Modeling of a combined cycle power plant

Abstract

The combined cycle power plant is a non-linear, closed loop system, which consists of high-pressure (HP) superheater, HP evaporator, HP economizer, lowpressure (LP) evaporator, HP drum, HP deaerator, condenser, HP and LP steam turbines and gas turbine. The two types of turbines in the plant i.e. the gas turbine and the HP and LP steam turbines operate concurrently to generate power to the plant. The exhaust gas which originate from the combustion chamber drives the gas turbine, after which it flows into the heat recovery steam generator (HRSG) to generate superheated steam to be used in driving the HP and LP steam turbines. In this thesis, the combined cycle power plant is modeled at component level using the physical method. Assuming that there is delay in transport, except for the gas turbine system, the mass and heat balances are applied on the components of the plant to derive the governing equations of the components. These time dependent equations, which are of first order differential types, are then solved for the mass and enthalpy of the component. The solutions were simulated using Matlab Simulink using measured plant data. Where necessary there is no plant data available, approximated data were used. The generalized regression neural networks are also used to generate extra sets of simulation data for the HRSG system. Comparisons of the simulation results with its corresponding plant data showed good agreements between the two and indicated that the models developed for the components could be used to represent the combined cycle power plant under study.