What question did this study set out to answer?

This work aims to improve control strategies for Permanent Magnet Synchronous Motors using reinforcement learning techniques.

May 16, 2026Open Access

Reinforcement Learning-Driven AI Control for PMSM with Field-Oriented Control

Key Points

This work aims to improve control strategies for Permanent Magnet Synchronous Motors using reinforcement learning techniques.
Implemented a Twin Delayed Deep Deterministic Policy Gradient (TD3) agent to control dq-axis current loops.
Validated performance using a 10 s staircase speed profile involving acceleration and braking transitions.
Achieved fast tracking with low overshoot during speed transitions.
Maintained stable dq current regulation throughout the tests.
Demonstrated improved robustness suitable for practical intelligent drive applications.

Abstract

This seminar work presents a reinforcement learn-ing based field-oriented control strategy for Permanent Magnet Synchronous Motor (PMSM) drives. A Twin Delayed Deep Deterministic Policy Gradient (TD3) agent is used to replace the conventional PI current controller in the dq-axis current loop. The controller is validated using a 10 s staircase per-unit speed profile with repeated acceleration and braking transitions. The obtained results show fast tracking, low overshoot, stable dq current regulation, and improved robustness for practical intelligent drive applications.

Read Full Paperexternally

Bookmark

View Full Paper

Cite This Study

Tejaswini Taware (Thu,) studied this question.

synapsesocial.com/papers/6a080ae2a487c87a6a40cef1 https://doi.org/https://doi.org/10.5281/zenodo.20187929

Bookmark

View Full Paper