融合无模型强化学习的永磁同步电机混沌抗扰控制

谭浚楷; 薛霜思; 郭子航; 曹晖

doi:10.11959/j.issn.2096-6652.202523

您当前的位置：

首页 >

文章列表页 >

融合无模型强化学习的永磁同步电机混沌抗扰控制

学术论文 | 更新时间：2025-11-18

- 融合无模型强化学习的永磁同步电机混沌抗扰控制
- Chaotic disturbance rejection control of permanent magnet synchronous motor integrating model-free reinforcement learning
- 智能科学与技术学报 2025年7卷第3期页码：304-315
- 作者机构：
  
  1.西安交通大学电气工程学院，陕西西安 710049
  2.西安交通大学电力设备电气绝缘国家重点实验室，陕西西安 710049
- 作者简介：
  
  [ "谭浚楷（2001- ），男，西安交通大学电气工程学院硕士生，主要研究方向为自适应动态规划与强化学习、逆强化学习等。" ]
  [ "薛霜思（1992- ），男，博士，西安交通大学电气工程学院助理教授，主要研究方向为数据驱动网络控制、智能系统控制。" ]
  [ "郭子航（2001- ），男，西安交通大学电气工程学院硕士生，主要研究方向为自适应神经网络控制、多智能体控制等。" ]
  [ "曹晖（1978- ），男，博士，西安交通大学电气工程学院教授，主要研究方向为数据驱动系统、智能系统控制。" ]
- 基金信息：
  
  中国博士后科学基金(2024M762602)
- DOI：10.11959/j.issn.2096-6652.202523
  中图分类号： TP39
- 收稿：2025-03-07，
  
  修回：2025-05-26，
  
  纸质出版：2025-09-15
- 稿件说明：
移动端阅览
谭浚楷,薛霜思,郭子航等.融合无模型强化学习的永磁同步电机混沌抗扰控制[J].智能科学与技术学报,2025,07(03):304-315.

TAN Junkai,XUE Shuangsi,GUO Zihang,et al.Chaotic disturbance rejection control of permanent magnet synchronous motor integrating model-free reinforcement learning[J].Chinese Journal of Intelligent Science and Technology,2025,07(03):304-315.
谭浚楷,薛霜思,郭子航等.融合无模型强化学习的永磁同步电机混沌抗扰控制[J].智能科学与技术学报,2025,07(03):304-315. DOI： 10.11959/j.issn.2096-6652.202523.

TAN Junkai,XUE Shuangsi,GUO Zihang,et al.Chaotic disturbance rejection control of permanent magnet synchronous motor integrating model-free reinforcement learning[J].Chinese Journal of Intelligent Science and Technology,2025,07(03):304-315. DOI： 10.11959/j.issn.2096-6652.202523.

摘要

针对永磁同步电机（permanent magnet synchronous motor，PMSM）控制中出现的由非线性震荡引发的混沌行为，提出了一种利用电机历史运行数据迭代获取最优控制策略的无模型强化学习方法，该方法解决了不同工作环境下由负载扰动不确定、电机模型参数不确定带来的问题。首先针对外部负载扭矩的不确定性，建立电机最优控制与外部最坏扰动之间的零和博弈，设计迭代形式的黎卡提方程，基于该方程进一步构建了一种基于模型的抗干扰最优控制器。在该控制器基础上，引入无模型强化学习思想，设计了PMSM混沌现象的数据驱动稳定方法，利用电机历史运行数据实现优化求解无模型抗扰动稳定控制器，实现对电机混沌行为的抗扰稳定控制。最后通过多个数值仿真实验对比验证了所设计方法的性能。实验结果显示，在外界负载扰动不确定的情况下，提出的方法相较传统有限时间控制方法在累计资源消耗损失性能方面提升了39.04%，在电机模型参数不确定的情况下，提出的方法相比线性二次型调节器的控制成功率提高10.71%。

Abstract

For the chaotic behavior of nonlinear oscillations occurring in the control of permanent magnet synchronous motor (PMSM)

a data-driven model-free reinforcement learning method was investigated

which only utilized the historical state data of the motor to obtain the optimal controller. This method solved the problem of uncertain external load and uncertain motor dynamic under varying and unknown working conditions. Firstly

to solve the uncertainty of external torque load

a zero-sum game between the controller and external disturbance was constructed. By redesigning the Riccati equation in iteration form

a model-based robust optimal controller was obtained. Based on this controller

by introducing the model-free reinforcement learning method

a data-driven stabilization method for chaotic phenomena in PMSM was proposed

which obtained a model-free robust optimal controller by learning from the historical operation data. Finally

the performance of the designed method was verified by comparing several simulations

and the results showed that the energy effort saving performance of our proposed method was improved by 39.04% compared with the traditional finite-time control method under the uncertainty of external load perturbation

and the success rate was improved by 10.71% compared with the linear quadratic regulator under the uncertainty of the motor model parameters.

