Induction motor design using neural network
K. Idir, Liuchen Chang, H. Dai · 2002
This paper presents the application of a neural network in optimizing design parameters of an induction motor. This approach is based on training the neural network with data generated from an optimization technique. A backpropagation with adaptive learning rate algorithm is utilized in training the network. Once trained, the neural network will be capable of producing a set of optimum motor design parameters for a given motor specification in a very short time and with little effort. The results shown in this study indicate that a well trained neural network can fulfil the task of a motor design successfully and therefore presents a good alternative approach in machine design that may have features of both speed and accuracy.