Analysis of Dynamic Characteristics of A Moving Secondary-Type Lim by the Neural Network and Fem

Abstract

In the moving secondary-type linear induction motor (LIM), because the secondary cannot get a continuous thrust, it is difficult to analyze some dynamic characteristics. In this paper, we construct a neural network of which training data sets are the thrusts calculated by a finite element method (FEM). Arbitrary positions of the secondary, slips and frequencies form the approximation space. Then, the dynamic characteristics of the carrier are determined by the equation of motion and the trained neural network. The computed velocity is compared with experimental values.