A neuro-fuzzy controller for grid-connected heavy-duty gas turbine power plants

Abstract

Frequent load fluctuation and set-point variation may affect the stability of grid-connected heavy-duty gas turbine power plants. To overcome such problems, a novel neuro-fuzzy controller is proposed in this paper for single-shaft heavy-duty gas turbines ranging from 18.2 MW to 106.7 MW. A neuro-fuzzy controller was developed using a hybrid learning algorithm and the effectiveness of the controller for all heavy-duty gas turbine plants (5, 6, 7, and 9 series) is demonstrated against load disturbance and set-point variation in a grid-connected environment. Various time domain specifications and performance index criteria of the neuro-fuzzy controller are compared with that of a fuzzy logic controller and an artificial neural network controller. The simulation results indicate that the neuro-fuzzy controller yields optimal transient and steady-state responses and tracks set-point variation faster than a fuzzy logic or artificial neural network controller. Hence, the neuro-fuzzy controller is identified as an optimal controller for heavy-duty gas turbine plants. The neuro-fuzzy controller proposed in this paper is also applicable to the latest derivative Speedtronic controller.