A hybrid model for optimal concurrent design of solid oxide fuel cell system considering functional performance and production cost

This research addresses the issues to identify the optimal design of solid oxide fuel cell (SOFC) system considering functional performance and production cost. In this research, modeling of the relations between design parameters and evaluation parameters is first discussed. Due to uncertainties of parameter relations, a hybrid model is introduced in this work to describe two types of parameter relations, mathematical relations and neural network relations, and associate these two types of relations through a parameter relation network. The optimal SOFC system design considering function performance and production cost is achieved by changing values of design parameters based on evaluation of performance and cost parameters through multi-objective optimization. A software system has been developed based on the introduced method. A case study has also been conducted to demonstrate the effectiveness of the optimal SOFC system design approach.