Electrochemical impedance spectroscopy (EIS) has been exploited as a rich source of Proton Exchange Membrane Fuel Cell (PEMFC) diagnostic information for many years. Several investigators have characterized different failure modes for PEMFCs using EIS and it now remains to determine how this information is to be obtained and used in a diagnostic or control algorithm for an operating PEMFC. This work utilizes the concept of impedance spectral fingerprints (ISF) to uniquely identify between failure modes in an operating PEMFC. Three well documented PEMFC failure modes, carbon monoxide (CO) poisoning, dehydration, and flooding were surveyed, modelled, and simulated in the time domain and the results were used to create a database of ISFs. The time domain simulation was realized with a fractional order differential calculus state space approach. A primary goal of this work was to develop simple and cost effective algorithms that could be included in a PEMFC on-board controller. To this end, the ISF was discretized as coarsely as possible while still retaining identifying spectral features using the Goertzel algorithm in much the same way as in dual tone multi-frequency detection in telephony. This approach generated a significant reduction in computational burden relative to the classical Fast Fourier Transform approach. The ISF database was used to diagnose simulated experimental PEMFC failures into one of five levels of failure: none (normal operation), mild, moderate, advanced, and extreme from one of the three catalogued failure modes. The described ISF recognition algorithm was shown to correctly identify failure modes to a lower limit of SNR = 1dB.

No description available.

Electrochemical impedance spectroscopy (EIS) has been exploited as a rich source of Proton Exchange Membrane Fuel Cell (PEMFC) diagnostic information for many years. Several investigators have characterized different failure modes for PEMFCs using EIS and it now remains to determine how this information is to be obtained and used in a diagnostic or control algorithm for an operating PEMFC. This work utilizes the concept of impedance spectral fingerprints (ISF) to uniquely identify between failure modes in an operating PEMFC. Three well documented PEMFC failure modes, carbon monoxide (CO) poisoning, dehydration, and flooding were surveyed, modelled, and simulated in the time domain and the results were used to create a database of ISFs. The time domain simulation was realized with a fractional order differential calculus state space approach. A primary goal of this work was to develop simple and cost effective algorithms that could be included in a PEMFC on-board controller. To this end, the ISF was discretized as coarsely as possible while still retaining identifying spectral features using the Goertzel algorithm in much the same way as in dual tone multi-frequency detection in telephony. This approach generated a significant reduction in computational burden relative to the classical Fast Fourier Transform approach. The ISF database was used to diagnose simulated experimental PEMFC failures into one of five levels of failure: none (normal operation), mild, moderate, advanced, and extreme from one of the three catalogued failure modes. The described ISF recognition algorithm was shown to correctly identify failure modes to a lower limit of SNR = 1dB.

No description available.

No description available.

Thesis (Ph. D.)--Ohio University, August, 1981. / Title from PDF t.p.

Thesis (M.S.)--University of Wisconsin--Madison, 1969. / eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1982. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis--Wisconsin. / Vita. Includes bibliographical references (leaves 196-203).

Electrochemical impedance spectroscopy (EIS) has been exploited as a rich source of Proton Exchange Membrane Fuel Cell (PEMFC) diagnostic information for many years. Several investigators have characterized different failure modes for PEMFCs using EIS and it now remains to determine how this information is to be obtained and used in a diagnostic or control algorithm for an operating PEMFC. This work utilizes the concept of impedance spectral fingerprints (ISF) to uniquely identify between failure modes in an operating PEMFC. Three well documented PEMFC failure modes, carbon monoxide (CO) poisoning, dehydration, and flooding were surveyed, modelled, and simulated in the time domain and the results were used to create a database of ISFs. The time domain simulation was realized with a fractional order differential calculus state space approach. A primary goal of this work was to develop simple and cost effective algorithms that could be included in a PEMFC on-board controller. To this end, the ISF was discretized as coarsely as possible while still retaining identifying spectral features using the Goertzel algorithm in much the same way as in dual tone multi-frequency detection in telephony. This approach generated a significant reduction in computational burden relative to the classical Fast Fourier Transform approach. The ISF database was used to diagnose simulated experimental PEMFC failures into one of five levels of failure: none (normal operation), mild, moderate, advanced, and extreme from one of the three catalogued failure modes. The described ISF recognition algorithm was shown to correctly identify failure modes to a lower limit of SNR = 1dB. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate