Optimized model of microbial fuel cell using AI-driven walrus optimization algorithm

Abstract

The objective of this paper is to present a robust and accurate model of microbial fuel cell (MFC) to use it for a parameter optimization and performance improvement. The proposed methodology contains main two phases: ANFIS modeling and parameter optimization by recent Walrus Optimization Algorithm (WOA). The ability of MFC to cleanse wastewater by removing organic matter and heavy metal ions while simultaneously generating power has been established. The power production and wastewater treatment efficacy of MFCs are significantly influenced by optimizing their operating parameters. Therefore, in this research work, the operation of an MFC for wastewater treatment, Cr(VI) removal, and power output was modelled and optimized using integration between ANFIS and WOA. Key determining factors include the substrate concentration on the anode side, the Cu(II)/Cr(VI) ratio on the cathode side (which acts as electron acceptor), and the applied external resistance between the anode and cathode. The Cr(VI) removal and power output are the output parameters. The findings from the ANFIS and the WOA demonstrated the effectiveness of the model and optimizer in determining the optimal operating conditions for maximizing power output and Cr(VI) removal. WOA successfully increased the power density from 75 to 86.65 mW/m², representing an approximate 15.33% improvement over the measured data. Additionally, it enhanced the Cr(VI) removal efficiency by approximately 18.9%, raising it from 36.76% to 43.74%. Under these conditions, the optimum Cu(II)/Cr(VI) ratio, external resistance, substrate concentration, and total objective function were 1.672 mg/mg, 1.756 kΩ, 1.756 g/L, and 130.39, respectively.