Table of Content

15 September 2022 Volume 35 Issue 9

  

An Improved Obstacle Detection Method for AGV

YANG Yingying,LIU Xiang,SHI Yunyu

Electronic Science and Technology. 2022, 35(9):  1-6.  doi:10.16180/j.cnki.issn1007-7820.2022.09.001

Abstract ( 276 )   HTML ( 15 )   PDF (1193KB) ( 52 )  

Figures and Tables | References | Related Articles | Metrics

To solve the problem that AGV obstacle detection algorithm performs poorly under the interference of uneven illumination and background texture in smart factories, this study proposes an improved Canny operator for obstacle detection. The method achieves the optimization of AGV obstacle detection in terms of color space, filtering method, gradient direction and adaptive threshold. Through Lab color space conversion, the b component is extracted and then filtered. The improved median filter and bilateral filter are merged to replace the Gaussian filter in the traditional Canny operator, which reduces the loss of edge details while achieving noise reduction, and improves the speed of the algorithm. The edge information is enhanced by increasing the gradient direction, and adaptive thresholding is obtained using Otsu algorithm. Experiments show that the proposed method can improve the accuracy of edge detection and reduce noise interference, thus achieving stable detection of obstacles.

Multi-Objective Optimization of Active Distribution Network Based on Particle Swarm Optimization

SHI Zhenli,WEI Yewen

Electronic Science and Technology. 2022, 35(9):  7-14.  doi:10.16180/j.cnki.issn1007-7820.2022.09.002

Abstract ( 454 )   HTML ( 11 )   PDF (1041KB) ( 84 )  

Figures and Tables | References | Related Articles | Metrics

With the large-scale grid integration of distributed renewable energy, the traditional distribution network has gradually developed from a single flow to a complex two-way flow. In view of the problem that the traditional dispatching method in active distribution network technology cannot be directly applied, this study explores improvement measures from two aspects of intelligent algorithm and optimization model. On the basis of considering the relevance of “source network load and storage”, aiming at improving the effect of peak shaving and valley filling, improving the economy of distribution network, and reducing the loss of distribution network, the forecast of wind and solar output has been carried out to improve the validity of the data, and a two-stage two-layer joint optimal dispatch model has been established. The study analyzes the advantages and disadvantages of the traditional particle swarm algorithm, and proposes to use the improved HE-MOPSO algorithm to solve the model. By solving the ZDT1~4 test function and using the extended IEEE33 node to perform the simulation calculation, the experimental results proved the superiority of the improved algorithm and model.

Photovoltaic Maximum Power Point Tracking Based on Improved Differential Evolution Algorithm

GE Chuanjiu,WU Peng,JIN Junzhe,DONG Xiangxiang,LOU Qikai

Electronic Science and Technology. 2022, 35(9):  15-21.  doi:10.16180/j.cnki.issn1007-7820.2022.09.003

Abstract ( 176 )   HTML ( 3 )   PDF (1023KB) ( 36 )  

Figures and Tables | References | Related Articles | Metrics

For the photovoltaic maximum power point tracking algorithm, the convergence of traditional algorithms will cause the problem of trapping into local extremes, while the convergence of intelligent algorithms is slow, and may produce large amplitude voltage oscillation in the process of convergence. In view of the above-mentioned problems, the study proposes corresponding improvement strategies. In the iterative process of particle swarm optimization, the individual order of the population is added to suppress the large amplitude voltage oscillation, the influence of the particles with poor fitness on the speed update is eliminated, and the convergence speed is improved through combining the competitive relationship during population update in the differential evolution algorithm. Single-peak and multi-peak examples are used to simulate the proposed strategy, and the maximum photovoltaic power obtained is 60 W and 122 W, respectively. The simulation results show that compared with particle swarm optimization and differential evolution algorithm, the proposed algorithm improves the convergence speed by 52.22% and 61.60% respectively, and reduces the mid-term voltage fluctuation amplitude by 15% and 30%, respectively.

Photovoltaic Grid-Connected System Based on Adaptive VSG Control

SUN Sinan,HAO Zhenghang

Electronic Science and Technology. 2022, 35(9):  22-29.  doi:10.16180/j.cnki.issn1007-7820.2022.09.004

Abstract ( 421 )   HTML ( 2 )   PDF (998KB) ( 43 )  

Figures and Tables | References | Related Articles | Metrics

In traditional VSG control, the introduction of the inertia moment and damping coefficient of the synchronous generator into the photovoltaic inverter will result in poor dynamic adjustment capabilities of the photovoltaic grid-connected system. In view of this problem, this study proposes an improved VSG control strategy. The relationship between power angle and active power and the transient process of the rotor angular frequency oscillation period are discussed through the mathematical model of the converter. The principles for selecting moment of inertia and damping coefficient are given, and the rate of change of moment of inertia and angular frequency as well as damping coefficient and angular frequency are analyzed. The relationship between the changes leads to the improvement strategy of photovoltaic grid-connected under the control of adaptive VSG. The simulation results show that the photovoltaic grid-connected system based on the adaptive VSG control strategy has a good dynamic response, which proves the effectiveness and superiority of the improved strategy.

