of-freedom rocking test bed simulates sea conditions. The rotor-bearing test bed is equipped with an eddy current sensor and an acceleration sensor to obtain experimental data. The programmable loader is used to control the rotor speed so as to carry out the experiment process at different speeds. Finally, the dynamic characteristics of rotor-bearing system under swinging condition are obtained.

In this paper, taking the feeding process as a form of impulsive and considering the time-delay in fermentation process. A robust model with the time-delay system as the control variable and the time-delay system as the constraint is established. In order to solve this optimal control problem, we have propose an particle swarm optimization method to solve problem. Numerical results show that 1,3-PD yield at the terminal time increases compared with the experimental result.

By optimizing the data detection performance of distributed wireless sensor networks, the data sensing and collecting ability of wireless sensor networks can be improved. Traditional methods adopt statistical characteristic parameter detection algorithm for distributed wireless sensor network data detection. Distributed wireless sensor network data has strong time-frequency coupling, so it is difficult to realize frequency domain spatial parameter clustering in frequency domain, and the detection performance is not good. A distributed wireless sensor network data detection algorithm based on non-stationary filtering and high-order statistical feature peak retrieval is proposed. The data model of distributed wireless sensor network is constructed under the interference of color noise. The weak vibration signal is subjected to time-frequency analysis and noise separation by non-stationary filtering, and the spectral peak of distributed wireless sensor network data is searched by the fourth-order cumulant slice post-operator to realize the optimal detection of signals. The simulation results show that the algorithm has a high probability of accurate detection, and has a good ability of suppressing noise and noise sidelobe information interference, which improves the probability of accurate detection of distributed wireless sensor network data under low signal-to-noise ratio.

This paper designs and implements a High Speed Redundant IO Bus for Energy Power Controller System. The physical layer adopts multi-point low-voltage differential signal standard. This bus has the characteristics of high real-time, high throughput and easy expansion. The controller communicates with IO module by A/B bus alternately, monitors link status in real time and collects IO module data. Non real time slots can be used to control non real time messages for IO modules such as time synchronizing and memory monitoring. The controller ARM core runs QNX real-time operating system, and transmits the message needed to communicate with IO modules to the FPGA through DMA. After receiving the message, the FPGA parses the message and automatically fills in the CRC check code and frame end flag at the end of the message. When the FPGA receives the data feedback from the IO module, it performs CRC verification. If the verification passes, it fills the corresponding module receiving buffer. Otherwise, it fills the CRC verification error flag in the register of the corresponding IO module to reduce the load of the arm core.

Average allocation of data rate to each user is inefficient since the resource a base station can allocate is limited. Thus, user selection and user scheduling need to be applied into multi-user massive multiple-input multiple-output (MIMO) downlink system. In this paper, we mainly focus on the methods of user selection. First, we establish a downlink system model including transmission model and channel model. Then, two user-rate based user selection algorithms via the signal-to-interference-plus-noise-ratio (SINR) are proposed, where the SINR is generated by MRC beamforming. Finally, simulation results are provided to compare the performance of two proposed algorithms and their fairness towards selected users. In the simulation results, location-based selection algorithm and random selection algorithm are jointly compared. The second proposed algorithm possesses the highest total sum-rate and is the optimal algorithms among the four algorithms.

In order to overcome the misjudgement and oscillation of the traditional perturbation observation method when tracking the maximum power point of the PV system, this paper proposes a fixed-zone variable-step hysteresis comparison algorithm. The bi-directional perturbation of the hysteresis loop can effectively avoid the occurrence of misjudgement; after setting the partition voltage threshold, different step lengths can be used to track specific regions, which can ensure the tracking speed and maintain the system stability. Finally, the proposed algorithm is validated in Matlab/simulink and the results show that the method is effective in tracking the maximum power point of the PV system in all aspects compared to conventional algorithms.

In recent years, multiple measurement systems fusion technologies have developed rapidly. Ultrasonic ranging is paid much attention because it has the advantages of directly measuring short-distance targets, high longitudinal resolution, and a wide application range. However, when the ultrasonic distance measurement is carried out by assembling the circuit board, it is found that the switch is triggered multiple times, and the measurement data is inaccurate. This paper proposes a new method that adding debounce to improve the accuracy of measurement. Besides, Some suggestions are provided to decrease the operation difficulty in the actual measurement process. Based on the experimental result, the accuracy and efficiency of ultrasonic distance measurement are greatly improved via this method proposed in this paper. By the way, the single-chip computer simulation technology simulates the ranging phenomenon in the experiment.

A wireless power transmission system based on LCC-S compensation network is designed. The transmission system is composed of power frequency power supply, LCC-S compensation resonant converter and DC/DC converter, and the output power supply for the battery system. Through the design of transmission coil and compensation network parameters, zero current input is realized. A wireless transmission coil with outer diameter of 47cm and spacing of 30cm is designed. A 2kW wireless transmission system prototype is built. Under 311V input, the actual transmission efficiency of high frequency inverter/wireless transmission/high frequency rectifier reaches 88.5%, and the wireless energy transmission with high distance to diameter ratio is realized.

Traditional virtual synchronous generators (VSG) control inverters. Inverter output frequency characteristic of the virtual inertia (J) and virtual damping (D) coefficient, and the virtual parameters need to be modified and adjusted according to the purpose. To solve this problem, this paper proposes a virtual parameter adaptive control strategy based on fuzzy control theory to adjust the frequency characteristics of VSG. MATLAB/Simulink is used to build a simulation model to verify the correctness of the proposed fuzzy control theory's adaptive virtual parameter theory.

