Multi-Model Switching Predictive Functional Control Of Boiler Main Steam Temperature

Main steam temperature of boiler is typically great inertial and long-time delay system. Traditional PID control process is that the deviation between the reference and the output of the process is returned to controller when the deviation is not zero. Because of the long pipeline, the effect that acted on by the control needs long time, so traditional PID control will cause the oscillation even the stability of control system.In order to eliminate the great inertial and long-time delay of main steam temperature, predictive functional control strategy is proposed in this paper, the variation trend of main steam temperature is sensed by predictive model in advance, the effect of the long-time delay of control is overcomed by predictive control algorithm.When boiler load varies, the characteristic of the main steam temperature is also varied, good control effect won't be obtained by a single predictive model, multi-model control strategy is introduced in this paper, when the boiler load is changed, the model is changed to the one that most similar to the process so that the time varying of the system could be overcomed.Simulation tools of MATLAB. Simulink is used in this paper, the optimal PID and predictive functional control parameters in each of the five working conditions, then using the controller we can get and compare the effect of the two different control effect, and prove the rationality the choice in this paper. The combination of the predictive functional control and multi-model switching, we could change the fitting model when the boiler load varied.The simulation result showed that multi-model switching predictive functional control could not only overcome the great inertial and long-time delay but also the adaptive the variation the boiler load.