Controlling disturbances of islanding in a gas power plant via fuzzy-based neural network approach with a focus on load-shedding system

Abstract

In a number of power plants such as the South Pars Gas Company, efficient control performance regarding the turbine generator is always needed to avoid facing disturbances of islanding in the supply of power, at each instance of time. It is particularly important to deal with through state-of-the-art techniques, due to the fact that the separation of the electrical grid may increase or reduce the speed of the above-referenced turbine generator and correspondingly the frequency, in its permitted range, may be deviated to stop power generation, abruptly. With this key goal, a set of control strategies through the representation of the gas turbine via Rowen is proposed in the present research. In a word, there are some classical and intelligence-based control approaches to cope with the disturbance of islanding with a focus on load-shedding system. In reality, the proposed investigation is mainly concentrating on the fuzzy-based neural network control approach to cope with the turbulence of frequency deviation. The main contribution made in this research is to deal with gas turbine model in association with generator in conjunction with load-shedding system, while intelligence-based techniques considering fuzzy, neural network and genetic algorithm are synchronously established to design the control strategy with high accuracy. In the sequel, the effectiveness of the proposed intelligence-based control approach is verified, while the classical proportional-integral-derivative control approach via the Ziegler–Nichols and also the simulated annealing tuning processes are taken as the benchmarks.