Excitation control of a power‐generating system based on fuzzy logic and neural networks

This paper presents a practical design of an intelligent controller (using fuzzy logic and neural network concepts) for the excitation control of an isolated power-generating system. The controller is suitable for realtime operation, with the aim of improving the dynamic characteristics of the generating unit by acting properly on the exciter input. At first, digital simulations of high- and reduced-order models of the above system are performed using conventional control techniques on five system cases which are based on previous work. Then, a fuzzy logic proportional-plus-derivative controller (FPDC) is designed and the dynamic performances of the mentioned five system cases and the FPDCs are presented by comparison. Finally, an enhanced controller is designed which involves a pre-trained neural network as a model dynamics capture. The computer simulation results obtained clearly demonstrate that the performance of the developed fuzzy logic and neural network controller offers better damping effects on the generator oscillations over a wider range of operating conditions, than the associated ones of the conventional excitation controller designs with output feedback applied to high- and reduced-order system models.