Optimal Control for Boiler Combustion System Based on Iterative Heuristic Dynamic Programming

Boiler combustion system is a complex nonlinear system which has characteristic of strong coupling and strong-disturbance. It is hard to build accurate mathematical model and achieve optimal control for it. In this paper, radial basis function (RBF) neural network model for boiler combustion system is built based on data driven method firstly, then performing the optimal control of the boiler combustion system via the iterative heuristic dynamic programming (HDP) algorithm, and improving the initial weights of neural network and the utility function. Finally compared with the traditional HDP algorithm in Matlab. The result shows that the optimization algorithm of the iteration HDP based on the RBF neural network gets better in overshoot, convergence speed, steady state error, adaptability and robustness.