The nuclear reactor control system plays a crucial role in the operation of nuclear power plants. The coordinated control of power control and steam generator level control has become one of the most important control problems in these systems. In this paper, we propose a mathematical model of the coordinated control system, and then transform it into a reinforcement learning model and develop a deep reinforcement learning control algorithm so-called DDPG algorithm to solve the problem. Through simulation experiments, our proposed algorithm has shown an extremely remarkable control performance.

In recent decades, the aging of the world's population has intensified, and the number of people with disabilities caused by various disasters and diseases has gradually increased. Most of the elderly, frail and disabled people will choose wheelchairs as their means of transportation. However, ordinary wheelchairs are unable to climb stairs, especially in cities, which seriously limits the range of activities of wheelchair users and affects their daily life. Hence, it is of great significance and value to design an intelligent building climbing wheelchair with appropriate price, stability and safety. This paper designs a multi-functional planetary height adjustable building climbing wheelchair with simple structure and low price. Firstly, a multi-functional building climbing wheelchair based on planetary gear train is designed by using modular design idea, including ground walking mechanism, in place steering mechanism, center of gravity adjustment mechanism and Seat leveling mechanism. The working principle and design characteristics of each mechanism are analyzed in detail. Then, through the analysis of the mathematical model of the climbing mechanism, the dimensional parameters of the three-star wheel set are determined. According to the power demand of wheelchair, the power system is selected and analyzed. It provided a new possibility for the wheelchair application and also for those in need.

In clinical practice, augmentative and alternative communication (AAC) is an effective training tool to improve the communication skills of individuals with language impairment, hearing impairment, and autism. However, there are few studies on the development of augmentative communication systems based on Mandarin Chinese in China. This study presents an augmentative communication system developed using the MATLAB GUI platform. The system consists of two parts: a learning part and a communication part, where the learning part provides the basis of the vocabulary content for the system, covering 5558 words, and its content is mainly a word list designed based on the content of special school language textbooks. The communication part provides the system with a complete discourse training content, mainly by arranging and combining the vocabulary content of the learning part to form a complete discourse. This aided communication system can expand the vocabulary of people with language expression disorders and improve their social communication skills. In addition, the effectiveness of the system was tested in a single-subject experiment, and the results showed that the augmentative communication system is effective in learning pronunciation, expanding vocabulary, and improving the social communication.

In this study, based on the computer vision technology, we developed a recognition system for nuts positioning to complete the automatic bolt assembly part of the automatic production line. The actual image of the nut was captured by an industrial camera, which would be processed by the following edge detection and Hough circle transformation. After that, the coordinates of the nuts were obtained in the pixel scale. Finally, the real position of nuts would be fed back to the robot arm, according to which the automatic assembly of the bolt would be completed. This computer vision based recognition system is an indispensable part for the efficiency and accuracy improvement of automatic production line.

Imitation is ubiquitous, yet what self-regulation orientations' role played in imitation strategies is poorly understood, which is particularly challenging in dynamic and uncertain environments. According to regulatory mode theory, we model two imitation strategies: assessment and locomotion. Assessment pays more attention on comparation among different alternatives, they repeatedly measure, evaluate, and compare desired means and try to find out the 'best' one. Contrariwise locomotion refers to 'keep moving', once choosing one alternative, they change some choices and learn from the resulting performance feedback. Using a computational model, we explore the performance implications of dynamic environments for these two imitation strategies. Consequently, when environment is stable, assessment is more effective in maintaining the lead, whereas locomotion prevails as environmental changes become more frequent and substantial. We contribute to the literatures on strategy, imitation, and NK studies.

In order to solve the problem of elevator braking test inspection, this paper combs the requirements of relevant national standards and inspection rules on braking performance, and develops an elevator braking test inspection instrument. With the help of the instrument, the elevator braking effect can be expressed quantitatively, and the inspectors can conveniently judge whether the braking test results meet the requirements of elevator inspection rules remotely. A way to standardize and rationalize the test is given. In order to solve the problem of elevator brake test inspection, this paper sorted out the national standards and inspection rules on braking performance, and developed an instrument of elevator brake test inspection. With the help of the instrument, we could quantitatively give the expression on the elevator braking effect, and the inspectors can conveniently judge whether the braking test results meet the requirements of the elevator inspection rules in the remote.

At present, transformer verification line of metering centre adopts fixed cycle inspection method manually. This method requires downtime for detection, which costs a lot of time and cost. Moreover, the inspection cycle is determined based on experience and lacks rigorous basis. To solve this problem, a hybrid delivery of inspection devices is proposed to realize non-stop detection and reduce the cost of inspection time. Considering impact of cost and false detection risk on inspection cycle, a multi-objective optimization model of inspection cycle based on inspection and false detection cost is proposed. Based on NSGA-II algorithm, perturbation population is introduced to enhance the global search ability, which aims to minimize the cost of inspection and false detection. Taking the verification line's inspection plan of the metering centre as an example. It is solved by ENSGA-II algorithm, and feasibility of hybrid delivery mode is verified, which reduced downtime by 14.58%. A more reasonable inspection cycle is obtained, inspection cost is reduced by 29.57%, and false detection cost is reduced by 6.34%. It provides a reference for the formulation of inspection plan in the actual production process.

Advanced Persistent Threats (APT) have caused severe damage to the core information infrastructure of many governments and organizations. APT attacks usually remain low and slow which makes them difficult to be detected. In this case, the way of correlatively analyzing massive logs generated by various security devices for effectively detecting the new type of cyber threat turns out to be more and more significant. In this paper, on the basis of analyzing the principles and characteristics of APT, we propose an intelligent threat detection method based on the expanded Cyber Attack Chain (CAC) model and the long short-term memory network (LSTM) autoencoder to extensively correlate malicious behaviors from spatial and temporal dimensions, which provides a brain new idea for the application and practice of complex network attack detection.

