Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章检索

首页 >> 文章检索 >> 最新索引

基于优化BP神经网络PID的永磁同步电机控制研究

来源:电工电气发布时间:2024-12-02 10:02 浏览次数:34

基于优化BP神经网络PID的永磁同步电机控制研究

王雷,王育安,崔玉鑫,眭晓倩,王毅
(河北科技大学 电气工程学院,河北 石家庄 050018)
 
    摘 要:针对传统 PID 控制在永磁同步电机控制系统中未能实现精准控制的问题,提出了一种基于改进蜣螂优化算法的 BP 神经网络 PID 控制器,该控制器由 BP 神经网络通过自适应方法来调整权重系数,解决了 PID 无法在线调节参数的缺点。针对 BP 神经网络在进行反向传播时陷入局部最优的概率较大,引入蜣螂优化算法通过适应度值不断更新 BP 神经网络核心参数,从而提高 BP 神经网络的优化速率。对于蜣螂优化算法中存在初始种群质量不高及搜索能力不足等问题,对蜣螂优化算法进行混合策略优化,大大提升了蜣螂优化算法求解效率和精度。实验结果表明该改进蜣螂优化算法可以有效地提高控制系统的响应速度,减小超调量,在转速和负载突变的情况下都有较强的鲁棒性。
    关键词: 永磁同步电机;改进蜣螂优化算法;BP 神经网络;PID 控制
    中图分类号:TM315     文献标识码:A     文章编号:1007-3175(2024)11-0030-07
 
Research on Control of Permanent Magnet Synchronous Motor Based on
PID of Optimized BP Neural Network
 
WANG Lei, WANG Yu-an, CUI Yu-xin, SUI Xiao-qian, WANG Yi
(College of Electrical Engineering, Hebei University of Science and Technology, Shijiazhuang 050018, China)
 
    Abstract: In view of the problem that traditional PID control fails to achieve accurate control in the permanent magnet synchronous motor control system, a PID controller of BP neural network based on improved dung beetle optimization algorithm is proposed. The controller uses BP neural network to adjust the weight coefficient by adaptive method, which solves the shortcoming that PID can not adjust parameters online.Then, aiming at the high probability of BP neural network falling into local optimum when performing back propagation, the improved dung beetle optimization algorithm is introduced to continuously update the core parameters of the BP neural network through the adaptive value, so as to improve the optimization rate of the BP neural network. Furthermore, addressing issues like low initial population quality and inadequate search capability in the improved dung beetle optimization algorithm, a hybrid strategy is implemented to optimize the improved dung beetle optimization algorithm, which greatly improves the solving efficiency and accuracy of the improved dung beetle optimization algorithm. Experimental results demonstrate that the improved dung beetle optimization algorithm effectively enhances the response speed of the control system, reduces overshoot, and it has strong robustness to the case of speed and load sudden change.
    Key words: permanent magnet synchronous motor; improved dung beetle optimization algorithm; BP neural network; PID control
 
参考文献
[1] 李政,胡广大,崔家瑞,等. 永磁同步电机调速系统的积分型滑模变结构控制[J] . 中国电机工程学报,2014,34(3) :431-437.
[2] 刘景林,公超,韩泽秀,等. 永磁同步电机闭环控制系统数字 PI 参数整定[J] . 电机与控制学报,2018,22(4) :26-32.
[3] 王松. 永磁同步电机的参数辨识及控制策略研究[D] .北京:北京交通大学,2011.
[4] 周佳,卢少武,周凤星. 基于 RBF 神经网络的永磁同步电机速度 PI-IP 控制[J]. 组合机床与自动化加工技术,2017(1) :116-118.
[5] XUE Jiankai, SHEN Bo.Dung beetle optimizer:A new meta-heuristic algorithm for global optimization[J].The Journal of Supercomputing,2023,79(7) :7305-7336.
[6] 赵强,王昊洁,谢春丽. 基于改进蜣螂优化算法的永磁同步电机参数辨识[J]. 重庆交通大学学报(自然科学版), 2024,43(6) :102-108.
[7] 邵明玲. 永磁同步电机的神经网络模糊自适应控制[D].青岛:青岛大学,2016.
[8] HARNEFORS L, PIETILAINEN K, GERTMAR L.Torquemaximizing field-weakening control:Design,analysis,and parameter selection[J].IEEE Transactions on Industrial Electronics,2001,48(1) :161-168.
[9] 王同旭,马鸿雁,聂沐晗. 电梯用永磁同步电机 BP 神经网络 PID 调速控制方法的研究[J] . 电工技术学报,2015,30(S1) :43-47.
[10] 叶德住. 基于 BP 神经网络的永磁同步电机控制[J] .微电机,2016,49(11) :57-61.
[11] 郭琴,郑巧仙. 多策略改进的蜣螂优化算法及其应用[J].计算机科学与探索,2024,18(4) :930-946.
[12] TIZHOOSH H R.Opposition-based learning:A new schemefor machine intelligence[C]//International Conference on Computational Intelligence for Modelling , Control and Automation, and International Conference on Intelligent Agents, Web Technologies and Internet Commerce,2005.
[13] WANG Xingyuan, JIN Canqi.Image encryption using game of life permutation and PWLCM chaotic system[J].Optics Communications,2012,285(4) :412-417.
[14] 郭振洲,王平,马云峰,等. 基于自适应权重和柯西变异的鲸鱼优化算法[J] . 微电子学与计算机,2017,34(9) :20-25.
[15] 李飞,陈勇弟,魏小城,等. 基于改进麻雀搜索算法的微电网优化调度[J] . 智能计算机与应用,2024,14(5) : 150-156.
[16] 付接递,李振东,郭辉. 基于教与学和逐维柯西变异的鲸鱼优化算法[J] . 计算机工程与科学,2023,45(5) :940-950.
[17] 付华,刘昊. 多策略融合的改进麻雀搜索算法及其应用[J]. 控制与决策,2022,37(1) :87-96.