The Application Of Intelligence Control Algorithm In Main Stream Temperature Control System

In thermal power plant,the dynamic behavior of the main steam temperature shows a characteristic of great delay,big inertia,time variance and non-linearity, which makes the conventional PID control cannot achieve satisfying control performance when the conditions change, therefore, the intelligence control method based on neural network was proposed. Firstly, the single neuron PSD adaptive algorithm based on Smith predictor was proposed when the precise model cannot be built. Secondly, the weights of Elman network through SA for identifying the non-linear thermal process. Thirdly, By combing the state feedback based on neural network state observer and the conventional PID, the unmeasurable states of the system was reconstructed through NN observer, the adaptivability of system was improved by state feedback.