This research designs an absolute-value detector with the function of threshold comparing. Specifically, it is an essential device in the spike detection of the brain-machine interface. The optimized design in the research can accomplish the main functions in spike detection and has good performance in both delay and energy consumption. It comes up with two types of design at the beginning. To make the design reliable and comprehensive, it decides to discuss both methods in this paper. The first design is using a full adder, multiplexer and comparator. The concept of its logic circuit is adding the logic one to the input when the given input data is negative, keeping the original information as the given input data is positive. To achieve the function of adding, this study chooses the full adders. The primary purpose of using multiplexers is to select from the processed input and original input, and the choice depends on the most significant bit (MSB) of the input data. To compare the absolute value of the input data with a given threshold, this research used a multi-bit comparator. The second design is based on the fundamental algorithms of calculating total numbers. It indicates that this study can operate it with the threshold value through a subtractor when the input is negative. On the contrary, an adder can be used when the information is positive. Based on the concept of logic optimization, this study chooses to use the only subtractors, and it just needs to focus on the borrow bit, which can indicate the more significant number. By connecting the MSB of the input with the subtractors through XOR gates, the selection can be achieved without using any multiplexer. In the process of removing and replacing the devices, it reached the optimization of the design. Then, this paper compared the minimum delay by calculating each stage's size and finding that the second design is better. Finally, based on the dual design, this essay computed the energy consumption in the circuit and implement V_DD optimization to obtain the minimum energy.

At present, artificial intelligence has increasingly become the most promising and essential major. The interdisciplinary cooperation between the integrated circuit and artificial intelligence brings infinite possibilities. In the convolutional neural network area, in order to achieve the valuable output, the comparison between values is often encountered. The output is obtained by comparison between the result after the convolution calculation with the designed threshold. Therefore, our team not only design a 4-bit binary comparator hardware logic circuit to complete the task but also discuss and verify the feasibility and performance characteristics of the program from the perspective of energy and time delay. As for the overall framework, we design a convenient circuit that converts the complement code into the original code and chooses a CLA adder to accomplish this part. Using this kind of adder can ideally help us reduce the time delay at the expense of a complicated circuit schematic. In the comparator part, we design a high-quality circuit framework. The strategy of our circuit is to compare the relationship between the four-bit binary code and the threshold bit by bit from MSB to LSB, which performs better than the original 4-bit comparator, and we design two outputs that can legibly illustrate the relationship between two values. We use logic effort to discuss the normalized delay in our project. Besides, we find the connection between the energy and the delay by calculation. Finally, we design a trade-off function to make the optimization of energy and delay together with respect to voltage.

In this paper, an automatic calibration device for multi-channel resistance strain gauge indicator is designed and its applicability and measurement accuracy are verified at laboratory. The calibration done by original resistance bridge calibrator is time-consuming for its manual operation and complex calibration process. With the intent to increase calibration efficiency, an automatic channel switch device was developed, and the resistance bridge calibrator was automated. The designed calibration device is completely computer controlled enabling a sequence of unmanned measurements. The calibration device was verified at laboratory that the maximum of error is 0.072%. It was applied to calibrate a 60-channel resistance strain gage indicator to approve its practical applicability. The result shows that the designed calibration device can realize automatic calibration and the efficiency is increased by 40%.

In order to improve the control performance of vehicle-mounted radar, a high precision leveling control system is designed through mechatronic-hydraulic integration technology. A leveling method combining angle error leveling method and height error leveling method is designed, and inclinometers are set in the front and rear of the vehicle body platform to effectively eliminate the virtual outrigger phenomenon in the leveling process. The mathematical model of hydraulic system is developed, the fuzzy PID control method is introduced, and the four-point support hydraulic leveling control system is developed by using STM32 microcontroller as the control core. The field test proves that the control system has good control performance, high leveling accuracy, short time, and meets the control requirements of the radar vehicle.

Rotating machinery often produces continuous impact during operation due to the change of load and speed, which shows the characteristics of unsteady state and time-varying. Its working state can not be comprehensively judged by a single vibration state parameter. Therefore, this paper proposes to use acoustic sensors to collect the fault noise signal of rotating machinery, and use the whole column of sensors to detect the fault noise signal. Based on the microphone array, this paper studies the adaptive beamforming algorithm (MVDR) to locate the fault source of rotating machinery in space. The effect of fault source location is verified by simulation and equipment measurement experiments. The acoustic sensor does not in contact with the equipment, which will not damage the generator set, but also provide more effective information for fault source location and fault diagnosis and analysis.

4-bit absolute-value detector (AVD), as one of the basic implementations of bit arithmetic with logic circuits, can help grab a better understanding about digital integrated circuits. Conventional 4-bit AVDs scheme in a multi-comparator and multiplexers, or need to consider multiple situations of overflow and carry-in, both of which could make the final circuit to be complex, labyrinthine and inefficient in the meantime. In this paper, a new design of 4-bit AVD is proposed, the topology of which includes a 2's complement calculator and a specially designed logic circuit known as chain carry adder (CCA). The whole circuit is concise and the critical path is rigorously considered to make it as short as possible. The delay is set to 1.5 times its minimum, which is positively corresponding to the length of the critical path, the energy accordingly reaches its lower limit. Gate sizing and Device Voltage (VDD) optimization are proceeded for the exact purpose of proving that the circuit energy is minimized.

