High-Frequency Modeling of Induction Motor Using Multilayer Perceptron

Abstract

This paper proposes an easy-to-use accurate high- frequency (HF) behavioral model of induction motor with a simple and straightforward parameterization process for conducted electromagnetic interference (EMI) analysis. The HF model contains only six circuit elements per phase with a simple structure and is intended for time-domain simulation of variable frequency drive (VFD) systems. The circuit elements are determined by a multilayer perceptron (MLP) neural network, in which only the phase-to-frame impedance is needed for training purposes. Compared with the parameterization methods in conventional motor modeling, the MLP significantly simplifies the parameterization process. Using a three-phase 5.5 kW star-connected induction motor as a case study, the capability and accuracy of the HF model are validated.