Research On The Model Indetification Method In Mu Ltivariate Thermal Systems

Paper is composed of two parts,The first part is derived coordinated control system model of the mechanism, and the model parameters is identified through the mechanism of genetic algorithm. The model of coordinated control system can capture the essential dynamics of a unit so it is typical for the process.Based on this model and applyingthe model,an parameter identification algorithm based on the improved genetic algorithm is proposed.The second part utilized a set of actual close-loop running data from a 600MW unit. A lot of work was done on dealing the data. After that the unitcoordinated control system was simplified into a set of two-input andtwo-output multi variable object models for each equilibrium point. The model was then decomposed into submodels. through the same method,subsubmodels were gained. by Extend_LS_Identify and Auxiliary variable recursive algorithm,the models structure and parameters were obtained,and a set of mathematical models linear is gained.