Research On Main Steam Temperature Model Identification And Controller Optimization Of Power Station Based On PSO

As a typical part of the thermal power plant,the main steam temperature system has large delay time and inertia factor,system structure is very complex,the the vast majority of current study on the main steam temperature system,simplyfy MISO system into SISO system for modeling,lack of detail study of MISO system,can not establish higher accuracy main steam temperature system multivariable model to fundamentally reflect the characteristics of the main steam temperature system.With the development of informatization in thermal power plant,real time operational data can be viewed in DCS and SIS,which provides the data foundation for main steam temperature system research.Based on above research background,this paper starts from the actual process of thermal power plants,and based on the real time operation data of the main steam temperature system,identifies the main steam temperature system and optimizes the controller parameters based on the identification result.This paper mainly expands from model identification theory,field data processing,particle swarm optimization,thermal power plant thermal model identification,model verification,controller parameter optimization.Through the research,the following results have been achieved:(1)Analyze the thermal process model structure.According to the dynamic characteristics of the main steam temperature system,design the transfer function model structure of the main steam temperature.(2)Field data processing.Due to the complexity of the field operation of the thermal power plant,the filed data collected by the data acquisition system often contain noise,low frequency components,and unavoidable random errors.Through data filtering,coarse processing,access to meet the identification requirements.(3)Optimize the particle swarm optimization.Aiming at the PSO problem of local optimum and the slow convergence speed,a particle swarm optimization algorithm based on natural selection rule is proposed based on the dynamic inertia coefficient,the recognition accuracy and convergence speed of the algorithm are improved.(4)Build MISO model of the main steam temperature system.The principal component analysis(PCA)was used to analyze the influencing factors of main steam temperature,and five variables that affected the main steam temperature are extracted as input variable of MISO system and use PSO algorithm to identify this system.(5)Based on the result of model identification,the coal-main steam temperature controller are optimized.