A Video Synopsis Generation Model Incorporating Object Speed Change Mechanism

NIU Jiafeng,SHI Yunyu,LIU Xiang,LI Rensi

Electronic Science and Technology. 2022, 35(9):  30-36.  doi:10.16180/j.cnki.issn1007-7820.2022.09.005

Abstract ( 198 )   HTML ( 2 )   PDF (1950KB) ( 22 )  

Figures and Tables | References | Related Articles | Metrics

Video synopsis is an effective technique for surveillance video browsing and storage. Existing video synopsis generation methods are prone to cause rearrangement objects collision and chronological disorder under the limitation of compression ratio. To solve this problem, a video synopsis generation model incorporating object speed change mechanism is proposed in this study. In the energy function of object rearrangement, object speed variable is added in addition to object start position variable, so that object start position and speed can be changed simultaneously to avoid collision and chronological disorder problems. Then the Markov chain Monte Carlo random sampling algorithm is used to solve the optimal value of the energy function, and the optimal solution of the target rearrangement scheme is obtained. Experiments show that when the compression rate is the same, compared with other methods, the summary video generated by the proposed model has fewer problems with target collision and timing disorder.

A Research on Distance Measurement Between Trains in Rail Transit Based on Machine Vision

BI Jiazhen,SHEN Tuo,ZHANG Xuanxiong

Electronic Science and Technology. 2022, 35(9):  37-43.  doi:10.16180/j.cnki.issn1007-7820.2022.09.006

Abstract ( 402 )   HTML ( 8 )   PDF (1587KB) ( 47 )  

Figures and Tables | References | Related Articles | Metrics

A safe distance between two moving trains is an important condition to avoid train rear-end collision. Since the image data obtained by machine vision is rich in information and can be integrated in many aspects based on the collected images, this study proposes a distance measurement method based on monocular machine vision. This method uses the constant distance between the two tracks of the train (1 435 mm) as a benchmark to estimate the distance between trains. The images collected by the monocular camera are processed and analyzed by the convolution neural network to extract the track features. Based on the existing small hole imaging principle, the mapping relationship between the world coordinate system and the pixel coordinate system is derived, so as to optimize the calculation formula of the distance between trains. The experimental results show that the error rate of the system is less than 6%, and the measurement time of the system is within 40 ms, indicating that the method realizes the effective fusion and integration of ranging and other information obtained in the image, and can be used to judge the braking distance of the train.

Optimization of the Internal Memory Architecture of Heterogeneous Multi-Core SoC Processors

ZHANG Xuan,ZHANG Duoli,SONG Yukun

Electronic Science and Technology. 2022, 35(9):  44-51.  doi:10.16180/j.cnki.issn1007-7820.2022.09.007

Abstract ( 304 )   HTML ( 7 )   PDF (1675KB) ( 49 )  

Figures and Tables | References | Related Articles | Metrics

The performance of microprocessors has been greatly improved by the development of heterogeneous multi-core technology. The bandwidth difference between the processor and external memory severely limits the performance of the processor, and the "Memory Wall" problem is becoming increasingly serious. For a heterogeneous multi-core SoC system in high-density computing, this study proposes a set of memory design scheme. The solution increases memory access bandwidth and reduces the frequency of accessing external memory by reusing some local free memory resources that have been idle for a long time as shared L2 cache. Meanwhile, the distributed high-speed shared L2 cache combined with the hierarchical storage structure of multi-channel parallel access to external storage alleviates the speed difference between system processing data and external storage, improves data access efficiency, and optimizes system performance. In terms of resource consumption and computing efficiency, the proposed design saves 69.36% of on-chip SRAM resources compared with ordinary L2 cache, provides 41.2% speedup ratio compared with non-cache structure, and reduces the overall task calculation time by about 40.6% on average.

Glasses Virtual Try-on System Based on 3D Frame Modeling

ZHANG Xindong,FU Dongxiang

Electronic Science and Technology. 2022, 35(9):  52-57.  doi:10.16180/j.cnki.issn1007-7820.2022.09.008

Abstract ( 534 )   HTML ( 19 )   PDF (1493KB) ( 72 )  

Figures and Tables | References | Related Articles | Metrics

In the virtual try-on system, virtual glasses need to be superimposed on the face image accurately, which determines the experience of the virtual try-on system. The key technology is to construct a real glasses model and quickly and accurately estimate the 3D face pose of the image. In response to this technical requirement, a virtual try-on technology combining 3D mirror frame modeling and head space pose estimation based on facial feature detection is proposed and a virtual try-on system is implemented. First, the method uses ERT cascade regression to locate the facial feature points of the face images. Then, the VTK visualization tool is adopted to construct a 3D mirror frame, and the processed mirror frame image is accurately fused to the face image according to the information of the face feature points and the rotation information of the head. Finally, the accuracy of the pose estimation method is tested using the AFLW data set. The results show that the algorithm has high precision and fast speed. It can still quickly and accurately realize three-dimensional and multi-angle virtual try-on under complicated conditions of large angles, many background interferences and poor light conditions, which basically meets the requirements of virtual try-on technology.

