DESIGNING POWER SYSTEM STABILIZER WITH ARTIFICIAL NEURAL NETWORKS

A power system stabilizer using artificial neural network has been designed and tested via simulation. The use of the neural network has made it possible to design a controller with a simpler structure, more rapid computation and more satisfactory performance. The structure of the ANN PSS is similar to state feedback, with the exception that its output is bounded and so the control signal does not exceed the field excitation. The design has been carried out by the minimization of an objective function via the error back propagation algorithm, therefore the proposed controller has optimal performance. The simulation results demonstrate the superiority of the proposed ANN PSS over CPSS in a wide range of operating conditions.