Adaptive PID control with BP neural network self-tuning in exhaust temperature of micro gas turbine

Mathematical model of exhaust temperature control in micro gas turbine is introduced. To obtain better performance, a self-adaptive PID control is applied to the exhaust temperature control. The parameters of PID control are tuned by back propagation (BP) neural networks. In the tuning process, the plantpsilas predictive output is used to modify the weights of neural networks. The plantpsilas output is also predicted by BP neural networks and it is nonlinear prediction which improves the predictive accuracy. The effectiveness and efficiency of the proposed control strategy is demonstrated by applying it to the exhaust temperature control. The simulations show that the dynamic responses of the exhaust control system can be effectively improved and the anti-disturbance of the proposed controller is better than that of the PID controller.