A neural network vector control of induction motor

This paper presents the practical implementation of a computer based slip frequency vector control scheme of an induction motor. A recurrent artificial neural network is used to transform the input parameters, speed error and rate of change of speed error, into an output quantity, the change in inverter output frequency. The recurrent neural network used has an output layer, an input layer and no hidden layer. A novel learning algorithm called the vector space searching algorithm is used to update the network weight. The control algorithm was coded in C++ and implemented on a 486 personal computer