Simulation about applying generalized predictive control algorithm based on improved Elman neural network to combustion system of circulating fluidized bed boiler

摘要

For circulating fluidized bed boiler that has many features such as complex nonlinear, multivariable, time-varying, large lag, and model is difficult to establish control and so on, this paper applied Elman neural network to the identification of combustion control system to establish prediction model of multivariable generalized predictive control. Generalized predictive control has advantages of wide range of application and strong robustness, so we can do online scroll optimization and feedback correction to achieve predictive control of combustion control system. Simulation results show that this method is effective.