Electrical Machine Design by optimization for E-motor Application: a Drive Cycle Approach

Abstract

This paper presents a methodology for the design by optimization of an electric machine considering several operating cycles. This is achieved by simplifying the operating cycles using a k-means clustering algorithm. K-means clustering is used to simplify three operating cycles: WLTP, highway and urban. The three simplified cycles are then considered in the design optimization of a permanent magnet synchronous machine with the objective of minimizing the losses of both the electrical machine and the voltage inverter. The machine is represented by a magnetic equivalent circuit model and the voltage inverter losses by an analytical model. Then, the three optima (obtained for each cycle) are characterized and compared in terms of performance over these drive cycles.