Degradation estimation using feature increment stepwise linear regression for PWM Inverter of Electro-Mechanical Actuator

Abstract

Electro-Mechanical Actuator (EMA) is a critical subsystem of the aircraft. The degradation of Pulse Width Modulation (PWM) Inverter has significant influence on EMA performance, which may lead to aircraft catastrophic consequence. However, strict limitation on sensor placement in aeronautical applications brings much difficulty to characterize the degradation state of EMA PWM Inverter. To effectively estimate the degradation of EMA PWM Inverter with external monitoring features, this paper proposes a feature increment stepwise linear regression (FI-SLR) model. In the FI-SLR model, the increments of features are extracted to estimate degradation state of EMA PWM Inverter. To evaluate the effectiveness of FI-SLR model, EMA simulation is performed to acquire more condition data. Moreover, both practical dataset from NASA and simulation data are utilized in experiments. Finally, experimental results show that FI-SLR has better estimation performance in the field of EMA PWM Inverter degradation estimation.