Current harmonics suppression strategy for motor drives using neural network algorithm

• Main Author: Lei, Xiaoyi
• Other Authors: Christopher H. T. Lee
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/175934
• Institution: Nanyang Technological University

Permanent Magnet Synchronous Motors (PMSMs) are widely used in various applications due to their simple structure, excellent torque characteristics, and precise control capabilities. However, dead time effect and inverter’s non-ideal behaviors would lead to an increase in voltage and current harmonic components, causing extra losses, torque ripples and vibration. This thesis focuses on investigating the nonlinear voltage distortion as well as developing a novel offline measurement compensation method to eliminate the voltage distortion. A novel numerical fitting method based on radical basis function neural network is introduced to build voltage error model. The proposed offline compensation method is evaluated through experiments on the test bench based on the dSPACE MicroLabBox and PMSM, demonstrating its effectiveness in mitigating voltage distortion and reducing current harmonics across various operating conditions, including steady-state, transient, load variations, and multiple speed increases. The experimental results illustrates that the proposed method achieves superior performance in harmonic suppression. The proposed method offers a promising solution for enhancing the performance of PMSMs in diverse application domains without requiring additional hardware or complex algorithms.