关键词

Keywords

references

刘凌, 司杰文, 林起联. 支持向量机预测可变参数的永磁同步电机快速终端滑模控制[J]. 西安交通大学学报, 2021, 55(6): 53-60.

LIU L, SI J W, LIN Q L. A fast terminal sliding mode control of permanent magnet synchronous motor with variable parameters predicted by SVM[J]. Journal of Xi'an Jiaotong University, 2021, 55(6): 53-60.

ZHAO J G, YANG C Y, GAO W N, et al. Reinforcement learning and optimal control of PMSM speed servo system[J]. IEEE Transactions on Industrial Electronics, 2023, 70(8): 8305-8313.

刘思源, 刘凌, 靳东松. 永磁同步电机新型有效磁链固定时滑模扰动辨识策略[J]. 西安交通大学学报, 2023, 57(9): 144-151.

LIU S Y, LIU L, JIN D S. A novel active flux fixed-time sliding mode disturbance identification method for permanent magnet synchronous motor[J]. Journal of Xi'an Jiaotong University, 2023, 57(9): 144-151.

邵保领, 刘凌, 胡铭遥. 面向永磁同步电机负载转矩观测的自抗扰改进滑模控制[J]. 西安交通大学学报, 2024, 58(3): 204-211.

SHAO B L, LIU L, HU M Y. Improved sliding mode control with disturbance rejection for load torque observation of permanent magnet synchronous motors[J]. Journal of Xi'an Jiaotong University, 2024, 58(3): 204-211.

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.

LI Z, HU G D, CUI J R, et al. Sliding-mode variable structure control with integral action for permanent magnet synchronous motor[J]. Proceedings of the CSEE, 2014, 34(3): 431-437.

毛北行, 王东晓. 分数阶永磁同步电机混沌系统自适应滑模同步[J]. 浙江大学学报(理学版), 2023, 50(5): 564-568.

MAO B X, WANG D X. Self-adaptive sliding mode synchronization of fractional-order permanent magnet synchronization motor chaotic systems[J]. Journal of Zhejiang University (Science Edition), 2023, 50(5): 564-568.

王江彬, 刘崇新. 4阶混沌电力系统的协同控制方法[J]. 西安交通大学学报, 2020, 54(1): 26-31.

WANG J B, LIU C X. Synergetic control method for four-dimensional chaotic power system[J]. Journal of Xi'an Jiaotong University, 2020, 54(1): 26-31.

李宁洲, 邱思旋, 卫晓娟, 等. 一类永磁同步电机混沌运动的GWO-RBFNN双参协同智能优化控制[J]. 哈尔滨理工大学学报, 2024, 29(3): 28-36.

LI N Z, QIU S X, WEI X J, et al. GWO-RBFNN dual-parameter collaborative intelligent optimal control of chaotic motion of a class of permanent magnet synchronous motor[J]. Journal of Harbin University of Science and Technology, 2024, 29(3): 28-36.

REN H P, LIU D. Nonlinear feedback control of chaos in permanent magnet synchronous motor[J]. IEEE Transactions on Circuits and Systems Ⅱ: Express Briefs, 2006, 53(1): 45-50.

WEI Q, WANG X Y, HU X P. Optimal control for permanent magnet synchronous motor[J]. Journal of Vibration and Control, 2014, 20(8): 1176-1184.

DO T D. Optimal control design for chaos suppression of PM synchronous motors[C]//Proceedings of the 2016 2nd International Conference on Control Science and Systems Engineering (ICCSSE). Piscataway: IEEE Press, 2016: 88-92.

HOU Y Y. Finite-time chaos suppression of permanent magnet synchronous motor systems[J]. Entropy, 2014, 16(4): 2234-2243.

YAN T F, LIU Q Y, DOU B W, et al. An adaptive dynamic programming method for torque ripple minimization of PMSM[J]. Journal of Industrial & Management Optimization, 2021, 17(2): 827-839.

陈强, 南余荣, 邢科新. 基于扩张状态观测器的永磁同步电机混沌系统自适应滑模控制[J]. 物理学报, 2014, 63(22): 117-124.

CHEN Q, NAN Y R, XING K X. Adaptive sliding-mode control of chaotic permanent magnet synchronous motor system based on extended state observer[J]. Acta Physica Sinica, 2014, 63(22): 117-124.

ZHANG J, SUN J H, GU H, et al. Control of PMSM chaos using backstepping-based adaptive fuzzy method in the presence of uncertainty and disturbance[J]. Systems Science & Control Engineering, 2021, 9(1): 673-688.

谢东燊, 杨俊华, 熊锋俊, 等. 永磁直线同步电机解耦自适应滑模混沌控制[J]. 计算机仿真, 2019, 36(5): 263-268.

XIE D S, YANG J H, XIONG F J, et al. Decoupling adaptive sliding mode chaos control of permanent magnet linear synchronous motor[J]. Computer Simulation, 2019, 36(5): 263-268.

毛北行. 分数阶整数阶多混沌系统的自适应滑模同步[J]. 中山大学学报(自然科学版), 2020, 59(4): 128-133.