Aiming at the high-frequency oscillation and estimation accuracy problems of traditional sliding mode observers in the control process of permanent magnet synchronous motors, a sensorless control method for permanent magnet synchronous motors based on a new approaching law sliding mode observer is proposed. Based on the construction of a permanent magnet synchronous motor two-phase static coordinate system model, a sliding mode observer is used to estimate the back electromotive force, and then the rotor speed and position information are obtained. Finally, a simulation experiment is carried out. The results show that the new sliding mode observer based on the new approaching law effectively reduces the high frequency chattering of the system, improves the estimation accuracy of the system, and has better control performance.

With the increase in the number of traction substations year by year, manual inspections are gradually being replaced by unattended inspections. Target detection algorithms based on deep learning are more widely used in intelligent inspections of power equipment. However, in practical applications, it is found that due to the small target to be detected, the accuracy of the deep learning model will decrease when the shooting angle is inclined and the light conditions are poor. This is because the algorithm's robustness is low, and the detection ability of the model will be seriously affected when the angle or illumination difference with the sample is large. Based on this, the feature fusion part of the YOLOv3 algorithm and the selection of the loss function and the size of the anchor frame are improved, and the improved ASFF fusion method is used to classify various images in the power equipment. Actual measurement and repeated experiments show that the proposed method can be effectively applied to image recognition of various power equipment, optimize robustness, and greatly improve the image recognition efficiency of power equipment.

The dynamic equivalence and modeling of large-scale wind farms is the basis for studying the problem of multi-temporal and spatial scale wind farms grid connection, and the establishment of a high-precision equivalence model is a key issue among them. This paper takes doubly-fed wind turbines as the research object. Aiming at the problem of large dynamic errors in direct aggregation of equivalents into a single unit, an equivalent model combined with particle swarm optimization algorithm is established to improve the accuracy of dynamic equivalents. The electrical parameters, reactive power reference values, active power controller parameters, and active DC voltage controller parameters are optimized using particle swarm optimization. The research results show that the equivalent model after optimizing the parameters of the active power and DC bus voltage controller is the closest to the detailed model in describing the dynamic characteristics of multiple machines.

In order to make use of fewer fault data samples to diagnose the main fault types of circuit breakers accurately in real time, an intelligent fault diagnosis method for circuit breakers based on convolutional neural network (CNN) and quantum particle swarm optimization (QPSO) is proposed. Firstly, the key features of the circuit breaker operational signal are extracted through the CNN model, and the extracted feature vectors are input into the support vector machine (SVM) for fault diagnosis. In order to improve the diagnostic performance, this paper uses QPSO algorithm to optimize the parameters of the classifier, it effectively solves the local optimal problem. The experimental results show that the method presented in this paper has achieved good results in fault diagnosis of circuit breakers, and the accuracy of diagnosis is up to 100%, which highlights the superiority of this method.

Cable intermediate joint is the weak link of power cable. Its operation determines the quality of power transmission and distribution in power system. The temperature rise of the intermediate joint is the direct manifestation of a variety of faults. Therefore, monitoring the temperature of the intermediate joint of power cable and finding and eliminating cable faults in advance is of great significance to ensure the safe and reliable operation of power system. Based on the analysis of the temperature characteristics of the cable intermediate joint, the design scheme of the temperature measurement system of the power cable intermediate joint is given, and its hardware system and software system are designed. Finally, the temperature measuring device for the intermediate joint of power cable is obtained, tested on the spot, and the power consumption of the data is analyzed. The debugging results show that the system has good characteristics and meets the design requirements.

Island for micro grid is vulnerable at the run time, which is caused by the problem of load fluctuation, voltage frequency deviation, and distributed power supply power. These issues are generated by the uncertainty of three-phase voltage waveform output at the same time. All of these problems are urgently needed to solve, this paper designs a linear active disturbance rejection control to do robust estimation. This controller is based on optimal finite impulse response (FIR) filter, which is also combined with the immunity of the micro grid system harmonic method. The proposed controller is simulated. And it is compared with the classical PI controller and the linear active disturbance rejection controller respectively too. The simulation results show that the performance and robustness of the proposed model are better than those of traditional algorithm in microgrids.

With the widespread application of power grid systems, the information security problems faced by power grids have become more obvious. Various internal and external intrusion attacks that occur frequently have become an important issue affecting the normal operation of power generation and operations. The purpose of this paper is to study the intrusion detection method of electric power information(PI) network in the cloud computing environment. With the help of the cloud platform's ability to process big data, and based on the analysis of the PI network structure, a DBN optimized BP network algorithm is proposed, and the optimized BP neural network is used as a runtime classification program. Experimental results show that MR-DBN-BP has a detection rate of 96.7% for intrusion detection of PI networks, which can effectively detect intrusions and effectively protect the power dispatch system network.

As a component of the Internet of things, high-voltage cables are the power supply infrastructure for the modern development of cities. The operation experience shows that the high-voltage cable has been broken down many times, due to the defective operation. At present, due to the limitation of detection technology, the research on detection and identification of defects in high-voltage cables is progressing slowly. Therefore, a new DR technology based on X-ray digital imaging is proposed in this paper to realize real-time detection of defects in the semi-conductive buffer layer of high-voltage cables, and intelligent detection of DR images of high-voltage cables by using image depth processing technology to realize intelligent identification of defects in the buffer layer of power cables. The results show that using the new DR technique proposed in this paper, the accurate and intuitive DR image of high-voltage cable can be obtained quickly, and the intelligent identification of defects can be realized.

