Modeling And Simulation Of Gas-steam Combined Cycle Power Plant At Off-design Conditions

In this thesis, the research is based on the1#gas-steam combined cycle unit in some steel factory in China.1#gas-steam combined cycle unit (Combined Cycle Power Plant) is built by the steel factory and GE. It is the first large capacity gas-fired combined cycle generating unit by burning high heat value COREX gas all over the world. Because of the multi variables, strong coupling, nonlinear, large delay and uncertain characteristics of the CCPP, mechanism modeling, data modeling and hybrid modeling methods have been used to achieve the modeling and simulation of the CCPP.This thesis first briefly describes the domestic and international CCPP status and the development trend, and also pointed out the significance and status of the study for the CCPP at off-design conditions characteristics. The research in this thesis is based on the above background. After a detailed analysis and research of the CCPP working principle, the gas turbines, heat recovery steam generator and steam turbine working characteristics, this thesis analyzes the major factors that impact on CCPP performance, including the load, the atmospheric conditions, etc. Then, based on each factor’s specific impact on CCPP and a rational assumption, the mechanism model and the simulation are carried out.After a detailed analysis of partial least squares (PLS) method and the kernel function theory, kernel-based partial least squares method (KPLS) is introduced and applied to the modeling process of CCPP. As is known from the data-drived model simulation, the satisfactory accuracy is obtained. Finally, to solve the low accuracy problem of the mechanism model, an error compensator is established in order to modify the mechanism model, simulation results show that compared with the mechanism model, the accuracy of the modified hybrid model has been greatly improved. Hence, the effectiveness of the proposed hybrid model is verified.