Multi-objective Teaching-Learning-Based Optimization of Combined Commercial Fuel Cells for Electricity Production

Abstract In this research, an organic Rankine cycle (ORC) is run by the recovered heat from different commercial fuel cells and optimized to find the best configuration from the thermos-economic point of view for different refrigerants. We considered a multi-objective optimization approach based on teaching learning based optimization, referred to as the MOTLBO, which allows efficient and accurate determination of the optimum solutions, As a result, both the total cycle efficiency and price of electricity that are in conflict with each other have been selected as the objective functions, and six design variables have been selected. It has been tried to find the best working fluid as well as the best fuel cells from an economic and thermal efficiency point of view. The paper incorporates equipment selection for the different fuel cells, and cost corrections to estimate the investment cost, with the overall goal of the designing an optimal ORC system by considering different working fluids. Based on multi-objective optimization, the paper finds that R123 is the optimal fluid for ORC based the thermo-economic performance in the cases of all fuel cells except the MCFC (300 kW). We also conducted a parametric study for determination of the effect of varying selected design parameters on the overall efficiency and electricity cost to make comparisons. In all fuel cells except the PEMFC, the highest and lowest pressure ratios are needed when applying R123 and R134a, respectively. In addition, the effect of ORC mass flow rate has been investigated for different configurations using different working fluids. Finally, the results achieved for the design parameters in the case of different fuel cells have been discussed and compared.