Rail transit has many advantages such as large capacity, fast speed, energy saving and environmental protection, and has gradually become an indispensable means of transportation in people's lives. At the same time, the safety and reliability of rail transit is also facing severe challenges. The running part plays the role of load bearing, running and traction guidance, etc. It is one of the most important systems of the train, and it is also the accident-prone part. Aiming at the failures of rail vehicle wheelsets, bearings, overall frame and suspension parts, this article uses CNKI, Wanfang Data Knowledge Service Platform, IEEEXplore and ScienceDirect as the search database to investigate and analyze the research results published at home and abroad in recent years. The fault diagnosis technology of the traveling department summarizes the sensor types, installation positions and usage methods used in various diagnosis technologies, and finally puts forward the prospect of the follow-up research. The work of this paper will help researchers to systematically grasp the relevant research trends of the fault detection of the walking part, and then make clearer and deeper thinking.

The present current differential protection for MMC-HVDC transmission lines has absolute selectivity and powerful ability to withstand high transition resistance, while it is easily affected by distributed capacitive current and data synchronization error. To solve the problem above, this article proposes a novel current differential protection scheme. The distributed capacitive current can be calculated by integrating the linear voltage distribution in real-time. Thus, the differential value of the midpoint currents of DC line, which are calculated based on the low-pass filtered measure voltages and currents on both sides, can be adopted to identify the fault. Besides, the data synchronization error can be eliminated based on the waveform matching of the calculated midpoint currents. This novel current differential protection has excellent performance and can solve the problems of traditional current differential protection for HVDC lines.

In order to solve the problem that the process of WAPI terminal connection is troublesome, this paper proposes an intelligent support and command system of WAPI terminal equipment based on centralized control AC + AP architecture. The system stores WAPI certificate on AC (centralized controller), WAPI terminal equipment is associated with AP (wireless access point), and AP equipment notifies AC of the associated events of the terminal, AC equipment and terminal equipment enter authentication and complete certificate authentication. Complete access authentication through AC and terminal equipment, and store the unicast and multicast keys generated by authentication negotiation on AC. This method is more convenient to maintain the certificate. At the same time, on the premise of ensuring the security of the certificate, it also improves the encryption and decryption efficiency and improves the user's business experience.

The parallel connection of IGBTs has been being applied in high power neutral point clamped (NPC) three level converters. This paper investigates the impact of gate parameters (gate resistor and capacitance) on dynamic current imbalance of parallel connected IGBT for NPC three level converter. A gate parameters calculation method is proposed in the paper, and the delay time and collector current difference can be analysed quantitatively. Experimental results have shown the effectiveness of the method.

With the proliferation of the distributed energy resources (DERs), the scheduling and control of the distribution network have become more complicated. To cope with the uncertainty nature of distributed generation, a multi-timescale optimal dispatch method in active distribution network (ADN) based on the model predictive control (MPC) is proposed in this paper. First, based on MPC, a hierarchical scheduling framework for ADN is established, including long-timescale stage, and short-timescale stage. Then, via coordinated control of various resources in the ADN, i.e., distributed generators, energy storage, capacitor banks and OLTC transformer, the impact of intermittent renewable energy and load forecast errors can be reduced. Finally, considering the coupling characteristics of active and reactive power in the ADN, a joint active and reactive power optimization model is proposed to further reduce the network loss. Numerical simulation on a modified IEEE-33 distribution network system verifies the correctness and superiority of the proposed scheduling approach.

With the rapid development of society and economy, various laboratories have become more complete in number and equipment, but subsequent laboratory accidents frequently occur. Therefore, it is imperative to construct a laboratory safety management system to ensure the safety of laboratory personnel and complete equipment. This article intends to design a smoke alarm, which mainly includes measuring the sensitivity of the smoke sensor to the concentration of smoke in the air, and the linear change of the sensor converting the corresponding analog signal into a digital signal. By designing the fan speed to be controlled by the change of smoke density, the speed of the motor is adjusted. And the alarm module can sound an alarm when the smoke concentration is certain, and at the same time, it can automatically cancel the alarm when the smoke concentration decreases. Finally, a simulation experiment is designed to simulate the effect of the sound alarm of the smoke alarm disperser on the dispersal of smoke under different working environments.

The drone light show is a popular form of performance. This paper designs a new type of entertainment drone with rotating LED display technology. This drone is equipped with a rotating LED display, a single drone can display a complete picture, and has a good display effect. The application of the new entertainment drone can reduce the cost and technical difficulty of the drone light show, and expand the applicable scenarios of the drone light show.

This paper introduces a design scheme of laser array harp based on multi-dimensional wavelet transform and audio signal reconstruction. The green light beams from multiple high-power lasers simulate harp strings, use photoresistors as the signal receiving end, and use a signal conditioning system composed of analog circuits and LM393 comparators to collect and adjust the resistance signal of the laser sensor[1], and finally it is adjusted to a level signal that can be recognized by the CPU. After receiving the signal, the CPU core board analyzes the string signal, and sends control commands to the audio processing system through the industrial bus according to the analyzed digital signal. After receiving the control command, the audio processing system uses the audio signal reconstruction technology composed of multi-dimensional wavelet packets, deep learning and other algorithms to simulate the audio signals of various string music, so as to achieve the purposes of using the lasers as virtual strings and imitating musical instruments for musical performance.[2]

Aiming at the problem of poor positioning accuracy of GNSS for urban canyon environment, this paper proposes a TDOA/AOA combined positioning technology based on 5G.Firstly, the AOA estimation algorithm based on conventional beam forming is compared with that based on MUSIC algorithm. Then, the weighted matrix is smoothed forward and backward.Finally, according to the measurement data of AOA and TDOA, Chan and Taylor combined localization algorithm is used to obtain the estimated location of the user.

The finite element modelling, mesh generation and modal analysis of the rotor worktable are carried out by combining the analysis technology of finite element software in this paper. Then, the software analysis results and the actual experimental results are compared and analyzed, so as to get the causes of error, which can provide good basic data for the use of the rotor table in the future that could better meet the needs of scientific research and teaching.

