Modeling of Boiler-Turbine Nonlinear Coordinated Control System Based on RBF Neural Network

Due to the characteristics of multi-variable, nonlinear, time varying, coupling and large time-delay of the boiler-turbine coordinated control system in power station, it´s impossible to establish accurate mathematic model by traditional methods. RBF neural network can be used to identify nonlinear model effectively. A 160 MW nonlinear coordinated control model described by a group of differential equations is chosen in this paper. The drum pressure, power and water level of the coordinated control model are identified by RBF. Simulation results show that this strategy is effective. At the same time, the coordinated control strategy based on RBF is also discussed in this paper.