Exergy analysis plays a major role in thermal systems. Using exergy, apart from finding components for a potential for further improvement, fault detection and diagnosis, performance optimization, and environmental impact assessment can be conducted. This chapter addresses the use of fuzzy systems for modeling exergy destructions in the main components of an industrial gas turbine. The details include: (i) system description and the challenges in developing first principle models, (ii) thermodynamic models for part load and full load operating conditions, (iii) model identification technique that uses fuzzy systems and a meta-heuristic nature inspired algorithm called Bat Algorithm, (iv) validation graphs for semi-empirical models, and (v) validation test for fuzzy models. In the validation of the fuzzy model, the inputs to the model are considered the same as the inputs as experienced by the gas turbine generator. The comparison tests between actual data and prediction demonstrate how promising the combined method is as compared to separate use of the fuzzy systems trained by a heuristic approach.