For modern operation and maintenance systems, they are usually required to monitor multiple types and large quantities of machine's key performance indicators (KPIs) at the same time with limited resources. In this paper, to tackle these problems, we propose a highly compatible time series anomaly detection model based on K-means clustering algorithm with a new Wavelet Feature Distance (WFD). Our work is inspired by some ideas from image processing and signal processing domain. Our model detects abnormalities in the time series datasets which are first clustered by K-means to boost the accuracy. Our experiments show significant accuracy improvements compared with traditional algorithms, and excellent compatibilities and operating efficiencies compared with algorithms based on deep learning.

With the development of modern technology today, image analysis technology has become an important research direction in the computer field. Computer image processing technology is a cross and fusion involving multiple disciplines, which includes multiple research directions such as physics, chemistry, and computer science. Starting from the fractal dimension, this article studies the image generation and detection technology. The purpose of the research topic in this article is to have a deep understanding of image generation technology so as to master image processing methods proficiently. The methods used in this article include case analysis, data, comparison, and experimental methods, etc., and carry out related researches on image generation detection technology and fractal dimension. The experimental results show that the algorithm accuracy of GAN in the detection technology of computer-generated images is higher than other algorithms, and the improvement space is about 0.06.

This paper comments on four works for the optimization of comparator design. Today, with the development of integrated circuits, the requirements for comparators about low power, low delay, few offset voltage, and low noise are highly desirable. Specifically, these works made progress in the conventional comparator, which comprises a preamplifier and a latch. They also solved some problems, such as decreasing power and delay. Some works employ a positive feedback cross-coupled pares to provide a larger gain in the preamplifier, use PMOS switch transistors to accelerate the definition phase, or a double-tail architecture to increase the latch regeneration speed. Other work designs a charge pump to improve speed.

Since the study in the field of fusion has gradually developed toward the long-pulse experiment mode, long-pulse data has gradually become one of the main data types for pulsed experiments in the field of fusion. For long-pulse data, which is a kind of pulse-type data, it will be more difficult to transmit and store than short-pulse data because of its significant characteristics. In addition, in the design of data acquisition and control system (DACS) in fusion field, Experimental Physics and Industrial Control System (EPICS) has now gradually become the main framework of experimental control system to meet the diversity of devices and complexity of subsystems in large experimental system. However, due to the limitation of EPICS, its effectiveness in handling data transmission and storage under high speed data acquisition is not satisfactory. To solve the data transmission and storage under high-speed sampling, this paper proposes a data transmission and storage solution based on TCP/IP protocol and MDSplus database, which is designed with the concept of segmentation, i.e., data generated from experiments longer than 100 seconds are uploaded and stored in a segmented form. Currently, this system has been tested and applied, and the test result shows that the solution is feasible and the overall test system operates stably and reliably.

Aiming at the problem of large amount of calculation in extracting image feature points in panoramic image mosaic by SIFT algorithm, a panoramic image mosaic algorithm based on image segmentation and Improved SIFT is proposed in this paper. The algorithm fully considers the characteristics of panoramic image stitching. Firstly, the stitched image is divided into blocks, and the maximum overlapping block of image pairs is extracted by using mutual information. The SIFT key points are extracted by SIFT algorithm, and the dog is filtered before the spatial extreme value detection of SIFT algorithm to eliminate the feature points with small intensity value; When establishing the feature descriptor, the 128 dimension of the original algorithm is reduced to 64 dimensions to reduce the amount of calculation. In the feature point registration process, the feature descriptor is reduced to 32 dimensions, the feature point pairs are roughly extracted by the optimal node first BBF algorithm, and the feature point pairs are registered and screened by RANSAC; Finally, the image transformation matrix is obtained to realize panoramic image mosaic. The experimental results show that the proposed algorithm not only ensures the panoramic mosaic effect, but also extracts the feature points in 11% of the time of the traditional SIFT algorithm, and the feature point registration speed is 27.17% of the traditional SIFT algorithm.

Permutation entropy (PE) and weighted permutation entropy (WPE) are indexes of the complex system. In this paper, the nonlinear sequence of two chaotic states is simulated by Logistic mapping, and the effectiveness of the PE and WPE algorithms in chaotic state recognition is compared and analyzed with different sliding step sizes. The results show that the value of WPE changes more obviously than PE in different chaotic states, and the recognition effect of WPE is better than PE in dynamic states, which can provide theoretical support for further analyzing the complexity characteristics of actual data.

Fault-tolerant design of cache is a key aspect of highly reliable processor design. In this paper, based on the key metrics in Cache architecture design: reliability, power consumption, latency and area, we divided the related research into two categories: one is to maximize reliability with guaranteed latency, power consumption and area, the other is to minimize latency, power consumption and area loss while ensuring fault tolerance reliability. Based on the classification, by analyzing different studies of Data and Tag in Cache, this paper gives the characteristics of these methods and the future development trend.

Large-aperture laser transmission unit (LLTU) is a device that focuses the laser beam to the center of the target, which are often designed as compliant mechanisms to achieve micro displacement adjustment. In the traditional mechanisms, they are designed as integrated micromanipulation systems, and driven by piezoelectric ceramics. However, most of these researches only focuses on motion accuracy, due to the lack of consideration of large load problems, the application is greatly limited. To this end, a flexible support module (FSM), as well as its stiffness model, was presented in this paper. Combined with finite element method (FEM) of FSM, structure size optimization was also completed, successfully solved the problems of stress concentration and load of FSM in engineering application. Moreover, a dual vision-based measurement method was introduced, to verify the stiffness model and analyze the repetitive error of FSM. From this result, the prototype enabled 5 mm and 0.007 rad of working area with average error of 0.3192 mm and -0.0036 rad. The repeatable error is within 7%, and will decreased to 4% with internal stress released in 5~15min.

