UDC 681.518 UDC 621.313.333

STUDY OF DIRECT NEURAL CONTROL OF AN INDUCTION MOTOR TRAINED IN A DIGITAL TWIN Study of Direct Neural Control of an Induction Motor Trained in a Digital Twin

Published in Proceedings of Petersburg Transport University · Volume 23, Issue 3, 2026 · Pages 596–603 · Rubric: PROBLEMATIC OF TRANSPORT SYSTEM
DOI: https://doi.org/10.20295/1815-588X-2026-3-596-603
Received: 13.05.2026 Accepted: 07.07.2026 Published: 07.10.2026 Language of publication: RUS
Objective: to assess the feasibility of direct neural control of an induction motor drive without a fully specified mathematical model of the motor and to determine whether this approach can reduce active power consumption while maintaining speed-control quality comparable to classical field-oriented control. Methods: the study was carried out on a digital-twin model of the inverter-induction motor-controller system. A reinforcement learning agent was trained to generate control actions in the d-q reference frame, and its performance was compared with FOC using the stator RMS current, active power, and speed-control error as evaluation criteria. Practical significance: the trained policy can potentially be transferred to an embedded drive-control platform. This reduces the dependence of control quality on accurate parameter identification and makes the approach relevant for energy-efficient electric drives.
electric drive, reinforcement learning, digital twin, neural network control, field-oriented control
Text References
Text (PDF)
Read Download

1. Kazemikia D. Reinforcement Learning for Motor Control: a Comprehensive Review // arXiv preprint. 2024. URL: https://arxiv.org/abs/2412.17936 (data

2. Field-Oriented of Reinforcement Learning // MathWorks Motor Control Blockset Example. URL: https://www.mathworks. com/help/mcb/gs/focofmsmsing-reinforcement-learning.html (data obrascheniya: 15.12.2025).

3. AI-Based Field-Oriented Control for Induction Motors / E. Benmalek [et al.] // IAPGOS. 2024. Vol. 14, № 4. Pp. 75–81. URL: https://doi.org/10.35784/ iapgos.6253 (data obrascheniya: 15.12.2025).

4. Enhanced Induction Motor Control Using Deep Reinforcement Learning and PWM with DTC / P. Saurabh [et al.] // Electric Power Components and Systems. 2023. Vol. 51, no. 12. Pp. 1083–1095. URL: https://doi.org/10.1080/15325008.2023.2274357 (data

5. Lenz I., Knepper R., Saxena A. DeepMPC: Control // Proc. of Robotics: Science and Systems. 2015. URL: https://doi.org/10.15607/RSS.2015.XI.012

6. Sensorless Speed Control of Induction Motor Drives Using Reinforcement Learning and Self-Tuning Simplified Fuzzy Logic Controller / Q. Abdullah [et al.] // IEEE Access. 2024. Vol. 12. Pp. 136485–136501. URL: https://doi.org/10.1109/ACCESS.2024.3435529 (data

7. Eschmann J., Albani D., Loianno G. RLtools: A Fast, Portable Deep Reinforcement Learning Library for Continuous Control // Journal of Machine Learning Research. 2024. Vol. 25, no. 301. Pp. 1–19. URL: https://www.jmlr.org/papers/v25/24-0248.html (data

8. TinyRL: Towards Reinforcement Learning on Tiny Embedded Devices / T. Szydlo [et al.] // Proc. of the 31st ACM Int. Conf. on Information and Knowledge Management (CIKM). 2022. Pp. 4985–4988. URL: https://doi.org/10.1145/3511808.3557206 (data obrascheniya: 15.12.2025).

9. Enhancing Field-Oriented Control of Electric Drives with Tiny Neural Network Optimized for Microcontrollers / M. J. Mouk Elele [et al.] // arXiv preprint. 2025. URL: https://arxiv.org/abs/2502.00532

10. A Survey of Reinforcement Learning-Driven Knowledge Distillation: Techniques, Challenges, and Applications / Z. Xu [et al.] // Preprints. 2025. URL: https://doi.org/10.20944/preprints202503.0903.v2