Workflow Scheduling Algorithm Based on Mobile Perception in Mobile Edge Environment

WEI Zefeng,ZHOU Yuanyuan

Electronic Science and Technology. 2022, 35(9):  58-64.  doi:10.16180/j.cnki.issn1007-7820.2022.09.009

Abstract ( 172 )   HTML ( 5 )   PDF (1389KB) ( 38 )  

Figures and Tables | References | Related Articles | Metrics

In the mobile edge computing environment, users can offload local computing-intensive tasks to the edge server, thereby shortening the completion time of the workflow and saving equipment energy consumption. However, many studies have neglected the influence of network connection changes caused by user movement on workflow scheduling. In view of the unreasonable unloading problem in the existing algorithms, a workflow scheduling algorithm MaWS is proposed. The algorithm predicts the user's movement trajectory to obtain a set of future communicable base stations, and incorporates genetic algorithms to formulate reasonable task execution sequence and execution position. The simulation results show that compared with algorithms such as HEFT and Greedy, the MaWS algorithm can effectively shorten the completion time of the workflow by 10% to 15% and reduce the energy consumption of the equipment by 8% to 13%, which indicates that the proposed MaWS algorithm is an effective solution for workflow scheduling under mobile edge computing.

Vector Control of Induction Motor Based on Adaptive Fuzzy Neural Network

JIN Aijuan,SHAO Feixuan,YAN Ziguang

Electronic Science and Technology. 2022, 35(9):  65-73.  doi:10.16180/j.cnki.issn1007-7820.2022.09.010

Abstract ( 285 )   HTML ( 3 )   PDF (1336KB) ( 44 )  

Figures and Tables | References | Related Articles | Metrics

In view of the problem that the induction motor has fixed parameters and is easy to overshoot in the traditional PI control, a method based on the adaptive fuzzy neural network PI control and the full-order adaptive observer is proposed in this study. According to the mathematical model of the induction motor, the model of the full-order adaptive observer is established, and the stability analysis and design are carried out using the Lyapunov stability theory, and the speed adaptive law is deduced. The motor speed outer loop PI is adjusted and optimized online by an adaptive fuzzy neural network inference system. Compared with traditional control schemes, this method is easy to implement, can effectively improve control accuracy, suppress external disturbances, and save sensor costs. MATLAB/Simulink simulation experiments show that the proposed scheme not only improves the dynamic performance of the speed sensorless induction motor vector control system, but also reduces the influence of external load disturbances, and improves the system's adaptability and robustness.

Numerical Simulation and Thermal Cycle Analysis of MAG Welding Temperature Field Based on ABAQUS

MING Can,MA Chunwei

Electronic Science and Technology. 2022, 35(9):  74-78.  doi:10.16180/j.cnki.issn1007-7820.2022.09.011

Abstract ( 835 )   HTML ( 6 )   PDF (2551KB) ( 69 )  

Figures and Tables | References | Related Articles | Metrics

In order to control the production cost of MAG welding and provide accurate technical support for MAG welding process, finite element software is used to simulate the welding process. Based on ABAQUS finite element analysis software, the temperature field of 8 mm-thick Q235 MAG is numerically simulated, the loading process of welding heat source is simulated, and the temperature field of each stage and the welding thermal cycle law of vertical and parallel points near the weld are analyzed. The results show that the temperature field is steady, the welding thermal cycle is stable, the peak temperature time away from the weld is slightly delayed, and the temperature gradient decreases. The numerical simulation results are in good agreement with the actual changes of the welding temperature field.

Establishment of a Predictive Model of the Process Parameters of Secondary Moisturizing Based on BP Neural Network

ZHOU Yongchang,HUANG Yayu

Electronic Science and Technology. 2022, 35(9):  79-86.  doi:10.16180/j.cnki.issn1007-7820.2022.09.012

Abstract ( 168 )   HTML ( 5 )   PDF (1054KB) ( 46 )  

Figures and Tables | References | Related Articles | Metrics

In this study, the influence of the process parameter setting of the hot-air leaf moisturizer on the quality index of the exit leaf during the secondary leaf conditioning of threshing and redrying is studied, and the corresponding prediction model is established. A BP neural network prediction model is established based on the characteristics of the secondary leaf conditioning process data. The current popular neural network writing framework TensorFlow's high-level API interface is called to construct the neural network structure. The activation function, optimizer, number of hidden layer neurons and other key parameters are gradually adjusted in the neural network structure to make the prediction result of the test set reach the best state. By inputting the parameters of the front steam nozzle pressure, front-end water flow rate, hot air temperature, return air temperature, feed blade temperature, and feed blade moisture combination, the two key tobacco leaf evaluation indicators, namely, outlet leaf moisture and temperature, are predicted. According to the mean square error, root mean square error, and average absolute error of the prediction results, it is concluded that when the number of neurons in the hidden layer is 7, the activation function selects ReLU, and the optimizer selects RMSprop, the effect is the best.