Self-adaptive neuron PID control in exhaust temperature of micro gas turbine

Abstract

Mathematical model scheme of exhaust temperature control in micro gas turbine is given out. To obtain better performance, a self-adaptive neuron PID control is applied to the exhaust temperature control in this paper. The neuron model and learning strategy are given. The effectiveness and efficiency of the proposed control strategy is demonstrated by applying it to the exhaust temperature control. The different learning velocity and neuron proportion of self-adaptive neuron PID control are simulated to analyze the control performance. It is found that the neuron proportion in self-adaptive neuron PID control is the most sensitive parameter, the learning velocities of proportion and integrator affect the rapidity of response, overshoot and static error, while the learning velocity of differentiator affects relatively little to the control performance. The simulations show that the dynamic responses of the exhaust control system can be effectively improved and the robustness of the proposed controller is better than that of the PID controller.