Model Identification And Simulation Analysis Of 300MW Power Unit

With continuous production of a variety of thermal power units with high-capacity, high-parameter, and automation become higher and higher, control quality also need higher requirements to process control system. Grasping the characteristics of controlled object is the basis to determine the optimal control parameters, well-designed system. In this paper, operating data of thermal power plant is taken as reference and a typical thermal power plant thermal process are taken as the research object, from a theoretical perspective, the model identification problem of the thermal power plants is studied. Classical identification methods, least-squares method, genetic algorithm, neural network method were adopted respectively to identify or thermal process and the advantages and disadvantages of each identification method were comparatively analyzed, and the adjusted controller parameters were simulated and analyzed.