Mitigation of acoustic noise and vibration in switched reluctance motor drive using neural network based current profiling

In order to optimize the performance of switched reluctance motor (SRM) drives, geometrical design of the motor as well as control strategies for its power converter should be adjusted. High levels of torque ripple and acoustic noise are the main drawbacks of SRM drives in many applications. This paper mainly focuses on control strategies to be used for minimizing the acoustic noise and vibration in SRM drives. Further, the major sources of acoustic noise and vibration have been addressed. In addition, the significance of various control variables (control angles and current profile) has been studied in detail. Furthermore, based on this analysis, a current shape optimization method has been introduced. Extensive simulations are given to support the proposed control strategies. Moreover, the impact of the noise reduction algorithm on the efficiency of the drive has been explored. In addition, artificial neural networks have been employed to generate optimized current profiles at various operating points. Finally, experimental results for an 8/6, 0.3 kW SRM drive are given.