Analysis and Validation of Torque Ripple Cancellation to Reduce Electric Motor Noise for Electric Vehicles

<div class="section abstract"><div class="htmlview paragraph">The electric motor is a significant source of noise in electric vehicles (EVs). Traditional hardware-based NVH optimization techniques can prove insufficient, often resulting in trade-offs between motor torque or efficiency performance. The implementation of motor control-based torque ripple cancellation (TRC) technology provides an effective and flexible solution to reduce the targeted orders. This paper presents an explanation of the mathematical theory underlying the TRC method, with a particular focus on the various current injection methods, including those that allow up to 4DOFs (degrees-of-freedom). In the case study, the injection of controlled fifth or seventh order current harmonics into a three-phase AC motor is shown to be an effective method for cancelling the most dominant sixth order torque ripple. A dedicated feedforward harmonic current generation module is developed the allows the application of harmonic current commands to a motor control system with adjustable magnitude and phase. Finite-element models are developed for both the electric motor and the motor fixture system to characterize the noise transfer path. The electromagnetic forces are calculated with and without TRC injections and subsequently mapped onto the stator teeth to predict the noise and vibration response at the motor fixture system. The predicted TRC inputs for the Id and Iq injection currents are then compared with the optimal parameters obtained from motor dynamometer tests. A good correlation is achieved between the predicted and measured optimal inputs, resulting in a reduction of over 10 dB at the most dominant sixth electric order, or the 24<sup>th</sup> mechanical order. The validated TRC injection analysis method and control algorithm have been successfully implemented in General Motors' Ultium electric propulsion systems.</div></div>