Deep reinforcement learning for permanent magnet synchronous motor speed control systems
- PDF / 1,596,491 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 88 Downloads / 331 Views
(0123456789().,-volV)(0123456789(). ,- volV)
ORIGINAL ARTICLE
Deep reinforcement learning for permanent magnet synchronous motor speed control systems Zhe Song1,2
•
Jun Yang1,3 • Xuesong Mei1 • Tao Tao1 • Muxun Xu1
Received: 21 May 2020 / Accepted: 7 September 2020 Ó Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract The permanent magnet synchronous motor (PMSM) servo system is widely applied in many industrial fields due to its unique advantages. In this paper, we study the deep reinforcement learning (DRL) speed control strategy for PMSM servo system, in which exist many disturbances, i.e., load torque and rotational inertia variations. The speed control problem is formulated as a Markov decision process problem, which is computed optimal regulation scheme corresponding to each speed and error state using the deep Q-networks. Simulation results are provided to demonstrate that compared with conventional proportion integral control, the proposed DRL control can improve the robustness against load disturbances and high performance of the PMSM speed control system. Keywords Permanent magnet synchronous motor (PMSM) Speed control Markov decision process (MDP) Deep reinforcement learning (DRL) Deep Q-networks (DQN)
1 Introduction In the past few decades, permanent magnet synchronous motor (PMSM) servo system is widely applied in industrial fields thanks to their advantages of compact structure, convenient control and good speed regulation. Those
& Zhe Song [email protected] Jun Yang [email protected] Xuesong Mei [email protected] Tao Tao [email protected] Muxun Xu [email protected] 1
State Key Laboratory for Manufacturing Systems Engineering, Xi’an Jiaotong University, Xi’an 710049, Shaanxi, China
2
Department of Electrical Engineering, Columbia University, New York, NY 10027, USA
3
State Key Lab of Digital Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan 430074, Hubei, China
applications mainly include industrial robotics, aerospace fields, CNCs, medical instruments and auto industries, etc. [1–4], as shown in Fig. 1. Speed control is an important issue in PMSM servo systems. As a classical control method, proportion-integral-derivative (PID) control strategy is the most commonly used method in many industrial areas. However, it can only reach the control requirements in a certain range [5]. For the PMSM system, in the situation of suffer from internal parameters change or subject to external disturbances such as load torque and rotational inertia, etc., it is difficult for PID control to meet highperformance control requirements [6]. To enhance the control performance, many advanced sciences and technologies are all piecemeal researched and applied to control the PMSM servo system, such as robust control [7], adaptive control [8], sliding mode control (SMC) [9], predictive control [10] and intelligent control [11], etc. These studies have achieved some certain results, but these control methods us
Data Loading...
