Use of fuzzy systems and bat algorithm for exergy modeling in a gas turbine generator

Exergy analysis plays a major role in thermal systems. Using exergy, apart from finding components for further improvement, fault detection and diagnosis can be conducted. This paper demonstrates the use of fuzzy systems for capturing exergy destruction variations in the main components of an industrial gas turbine. The models cover part load operating conditions. The fuzzy models are trained applying locally linear model tree algorithm followed by a meta-heuristic nature inspired algorithm called bat algorithm. The data for model training and validation are generated using semi-empirical models developed by the authors. However, the inputs to the model are kept the same as the inputs as experienced by a real gas turbine generator. The comparison between actual data from a different day and the prediction by the proposed method showed a match that is close enough to be considered as reliable. The models could be used for performance optimization and condition monitoring.