Neural network self-tuning PID control for boiler-turbine unit

In thermal power plant, conventional PID controller and direct energy balance (DEB) control strategy of coordinated control system (CCS), which are tuned at typical operating point, can hardly work well at different unit load. A novel self-tuning PID control strategy based on a two-level neural networks (NN) is proposed for CCS. The two level NNs are called static NN (SNN) and dynamic NN (DNN) respectively. SNN is used for PID controller arguments' primary tuning according to the system operating point such as unit load, in order to follow the wide range load changing; The two DNNs are used for PID fine tuning according to the error and error rate of the CCS, in order to overcome the small range load changing, system parameters' slow variance and some disturbance. Simulation results show that good dynamic regulating performance can be obtained by using the presented new method, and stronger robustness is obtained.