Speed Control of DC Motor Using Novel Neural Network Configuration

Abstract

This paper uses Artificial Neural Networks (ANNs) in estimating speed and controlling it for a separately excited DC motor. The rotor speed of the dc motor can be made to follow an arbitrarily selected trajectory. The purpose is to achieve accurate trajectory control of the speed, especially when the motor and load parameters are unknown.

Such a neural control scheme consists of two parts. One is the neural identifier which is used to estimate the motor speed. The other is the neural controller which is used to generate a control signal for a converter. These two neural networks are trained by Levenberg-Marquardt back-propagation algorithm. ANNs used in this are the standard three layers feedforward neural network with sigmoid activation functions in the input and hidden layers while linear activation function is employed for the output layer.

The conventional constant gain feedback controller fails to maintain the performance of the system at acceptable levels under unknown dynamics in load torque. On the other hand, ANNs act as an effective tool to implement the model and adaptive control in a complicated non-linear system having expansive allocations.

The adaptive learning algorithm is formed in such a way that the learning rate is as large as possible while maintaining the stability of the learning process. This simplifies the learning process in terms of computation time, which is of special importance in real-time implementation.