Accuracy Improvement of High-Frequency Transmission Line Model of Induction Motor Through Multilayer Perceptron

The high-frequency transmission line (HFTL) model of the induction motor has been used to analyze and estimate some high-frequency behaviors in the motor drive system. The per-unit-length (PUL) equivalent circuit model serves as a promising HFTL model because of its simple structure and ease of implementation. However, this model does not consider the influence of position, assuming that the model elements at different positions remain unchanged. Due to layout differences and manufacturing tolerances, this assumption may affect the model accuracy. In order to resolve this issue, this paper proposes an improved form of the PUL equivalent circuit model that uses a non-uniform transmission line with the position effect represented by the k-coefficients. A multilayer perceptron (MLP) neural network is used to determine the values of the k-coefficients. Using a 5.5 kW induction motor as a case study, the experimental results verify the ability of the proposed method to improve the accuracy of the PUL equivalent circuit model.