Dynamic fuzzy neural network based predictive control for alternating current excitation generators

Alternating current excitation generators (ACEG) can adjust the active power and inactive power flexibly and improve the stability of power system. The key to enhance the power system's stability is to choose appropriate ACEG's excitation control method. Con-ventional excitation controllers are unable to perform optimally over the full range of operation conditions and disturbances, due to the highly complex, nonlinear nature of power systems. In this paper, dynamic fuzzy neural network based predictive control is proposed to cope with the problem. Fuzzy neural network is employed to predict power angle and stator voltage of ACEG excitation control system, in order to achieve good dynamics of fuzzy neural network, genetic algorithm is introduced to optimize network parameters. Based on the model output, branch-and-bound optimization method is adopted, which generates proper value of excitation control variable of ACEG. Fuzzy neural network based model predictive algorithm is used in internal model control scheme to compensate for process disturbances, measurement noise and modeling errors. Simulation test under large disturbance at various operating points is made. The results show the controller is effective and feasible.

I. Introduction

Alternating Current excitation generators (ACEG) is one kind of new equipments in power system. It can adjust the active power and inactive power flexibly and improve the stability of power system. The key to enhance the power system's stability is to choose appropriate ACEG's excitation control method. Research has shown that conventional fixed-parameter excitation controllers are unable to perform optimally over the full range of operation conditions and disturbances, due to the highly complex, non-linear nature of power systems. In order to overcome this difficulty, self-tuning systems have been applied and are shown to improve the overall performance of ACEG by allowing the parameters of the controller to adjust as the operating conditions change. However, problems can arise with parameter estimator, which is susceptible to numerical difficulties [1], [2]. In addition, severe disturbances may force the estimator to operate outside the linear range, thus violating the preconditions for parameters estimation [3], [4].