Monitoring System for the Online Estimation of the Internal Parameters of a Microbial Fuel Cell Operated Intermittently

This research study describes the operation of a Microbial Fuel Cell (MFC) under pulse-width modulated connection of the external resistor (RPWM mode). The study is composed of 2 main sections: the first one describes the MFC operation at low and high frequencies, and under varying periods of external resistance connection/disconnection or duty cycle (D). The second part blends the RPWM mode of operation to the analytical solution of a proposed Equivalent Circuit Model (ECM) in order to develop an on-line monitoring procedure. This monitoring strategy allows the periodic follow-up of the MFC internal parameters. Regarding the first part of this project, the analysis of the output voltage profiles acquired during R-PWM tests showed the presence of slow and fast dynamic components, which can be described by a simple equivalent circuit model (ECM) suitable for process control applications. At operating frequencies above 100 Hz a noticeable improvement in MFC performance was observed with the power output increase of 22–43% as compared to MFC operation with a constant external resistance. Based on the identified dynamics for the MFC dynamics at low and high frequency, the analytical solution for the equivalent circuit model is used as a basis for an on-line monitoring strategy. This algorithm allows the online estimation of the MFC internal parameters. To validate the accuracy of the estimations, low and high frequency voltage profiles are generated and statistically compared to actual voltage profiles for the MFC. When actual and online estimated profiles are compared, mean square errors of 5.26e-4 for low frequency profiles, and 9.69e-6 for high frequency profiles were obtained. On-line estimations of the MFC internal parameters are also compared to off-line estimated parameters, and similar evolution trends are observed. The on-line monitoring strategy was tested under 2 different operating conditions: first, one MFC was used in under to study parameter evolution under varying organic load. Then, a second MFC was newly built in order to make a periodic follow-up of the internal parameters' evolution in time, from the MFC start-up to its maturity.