Load Frequency Control using ANN-PID Controller

摘要

A control scheme of ANN based proportional-integral-derivative (PID) controller is developed here to maintain the system frequency at nominal value. Due to some complication of modern industrial system, the conventional PID controller is not capable to meet our requirement. Neural networks have great capability in solving complex, nonlinear mathematical problems. This paper introduces the design of neuro-PID controller model to improve the response and performance of conventional PID controller. The weights of neural network connecting links are modified by faster descent mean using the back propagation algorithm. The performance of ANN-PID controller is compared with different existing techniques. The simulation result presented in the paper reveal that the proposed ANN-PID technique rapidly track the reference frequency of system and better result in respect of rise time, settling time, peak time, overshoot etc.