Application of neural network model reference adaptive control in coal-fired boiler combustion system

This paper proposes a neural network model reference adaptive PID control method based on RBF neural network identification. This approach can identify the controlled plant on-line with the RBF neural network identifier (NNI), and the weights of the adaptive PID controller (NNC) are adjusted timely based-on the identification of the plant. So the controller is adaptive and the system can be controlled effectively. This approach is also applied to the re-heated temperature plant with long time-delay, large inertia and time-variation in power plant. Research result shows that the controller performs very well when there is disturbance or when plant parameter varies. The robust plant has adaptive abilities that can be easily accomplished on-line.