Prediction of acoustic noise in switched reluctance machines

Abstract

This paper presents a fast and precise acoustic noise imaging technique for switched reluctance machines (SRM). This method is based on distribution of radial vibration in the stator frame of the SRM. Radial vibration of the stator frame, at a network of probing points, is computed using input phase current and phase voltage waveforms. Sequentially, the acceleration of the probing network will be expanded to predict full acceleration on the stator frame surface, using which acoustic noise emission caused by the stator can be calculated using Boundary Element Method (BEM).