Projektdetaljer
Beskrivelse
Abstract:
Over the past two decades, artificial intelligence (AI) has significantly advanced, enabling modern power electronic systems to think, learn, and reason. Despite the prevalent use of AI, particularly machine learning, in controlling power electronic systems, there is limited research on applying these techniques to improve sensorless control for Permanent Magnet Synchronous Motor (PMSM) drives. Addressing this gap runs into an inevitable problem, that is the inherent position and speed estimation error. The inaccurate feedback information makes it quite a challenge to decide the proper current command and to optimize the dynamic performance of sensorless drive. The system's nonlinear nature, compounded by multiple interacting control loops, further complicates this issue.
Building on recent innovative approaches, this project aims to explore a hybrid approach to integrate machine learning algorithm into PMSM drive system with respect to obtaining higher dynamic performance. Experimental validation of system performance and limitations will be made to give comments on the proposed method.
Funding:
Self-funded
Over the past two decades, artificial intelligence (AI) has significantly advanced, enabling modern power electronic systems to think, learn, and reason. Despite the prevalent use of AI, particularly machine learning, in controlling power electronic systems, there is limited research on applying these techniques to improve sensorless control for Permanent Magnet Synchronous Motor (PMSM) drives. Addressing this gap runs into an inevitable problem, that is the inherent position and speed estimation error. The inaccurate feedback information makes it quite a challenge to decide the proper current command and to optimize the dynamic performance of sensorless drive. The system's nonlinear nature, compounded by multiple interacting control loops, further complicates this issue.
Building on recent innovative approaches, this project aims to explore a hybrid approach to integrate machine learning algorithm into PMSM drive system with respect to obtaining higher dynamic performance. Experimental validation of system performance and limitations will be made to give comments on the proposed method.
Funding:
Self-funded
Status | Ikke startet |
---|---|
Effektiv start/slut dato | 15/03/2025 → 14/05/2027 |
Fingerprint
Udforsk forskningsemnerne, som dette projekt berører. Disse etiketter er oprettet på grundlag af de underliggende bevillinger/legater. Sammen danner de et unikt fingerprint.