MAO B X. Self-adaptive sliding mode synchronization of fractional-order and integer-order uncertain multi-chaotic systems[J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2020, 59(4): 128-133.

ZHANG J X, WANG S L, ZHOU P, et al. Novel prescribed performance-tangent barrier Lyapunov function for neural adaptive control of the chaotic PMSM system by backstepping[J]. International Journal of Electrical Power & Energy Systems, 2020, 121: 105991.

YIN X, SHE J H, LIU Z T, et al. Chaos suppression in speed control for permanent-magnet-synchronous-motor drive system[J]. Journal of the Franklin Institute, 2020, 357(18): 13283-13303.

SAHOO A, XU H, JAGANNATHAN S. Near optimal event-triggered control of nonlinear discrete-time systems using neurodynamic programming[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 27(9): 1801-1815.

廖泽华, 梁子钰, 周天民, 等. 基于平行控制的离散非线性系统的事件触发近似最优控制[J]. 智能科学与技术学报, 2021, 3(4): 435-443.

LIAO Z H, LIANG Z Y, ZHOU T M, et al. Parallel control-based event-driven approximate optimal control of discrete-time nonlinear systems[J]. Chinese Journal of Intelligent Science and Technology, 2021, 3(4): 435-443.

JIANG Y, JIANG Z P. Computational adaptive optimal control for continuous-time linear systems with completely unknown dynamics[J]. Automatica, 2012, 48(10): 2699-2704.

LU J W, WEI Q L, ZHOU T M, et al. Event-triggered near-optimal control for unknown discrete-time nonlinear systems using parallel control[J]. IEEE Transactions on Cybernetics, 2023, 53(3): 1890-1904.

LU J W, WEI Q L, WANG F Y. Parallel control for nonzero-sum games with completely unknown nonlinear dynamics via reinforcement learning[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2025, 55(4): 2884-2896.

PING Z W, JIA Y J, XIONG B G, et al. Optimal output regulation for PMSM speed servo system using approximate dynamic programming[J]. Science China Information Sciences, 2023, 66(7): 170206.

EL-SOUSY F F M, AMIN M M, AL-DURRA A. Adaptive optimal tracking control via actor-critic-identifier based adaptive dynamic programming for permanent-magnet synchronous motor drive system[J]. IEEE Transactions on Industry Applications, 2021, 57(6): 6577-6591.

FAN Z X, LI S H, LIU R J. ADP-based optimal control for systems with mismatched disturbances: a PMSM application[J]. IEEE Transactions on Circuits and Systems Ⅱ: Express Briefs, 2023, 70(6): 2057-2061.

SHI T Y, ZOU W C, GUO J, et al. Adaptive speed regulation for permanent magnet synchronous motor systems with speed and current constraints[J]. IEEE Transactions on Circuits and Systems Ⅱ: Express Briefs, 2024, 71(4): 2079-2083.

EQRA N, VATANKHAH R, EGHTESAD M. A novel adaptive multi-critic based separated-states neuro-fuzzy controller: architecture and application to chaos control[J]. ISA Transactions, 2021, 111: 57-70.

夏丽娜, 李擎, 宋睿卓, 等. 未知异构多智能体系统无模型自适应动态规划同步控制[J]. 智能科学与技术学报, 2021, 3(4): 444-448.

XIA L N, LI Q, SONG R Z, et al. Synchronization control of unknown heterogeneous multi-agent system via model-free adaptive dynamic programming[J]. Chinese Journal of Intelligent Science and Technology, 2021, 3(4): 444-448.

李金娜, 程薇燃. 基于强化学习的数据驱动多智能体系统最优一致性综述[J]. 智能科学与技术学报, 2020, 2(4): 327-340.

LI J N, CHENG W R. An overview of optimal consensus for data driven multi-agent system based on reinforcement learning[J]. Chinese Journal of Intelligent Science and Technology, 2020, 2(4): 327-340.

李涛, 魏庆来. 基于深度强化学习的智能暖气温度控制系统[J]. 智能科学与技术学报, 2020, 2(4): 348-353.

LI T, WEI Q L. Intelligent heating temperature control system based on deep reinforcement learning[J]. Chinese Journal of Intelligent Science and Technology, 2020, 2(4): 348-353.

ZHANG L J, CHEN Y. Finite-time adaptive dynamic programming for affine-form nonlinear systems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2025, 36(2): 3573-3586.

TAN J, XUE S, GUAN Q, et al. Finite-time safe reinforcement learning control of multi-player nonzero-sum game for quadcopter systems[J]. Information Sciences, 2025, 712: 122117.

浏览量

135

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

DK-Port：基于大语言模型和强化学习的港口自动驾驶仿真环境构建与验证

基于改进DQN算法的复杂海战场路径规划方法

机器人自动轴孔装配研究进展

基于动态自选择参数共享的合作多智能体强化学习算法

异构多智能体系统的输出同步：一个基于数据的强化学习方法