Increasing Output Power of a Microfluidic Fuel Cell Using Fuzzy Modeling and Jellyfish Search Optimization

Abstract

An efficient electrochemical energy conversion system with little to no environmental impact is the fuel cell (FC). FCs have demonstrated encouraging results in various applications and can even run on biofuel, such as bio-glycerol, a by-product of biodiesel. The most effective ways to operate FCs can significantly enhance their effectiveness. Incorporating fuzzy modeling and metaheuristic methods, this work used artificial intelligence to determine the ideal operating parameters for a microfluidic fuel cell (MFC). The concentrations of the following four variables were considered: bio-glycerol concentration, anode electrocatalyst loading, anode electrolyte concentration, and cathode electrolyte concentration. The output power density of the MFC was used to assess its performance. The output power density of the MFC was modeled using fuzzy logic, taking into account the aforementioned operational parameters. A jellyfish search optimizer (JSO) was then used to find the ideal operating conditions. The results were contrasted with response surface methodology (RSM) and experimental datasets to demonstrate the superiority of the proposed integration between fuzzy modeling and the JSO. In comparison with the measured and RSM approaches, the suggested strategy boosted the power density of the MFC by 9.38% and 8.6%, respectively.