In modern coal-fired power plants, the main steam temperature control is very strict. But the main steam temperature is an uncertain object with the features of large delays, large inertia, non-linear and so on, which make its control more difficult. Analysis of the BP neural network and genetic algorithm advantages and disadvantages are carried out and, making full use of the robustness and the learning ability of neural networks as well as the global random search capabilities of genetic algorithms, a new control strategy is proposed based on the conventional PID control, and has been carried on the computer simulation experiment. The simulation result indicates that the control policy has the better control quality and broad application prospect based on genetic algorithm - BP neural network PID.
|