The heavy duty gas turbines evolution led to higher combined cycle efficiencies. Thus, more complex heat recovery steam generators were developed in order to maximize the use of that energy potential. Therefore, computational models capable to predict the operational conditions of the equipment may be needed in order to analyze the system behavior for different situations. This article describes a computational model able to simulate the off-design behavior of a heat recovery steam generator (HRSG) operating in a combined cycle power plant. The model was developed so that it can be used in both model-based diagnostics systems and performance evaluation systems. Each heat exchanger inside the HRSG was designed individually and arranged according to the analyzed equipment. The computer code’s architecture was built in such a way that it can be easily changed, allowing the analysis of other HRSG’s configurations with simple structural changes, given the program’s modularity. In order to deal with the lack of details of the power plant equipment, which means not enough geometrical information of each heat exchanger, a generic algorithm tool was used to calibrate the heat exchangers models using only the measured data of the power plant SCADA. The developed program was validated against operational data from a real plant and showed satisfactory results, confirming the robustness of this model.

In recent years, combined cycle power plants showed remarkable progress in the safe operation and reliability of their equipment, mostly because of the reliable control and instrumentation systems available today. However, these systems cannot detect and evaluate inconsistencies in the behaviour of equipment due to failures and avoid trips caused by catastrophic events. Computer models developed to simulate the power plant equipment are often employed in diagnosis tools in order to provide accurate healthy parameters that are compared to the field measured parameters. In this work, the computer models built for the simulation of some of the main bottoming cycle equipment of a real power plant (steam turbine, HRSG, boiler feed water pumps and condenser) are described. These models were developed through characteristics maps and constitutive equations related to the fluid path analysis, implemented in Fortran language. The results provided by the developed models for each equipment show good agreement with operational data at base and partial load in simulations that covered a good part of the load domain. Due to the good agreement between the measured parameters values and those calculated through simulation, these models are intended to be included in an on-line fuzzy-based diagnosis system.