Variance-based global sensitivity analysis of a multi-population, single-chamber microbial fuel cell operating in continuous flow mode at steady state

Microbial fuel cells (MFCs) are environmentally friendly devices which are used to convert chemical energy in organic wastes to electrical energy. MFCs have a strong non-linearity that requires a very sophisticated controlling system. Consequently, this makes optimization and performance study of MFCs a difficult process. For better estimation of the constants used for optimization of MFCs, global sensitivity analysis is performed. The global sensitivity method based on Sobol’s indices coupled with Monte Carlo simulations was applied on multi-population, single-chamber MFC operating in a continuous flow at steady state for the first time. In this paper, first-order and total-order sensitivity indices were used to visualize the impacts associated with six main parameters resulted from the maximization of power density using Matlab. Such parameters are maximum anodophilic-specific growth rate, half-rate constant of anodophilics, curve steepness factor, mediator half-rate constant, number of electrons transferred per mole mediator and decay rate constant of anodophilic bacteria. The results showed that the curve steepness factor has almost no impact on the power density of MFC. While all other studied, factors are sensitive parameters that impact the power density of MFC. It is worth mentioning that maximum anodophilic growth rate and the number of electrons transferred per mole of mediator are the most sensitive parameters that affecting the power density production having total indices of 0.74 and 0.624, respectively. While the half-rate constant of anodophilics, mediator half-rate constant and decay rate constant of anodophilics have almost similar impact by having total-order indices of 0.127, 0.144 and 0.192, respectively. The findings herein are critical in understanding and further model improvement of microbial fuel cells as the most impacting parameters on MFC power density can be optimized further to reduce uncertainty associated with the experimental parameters in the model. Graphical abstract