Recursive relation mainly describes the unique law satisfied by a sequence, so it plays an important role in almost all branches of mathematics. It is also one of the main algorithms commonly used in computer programming. This paper first introduces the concept of recursive relation and two common basic forms, then starts with the solution of linear recursive relation with non-homogeneous constant coefficients, gives a new solution idea, and gives a general proof. Finally, through an example, the general method and the new method given in this paper are compared and verified.

In order to study the effect of transmission distance on the performance of optical code division multiplexing passive optical network systems based on fiber Bragg Grating (FBG), a two-dimensional OCDMA-PON system based on cascaded FBGs is designed and constructed by using the VPI optical communication software. The spectrum, waveform and eye diagram of the output signal are analyzed by setting different fiber lengths in the system. The experimental results show that different fiber lengths have different effects on the system. Within a certain transmission distance, the longer the fiber length, the smaller the error generated by the output signal and the better the signal quality; beyond this certain distance, the quality of the output signal deteriorates sharply and the error generated becomes larger.

With the wide use of automatic identification system (AIS), a large amount of ship-related data has been provided for marine transportation analysis. Generally, AIS reports the type information of ships, but there are still many ships with type unknown in AIS data. It is necessary to develop algorithms which can identify ship type from AIS data. In this paper, we employ random forest to classify ships according to the static information from AIS messages. Moreover, the importance of static features is discussed, which explains the reason why some classes of ships are misclassified. The method of this paper is proved to be effective in ship classification using static information.

The uncertainty of positioning trajectory and the existence of abnormal trajectory points are major challenges in the field of navigation and positioning. The method that only relies on global positioning system (GPS) has a large error, and the traditional location algorithms still have shortcomings in location efficiency and accuracy. Due to the successful exploration of deep learning in the field of navigation and positioning, an accurate positioning algorithm based on improved gated recurrent unit (GRU) and multi-source data fusion was proposed in this paper. Firstly, interpolation was used to fuse GPS data and inertial measurement unit (IMU) data to overcome the defect of single data. Then, an improved gated recurrent unit was introduced to process the data fusion. On this basis, the attention mechanism (AM) was added to reduce the loss of historical information and strengthen the influence of important information, so as to achieve accurate positioning. Compared with other advanced algorithms with regard to L2 norm and time of training iteration, the training time of this algorithm is reduced by more than 27%, and the highest optimization rate is 12.08%. Therefore, this algorithm can obtain higher positioning accuracy in a shorter time.

The volume of the data has been rocketed since the new information era arrives. How to protect information privacy and detect the threat whenever the intrusion happens has become a hot topic. In this essay, we are going to look into the latest machine learning techniques (including deep learning) which are applicable in intrusion detection, malware detection, and vulnerability detection. And the comparison between the traditional methods and novel methods will be demonstrated in detail. Specially, we would examine the whole experiment process of representative examples from recent research projects to give a better insight into how the models function and cooperate. In addition, some potential problems and improvements would be illustrated at the end of each section.

To solve the operation sequencing problem in CAPP that is a difficult problem, combining the idea of genetic algorithm, an GA-Jaya algorithm is proposed to minimize the total cost. In the GA-Jaya, the population is initialized according to the procedure priority adjacency matrix which makes the population all meet the process priority relationship. Mutation iteration operator and two kinds of crossover iteration operator are designed for process sequence and processing resource evolution. The GA-Jaya algorithm is applied to a typical case, and compared with the existing genetic algorithm, particle swarm optimization algorithm and ant colony optimization algorithm. The results show that the average quality of the solution obtained by the GA-Jaya algorithm is better than the existing genetic algorithm, particle swarm optimization algorithm and ant colony optimization algorithm.

Robots play an irreplaceable role in the manufacturing industry. The use of machines instead of humans'activing. It is a long-term development trend, but it is difficult to realize the research of real robots, and the data that the robot can collect is less. In response to the above problems, this article mainly proposes a method of physical modeling using matlab/simulink. This method can realize research without robots, and can obtain more robot-related data. This method includes robot physical modeling, importing into simulink and packaging in simulink. This article is mainly aimed at IRB6700 robot to complete the simulation platform construction.

At present, with the development of the times and technology, in order to meet the needs of industrial production, robots have begun to be widely used in factories. Although the production efficiency of the robot is very high and relatively cheap compared to the labor cost, when an industrial robot fails, it will have a serious impact on the production capacity of the entire production line. Therefore, it is necessary to monitor various data of the robot. If some abnormal data can be found, it can perform effective predictive maintenance on the robot. This article mainly proposes a method for various data acquisition of ABB robots. The data acquisition interface is developed using C# and the dynamic link library is called. The data of ABB robots can be read in real time. It can realize real-time monitoring of each data of the robot.

Based on the high control performance requirement of laser-guided mortar control system, the permanent magnet synchronous motor (PMSM) is adopted in this paper as the electromechanical actuator of the system, the mathematical model of the motor is analyzed, and the vector control technology is adopted to achieve precise control of position, speed and torque of the electromechanical actuator. Aiming at the characteristics of non-linearity, strong coupling and large parameter changes of the system in flight, an improved fuzzy neural network PID control method is proposed by combining the classical PID control algorithm with fuzzy control and neural network control algorithm to realize the real-time tuning and optimization of PID parameters. The mathematical model of the electromechanical actuator control system is established and simulated. The results show that the fuzzy neural network PID control has good tracking performance, small amplitude error, and strong adaptability to load changes.

