Study on New Control Method for Boiler Combustion System

Chain boiler combusting system is often seen and difficultly controlled in industry due to large delay time, varying coal´s quality and steam load. In this paper a neural network identification method for time delay system and the parameter of delay time is discussed. We can acquire the best control value in term of the errors between predictive output and reference input, adjusting the parameters of PID controller based on neural network. Combining the discrimination of long time delay system based on neural network with the control method based on model prediction, it can be applied to the control of the long delay time system with variable parameters or unknown delay time. In this paper, simulating with a 10t/h chain boiler model, compared with normal PI controller, pole assignment PI with Smith predictor (STPI) and model reference adapted controller (MRAC), the results show it has much better advantage of celerity and robustness