An Integrated Fault Isolation and Prognosis Method for Electric Drive Systems of Battery Electric Vehicles

Abstract

The electric drive system is a key subsystem of battery electric vehicles (BEVs). Abnormalities in the electric drive system components may lead to performance degradation in the drive system and, more severely, loss of power in the vehicle. This article presents an integrated prognosis system for early detection and isolation of the electric drive system and component faults. The system first calculates multiple health features, known as health indicators derived from the available onboard sensor signals. Then, an integrated prognostic and fault isolation strategy is used to isolate the root cause of electric drive faults by monitoring the performance of the multiple health indicators. The prognostic system uses a hierarchical approach: it first determines if there is a degradation in the electric drive system by comparing the system achieved torque with the estimated torque and, then, goes further to check multiple component-level health indicators for the various components comprising an electric drive system, including motor stator winding, three-phase current sensors, and resolver. After a component has been detected to degrade to a certain level, the prognosis system sends out an alert before the severe performance reduction of the drive system occurs, thus protecting the customers from loss of propulsion and walk home situations.