In this paper, an intelligent early warning scheme for rail transit line trip based on PSCADA system is proposed. The scheme takes into account the defects of low prediction accuracy and real-time prediction caused by the lack of power data in the traditional line trip prediction method. At the same time, a large number of power data generated by PSCADA system in the long-term application process are ignored in the field of rail transit[1]. Based on this situation, the prediction data set is constructed by combining the historical power data collected by PSCADA system in rail transit and the lightning weather data in traditional prediction methods. On this basis, the lightgbm machine learning intelligent algorithm is used to compare the similar support vector machine (SVM) and logistic regression algorithm to obtain a model with good prediction effect. In practical application, the real-time data set is constructed by using the real-time power data and real-time weather data collected by PSCADA system to predict, and an intelligent early warning system with the dual advantages of real-time and high accuracy is obtained.

For the problem of road surface condition recognition, this paper proposes a real-time tracking method to estimate road surface slope and adhesion coefficient. Based on the fusion of dynamics and kinematics, the current road slope of the vehicle which correct vertical load is estimated. The effect of the noise from dynamic and kinematic methods on the estimation results is removed by designing a filter. The normalized longitudinal force and lateral force are calculated by Dugoff tire model, and the Jacobian matrix of the vector function of the process equation is obtained by combining the relevant theory of EKF algorithm. The road adhesion coefficient is estimated finally. The effectiveness of the algorithm is demonstrated by analyzing the results under different operating conditions, such as docking road and bisectional road, using a joint simulation of Matlab/Simulink and Carsim.

sEMG(Surface electromyography) signal was widely applied in human-machine interactive field, especially in robotic arm control. In this paper, we built the Attention-MLP (Multilayer Perceptron) model to implement a type of continuous joint angle estimation method based on sEMG for six grasp movements, we tested this model on Ninapro dataset and the average Pearson correlation coefficient (CC) and the average root mean square error (RMSE) of the proposed Attention-MLP method achieved 0.812±0.02 and 10.51±1.98; the average CC and RMSE of this method are better than Sparse Pseudo-input Gaussian processes (SPGP), its average CC and RMSE are 12.14±2.30 and 0.727±0.07. Compared with the traditional method SPGP, our model performed better on continuously estimation of ten main hand joint angles under 6 grip movements.

Most real-world systems consist of a large number of interacting entities of many types. However, most of the current researches on systems are based on the assumption that the type of node or link in the network is unique. In other words, the network is homogeneous, containing the same type of nodes and links. Based on this assumption, differential information between nodes and edges is ignored. This paper firstly introduces the research background, challenges and significance of this research. Secondly, the basic concepts of the model are introduced. Thirdly, a novel type-sensitive LeaderRank algorithm is proposed and combined with distance rule to solve the importance ranking problem of content-associated heterogeneous graph nodes. Finally, the writer influence data set is used for experimental analysis to further prove the validity of the model.

Solving large-scale linear equations is of great significance in many engineering fields, such as weather forecasting and bioengineering. The classical computer solves the linear equations, no matter adopting the elimination method or Kramer's rule, the time required for solving is in a polynomial relationship with the scale of the equation system. With the advent of the era of big data, the integration of transistors is getting higher and higher. When the size of transistors is close to the order of electron diameter, quantum tunneling will occur, and Moore's Law will not be followed. Therefore, the traditional computing model will not be able to meet the demand. In this paper, through an in-depth study of the classic HHL algorithm, a small-scale quantum circuit model is proposed to solve a 2×2 linear equations, and the circuit diagram and programming are used to simulate and verify on the Origin Quantum Platform. The fidelity under different parameter values reaches more than 90%. For the case where the matrix to be solved is a sparse matrix, the quantum algorithm has an exponential speed improvement over the best known classical algorithm.

Aiming at the problem that most of the existing grid simplification algorithms for 3D models can not deal with a large number of boundary or non-popular grid models, this paper proposes a grid simplification algorithm for 3D models based on traditional algorithms. The algorithm mainly studies the geometric features of the model, considering the calculation methods and characteristics of edge shrinkage, and introduces the edge feature factors on the basis of the traditional algorithm, that is, the triangular area and side length factors of local area are introduced in the calculation of folding cost. In addition, the gaussian curvature characteristics of the 3D model are also included. Experimental results show that the proposed algorithm can keep the detail features of the mesh model well, and greatly reflect the quality and effect of mesh simplification after simplification.

The introduction of knowledge graph as the auxiliary information of recommendation system provides a new research idea for personalized intelligent recommendation. However, most of the existing knowledge graph recommendation algorithms fail to effectively solve the problem of unrelated entities, leading to inaccurate prediction of potential preferences of users. To solve this problem, this paper proposes a KG-IGAT model combining knowledge graph and graph attention network, and adds an interest evolution module to graph attention network to capture user interest changes and generate top-N recommendations. Finally, experimental comparison between the proposed model and other algorithms using public data sets shows that KG-IGAT has better recommendation performance.

Deep reinforcement learning (DRL) has greatly improved the intelligence of AI in recent years and the community has proposed several common software to facilitate the development of DRL. However, in robotics the utility of common DRL software is limited and the development is time-consuming due to the complexity of various robot software. In this paper, we propose a software engineering approach leveraging modularity to facilitate robot DRL development. The platform decouples learning environment into task, simulator and hierarchical robot modules, which in turn enables diverse environment generation using existing modules as building blocks, regardless of the underlying robot software details. Experimental results show that our platform provides composable environment building, introduces high module reuse and efficiently facilitates robot DRL.