Dynamic Simulation Of Ultrasupercritical Boiler Milling And Air & Flue System And The Parameter Self-tuning

The power station simulator is mainly used for training of operator,optimization of control system,fault analysis,design analysis of the unit,etc,its existence significantly guarantees the safety and economy of power station operation.In the development process of the simulator,the modeling process using advanced graphical modeling methods is more efficient,and the debugging process with inevitable trial-and-error work is less efficient.The research on dynamic process parameter self-tuning of simulation models is essential to shorten the development time of simulator and improve debugging process efficiency and accuracy.In this paper,IMAGE is used to establish the dynamic simulation model of boiler milling and air & flue system.Combined with field data and genetic algorithm,verification models of system important models are identified.Genetic algorithm is used to realize dynamic parameter self-tuning of corresponding simulation model.The main contents are as follows:Firstly,based on the unit design and structural data,the dynamic simulation model of milling and air & flue system is established.The established model is connected with other subsystem models of the unit and the virtual DCS system,and the whole process simulation of the simulator from cold start to full load to cold stop is completed.The dynamic simulation experiment of the simulator is carried out,and the dynamic response of the output parameter is analyzed,which verifies the rationality of model’s dynamic characteristics.Secondly,baseing on the identification ideology and field data,after simulation object’s closed-loop identifiability verification,using genetic algorithm to realize the dynamic model parameter identification of coal mill and air preheater.The reliability of the identification model is validated by comparing the operation data in different periods.Identification model as a criterion model for dynamic parameter self-tuning of simulation model and verification model for tuning effects.Finally,the step input disturbance is applied to the simulation model,according to the input and output data of the model,the specific parameter values of the transfer function are identified.The difference between the simulation model and the verification model is analyzed by comparing the specific parameter value of transfer function model.Basing on small disturbance analysis and Laplace transform,the transfer function form of the simulation model is derived,the physical meaning of each transfer function parameter and the reason for the existence of simulation error are analyzed.In order to reduce the simulation error,the concept of dynamic characteristic correction coefficient is proposed,basing on genetic algorithm to realize the main models dynamic parameter self-tuning.