625 / 2019-01-20 14:52:40

Dynamic modeling of coal-fired power plant with mechanism and machine learning

dynamic model, mechanism, machine learning, coal-fired power plant

全体主题 > 操作和维护

With the rapid development of renewables, a continuous demand for coal-fired power plant participation in load regulation is expected. There is an urgent need to improve the operation and control in the new era. A simplified dynamic model, which aimed at system analysis and controller design, was developed with fundamental physical laws and advanced machine learning approaches. The dynamics of pulverizing system, boiling system, overheating system, attemperators and the turbine system were studied. The fuel command, the flow of feed water and the turbine governor valve position were considered as the inputs, and the turbine power, the main steam pressure and temperature were considered as the outputs. A massive of operational data from a real power plant was employed to determine the parameters in the mechanism model parts and to obtain the machine learning based submodels. Open-loop simulation and close-loop simulation upon history operational data were conducted to verify the developed model. The results demonstrate that the model is capable to catch the dynamic characteristics of the actual unit and can predict output changes with certain accuracy. The model can provide essential guidelines in controller improvement and design.