Extended Kalman Filter for PEM Electrolyzer Condition Monitoring

Renewable hydrogen produced by water electrolysis is a key technology in the path to cross-sectoral decarbonization. The degradation of electrolysis cells is a central issue for the lifetime of the equipment. Literature describes several degradation mechanisms that affect different cell components, which are best detected in analysis while the system is not in operation. During operation, while the voltage response to an applied current is monitored and used as a measure of efficiency decrease, methods able to quantify aging of the individual cell components during operation are very limited. This work proposes a method based on an extended Kalman filter for estimation of non-measurable degradation effects. The filter is able to separate degradation effect of a proton exchange membrane electrolysis cell and allocating it to components of the membrane electrode assembly, namely to the anode and the membrane. The extended Kalman filter gives an estimation of degradation of catalyst and membrane, using input (current) and output (voltage) data alone.