A discrete Fourier transform-based fuel concentration and permeation sensing scheme for low temperature fuel cells

Abstract

Due to the fuel permeation phenomenon, fuel cells of low temperature suffer limitations in power density and energy density. It is especially crucial to a direct methanol fuel cell. Controlling the fuel concentration within an appropriate range thus becomes significant. This paper proposes a practical sensing scheme to estimate the methanol concentration and permeation fluxes. It can also serve as a gas sensor to measure the oxygen concentration as well. By acquiring the transient response of the open-circuit voltage under oxygen starvation conditions, a measurement index is derived via the Discrete Fourier Transform (DFT) and thereby performs the estimation. The approach takes only 3s to sample the transient open-circuit voltages, while a root mean square error of the estimated methanol concentration is 1.4% relative to the maximum measurement magnitude of 2.5M. It is also validated for a wide range of oxygen concentration in the illustrative examples. There is no strict requirement for the data acquisition unit. A sampling frequency of 5Hz is enough, and this enhances its feasibility and reduces its cost. Based on the pre-established data, it makes the estimation of methanol permeation fluxes portable without carrying delicate equipments. Finally, an area-oriented index deduced from the zero-frequency term of the DFT serves as an alternative measurement basis that provides more pathways to evaluate the sensing results. The notion proposed in this paper is not only suggested for small portable fuel cells. It possesses potential for developing other kinds of electrochemical sensors as well.