A Robust Control Scheme for PMSM Based on Integral Reinforcement Learning
- 10 September 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 11 (1) , 4214-4223
Abstract
This paper proposes an integral reinforcement learning (IRL)-based H ∞ control algorithm for permanent magnet synchronous motor (PMSM) drives with excellent performance and guaranteed stability. Owing to its model-free nature, this algorithm achieves superior current regulation without any prior knowledge of motor parameters. Unlike the traditional offline reinforcement learning (RL) algorithms, which rely heavily on the quality of pre-sampled data for training, the proposed algorithm optimizes the control strategy online using real-time data. The convergence of the proposed algorithm is proved. Moreover, a simple actor-critic structure based neural network is employed to iteratively update the control policy by recursive least square (RLS) approach with low computational burden. The effectiveness of the proposed algorithm is experimentally verified on a 2kW PMSM prototype.Keywords
This publication has 26 references indexed in Scilit:
- Improved ROGI-FLL-Based Sensorless Model Predictive Current Control With MRAS Parameter Identification for SPMSM DrivesIEEE Journal of Emerging and Selected Topics in Power Electronics, 2022
- A New Model-Free Deadbeat Predictive Current Control for PMSM Using Parameter-Free Luenberger Disturbance ObserverIEEE Journal of Emerging and Selected Topics in Power Electronics, 2022
- A Deep Q-Learning Direct Torque Controller for Permanent Magnet Synchronous MotorsIEEE Open Journal of the Industrial Electronics Society, 2021
- ${H_\infty}$ Robust Control of Permanent Magnet Synchronous Motor Based on PCHDIEEE Access, 2019
- Application of Off-policy Integral Reinforcement Learning for H∞, Input Constrained Control of Permanent Magnet Synchronous Machine2019
- Off-Policy Reinforcement Learning for $ H_\infty $ Control DesignIEEE Transactions on Cybernetics, 2014
- Integral Reinforcement Learning for Linear Continuous-Time Zero-Sum Games With Completely Unknown DynamicsIEEE Transactions on Automation Science and Engineering, 2014
- Adaptive dynamic programming for online solution of a zero-sum differential gameControl Theory and Technology, 2011
- H∞-Optimal Control and Related Minimax Design Problems2008
- L2-Gain and Passivity Techniques in Nonlinear Control1996