Skip to main content

Assessment and optimization of an integrated energy system with electrolysis and fuel cells for electricity, cooling and hydrogen production using various optimization techniques


Amir H. Keshavarzzadeh, Pouria Ahmadi, Mohammad Reza Safaei*

Source title: 
International Journal of Hydrogen Energy, 44: 21379-21396, 2019 (ISI)
Academic year of acceptance: 

In this research study, a novel integrated solar based combined, cooling, heating and, power (CCHP) is proposed consisting of Parabolic trough solar collectors (PTSC) field, a dual-tank molten salt heat storage, an Organic Rankine Cycle (ORC), a Proton exchange membrane fuel cell (PEMFC), a Proton exchange membrane electrolyzer (PEME), and a single effect Li/Br water absorption chiller. Thermodynamics and economic relations are used to analyze the proposed CCHP system. The mean of Tehran solar radiation as well as each portion of solar radiation during 24 h in winter is obtained from TRNSYS software to be used in PTSC calculations. A dynamic model of the thermal storage unit is assessed for proposed CCHP system under three different conditions (i.e., without thermal energy storage (TES), with TES and with TES + PEMFC). The results demonstrate that PEMFC has the ability to improve the power output by 10% during the night and 3% at sunny hours while by using TES alone, the overnight power generation is 86% of the power generation during the sunny hours. The optimum operating condition is determined via the NSGA-II algorithm with regards to exergy efficiency and total cost rate as objective functions where the optimum values are 0.058 ($/s) and 80%, respectively. The result of single objective optimization is 0.044 ($/s) for the economic objective in which the exergy efficiency is at its lowest value (57.7%). In addition, results indicate that the amount of single objective optimization based on exergetic objective is 88% in which the total cost rate is at its highest value (0.086 $/s). The scattered distribution of design parameters and the decision variables trend are investigated. In the next step, five different evolutionary algorithms namely NSGA-II, GDE3, IBEA, SMPSO, and SPEA2 are applied, and their Pareto frontiers are compared with each other.