The Supercritical Main Steam has a large inertia, delay and nonlinear and dynamic characteristics change with the operating conditions, it is difficult to establish the precise mathematical model, this algorithm based on RBF neural network GGAP posed a direct neural network predictive controller, the combination of online learning and control to a supercritical power plant main stream temperature as the research object, MATLAB simulation results show that the superheated steam temperature system can achieve effective control, performance than the conventional PID control has greatly improved.
The traditional PID control with RBF function the nerve network integration and made based on RBF the nerve
network the PID controller. the controller for the supercritical main system, the network for temperature control system
RBF an on-line to identify and build up the reference in the line and PID controller model for providing information and
controllers of the online study to their control, adjust the parameter from online, the performance indicators. MATLAB
simulation results show that the controller for temperature system has good control, not only keep track of good
performance and robustness better.