Model-Based Fault Diagnosis in Electric Drives Using Artificial Neural Networks
ARMY TANK AUTOMOTIVE RESEARCH DEVELOPMENT AND ENGINEERING CENTER WARREN MI
Pagination or Media Count:
A model based fault diagnostics study of the power electronics inverter based electrical drive is proposed. The power electronics inverter is considered to be the weakest link in such a system, hence the focus of the work is initially on fault conditions of the inverter. A faulted model for the inverter and the motor is used to generate various fault condition data, which are then compared against data generated by a normally functioning model. An artificial neural network is used to detect these faults based on features extracted from signals. The technique is viable for quick fault detection, and also the time of a fault. The concepts introduced in the paper can be effectively applied for real-time fault diagnostics in electric and hybrid vehicles, and other applications where electrical drives are used.
- Surface Transportation and Equipment
- Electric and Ion Propulsion