Development of Low-Loss, Sustainable Reluctance Motors: Implementation and Control of Ripple Optimization Methods

This work focuses on developing efficient control to optimize electric motors, increase battery autonomy, and promote sustainability when applied to vehicle electrification. It builds upon previous efforts by introducing a simulation framework designed to investigate the torques and currents of a three-phase reluctance motor within a Python environment. The overarching objective is to extract and explore various parameters associated with synchronous reluctance machines (SyRM) to optimize the performance of these motors. Synchronous Reluctance Motors (SyRM) represent a class of electric motors distinguished by their notable advantages, including low production costs and the absence of magnets or windings in the rotor. However, these motors have challenges associated with high torque ripple and the need for efficient control mechanisms at elevated speeds, consequently restricting their applicability across diverse scenarios. The results were categorized based on the torque ripple obtained after simulations were conducted at different current values. Subsequently, this data was prepared for neural network training, with MATLAB being employed to analyse training and testing indicators. The training phase revealed high levels of accuracy; training accuracy tended to decrease as the current increased, but this was mitigated by the increase in the random variation of Fourier coefficients. The study highlights the potential of integrating simulation techniques with artificial intelligence to significantly contribute to developing and optimizing synchronous reluctance motors, thereby enhancing their performance and efficiency. Furthermore, the results point toward a promising trajectory for future research and developments in this field.