Modeling Optimization Of Thermal Power Plant Simulation System Based On SIS Data

The research object of this paper is 600 MW subcritical unit in Faer power plant of China DaTang Corportion in Guizhou province. Using the method of mechanism to analyze the mathematical models of the main equipments, such as superheater and reheater, turbing iroper, high-low heater and so on, this paper find the key factor which influences the dynamic and static characteristics of the model. It establishes each subsystem’s models which concluded out of primary and secondary systems in the PanySimu simulation support platform:the main reheat steam system, Total System Intervention(TSI), the regenerative extraction system and so on. It studies the impact of the changing of the key factor influences the output characteristic in the model, so we can determine the scope of the key coefficient of the optimization. By using empirical mode decomposition algorithm to deal with the on-site data which exists burr and high frequency interference, and by using particle swarm optimization algorithm to optimize the key factor of the model, we make the when the optimized model is inputting the input value which is the SIS data, the output can follow the output which is data in SIS as much as possible.The optimization procedure above is realized through Window Procedure which is developed by C++Builde. The procedure as the plugin of PanySimu simulation support platform, by leading a few SIS data to intelligent optimize the designated factor automatically, then to make the relevant parameters of the typical load in accordance with the field data reaches a steady state simulation system accuracy requirements and dynamic response curve is consistent with the trend in the field. This procedure concludes superior human-computer interface, and have certain universality. Its optimization process can through changing the arithmetical parameters to optimize for different models. This paper have taken the typical equipment of boiler superheater final model as an example and used this method to optimize its coefficient. As the same time, added variable coefficients design to make sure it will have small errors in the continuous operating condition. At the end the experimental results verify the correctness and validity of this method.