Intelligent Model Predictive Control for Boiler Temperature

This paper introduces a new method based on intelligent model predictive control to control the temperature of a boiler. First, several linear local models (as a transfer function) are obtained at different operating points for the boiler system. A fully recurrent radial function neural network is then used to interpolate these linear models. This method is very useful in practice and has a high efficiency, because the boiler system acts as a local linear system at different work points. In simulations, various uncertainties are applied to the system to challenge the proposed control method. The simulation results show that the proposed method has a good performance and especially with increasing uncertainty, the difference between the proposed method and other existing methods becomes more prominent.