Developing a novel methodology based on the adaptive neuro-fuzzy interference system for the exergoeconomic optimization of energy systems

Optimal control and design of energy systems in some instances require the fast exergoeconomic optimization. Fuzzy inference systems (FIS) were previously employed for either the computerized iterative exergoeconomic optimization or fast optimization of the energy system. The shortcoming of the FIS system was that the requirement to have numerous fuzzy rules and fuzzy membership function that must be collected based on experts' knowledge and usually with try and error steps. On the other hand, conventional optimization methods such as genetic algorithm consume significant calculation time that makes them unsuitable for the fast optimization. In this paper, the adaptive neuro-fuzzy interference system known as the ANFIS was introduced for fast exergoeconomic optimization of energy systems. The ANFIS system automatically developed the required fuzzy items and used them for fast optimization of energy systems. It was employed on two case studies, one for exergoeconomic design and optimization of a benchmark energy system known as the CGAM problem. It was shown that the ANFIS could achieve the optimal solutions of systems with reasonable accuracy, very low computation time, and without dependency on experts' knowledge to develop fuzzy data. The ANFIS was found as an optimistic alternative for the fast optimization of energy systems.