Maximizing power production in a stack of microbial fuel cells using multiunit optimization method

Abstract

This study demonstrates real-time maximization of power production in a stack of two continuous flow microbial fuel cells (MFCs). To maximize power output, external resistances of two air-cathode membraneless MFCs were controlled by a multiunit optimization algorithm. Multiunit optimization is a recently proposed method that uses multiple similar units to optimize process performance. The experiment demonstrated fast convergence toward optimal external resistance and algorithm stability during external perturbations (e.g., temperature variations). Rate of the algorithm convergence was much faster than in traditional maximum power point tracking algorithms (MPPT), which are based on temporal perturbations. A power output of 81-84 mW/L A (A ¼ anode volume) was achieved in each MFC. IntroductionPower generation in a microbial fuel cell (MFC) from renewable carbon sources is a potential alternative to fossil fuel utilization.1,2 In a MFC, anodophilic microorganisms degrade organic matter and transfer electrons to the anode via nanowires or self-produced mediators. [3][4][5] Since the late 90s, when intensive MFC development began, power density in MFCs increased by several orders of magnitude. 6 Yet, attainable power density of a single MFC is relatively low, in a range of 50-200 W/m 3 A (A ¼ anode chamber volume) and the working voltage is limited to 0.3-0.5 V. 7 Consequently, a stack of MFCs might be required to obtain the desired power output. 7,8 As in any other battery, power generation in a MFC strongly depends on the external resistance (load) so that maximum power is produced when the external load is equal to the internal resistance of the cell. 9 As MFC is a biological system, the internal resistance depends on environmental factors such as temperature and influent composition. As a consequence, timely adjustment of the external load is required to maximize power production.The classical approach of real-time optimization consists of two steps.10,11 First, a model of the process is used to numerically calculate the optimum. Next, the model is updated using the available measurements and the updated model is then used for numerical optimization. However, building and maintaining a sufficiently detailed model of a MFC represents a challenge in itself. Also, the solution does not converge to the optimum if the model structure does not adequately describe the process. 12Extremum-seeking is an alternative approach 13,14 where optimization is achieved by following the necessary conditions of optimality, i.e., in an unconstrained case, forcing the gradient to zero. For gradient estimation, perturbation methods 15 can be used, when measurements of the performance criterion are available. If only auxiliary measurements are available, a model-based gradient estimation approach is needed.16 Maximum power point tracking (MPPT) algorithms used in photovoltaic systems are also based on an estimation of the gradient. These algorithms can also be used to maximize the electrical power of microbial fuel cells in real-time....

Microbial Fuel Cell Operation With Periodic Connection of the External Resistance

“…2A). These estimations are in agreement with an external resistance of 18 -25 Ω used during MFC operation by the computer-controlled perturbation/observation algorithm, which maximized MFC power output by real time external resistance correction (Pinto et al 2011;Woodward et al 2009).…”

“…However, an aspect that has been overlooked is how the produced energy is harvested. MFC power output is maximized when the electrical load (external resistance) connected to the cell is equal to the total internal resistance (Aelterman et al 2008;Logan 2008;Woodward et al 2009). In laboratory tests, MFCs are often operated with a constant external resistance.…”

Application of iron-based cathode catalysts in a microbial fuel cell

“…The MFC was equipped with lines for influent, effluent and liquid recirculation. More details on MFC construction can be found elsewhere [22].…”

“…Throughout the MFC tests, the power output was maximized using a perturbation and observation maximum power point tracking (MPPT) algorithm, which constantly adjusted the value of the external resistance (R ext ) connected to the MFC so that it was close to the MFC internal resistance [22,23]. Polarization tests were periodically carried out as follows.…”

The influence of operational conditions on the performance of a microbial fuel cell seeded with mesophilic anaerobic sludge

“…However, a cautious approach should be taken as to avoid MFC operation at external resistances below the internal resistance value as this results in the abrupt performance decline. In this regard, an on-line control of electric load must be considered in order to maximize MFC power output at varying operating conditions [31,32].…”