An adaptive turbogenerator excitation controller based on artificial neural network

In this paper, an adaptive turbogenerator excitation controller based on an artificial neural network (ANN) is developed. The proposed controller is a four layered feedforward neural network. The third layer produces the control signal and the fourth layer gives the output signal. The signs of plant error are introduced. The control strategy is reached continuously by adjusting weights and thresholds of the network. This controller is applied to a sample excitation system and is compared with an optimized PID controller. The computer simulation shows that the result is effective.<>