State estimation for gas purity monitoring and control in water electrolysis systems

TL;DR

This work tackles gas-purity safety in alkaline water electrolysis by estimating upstream impurity with an Extended Kalman Filter (EKF) using a simple, noise-driven model. The approach enables inferential control of the hydrogen-to-oxygen purity ratio $HTO$ at locations lacking direct measurements, by regulating the separator pressure through a cascade PI loop with a switch between measured and estimated feedback. Simulations show that upstream state estimation can reduce the time spent in hazardous $HTO$ conditions by up to an order of magnitude compared to conventional downstream measurements. The results highlight the potential for safer, more flexible electrolysis operation and point to experimental validation and broader control-method extensions as valuable future steps.

Abstract

Green hydrogen, produced via water electrolysis using renewable energy, is seen as a cornerstone of the energy transition. Coupling of renewable power supplies to water electrolysis processes is, however, challenging, as explosive gas mixtures (hydrogen in oxygen) might form at low loads. This has prompted research into gas purity control of such systems. While these attempts have shown to be successful in theoretical and practical studies, they are currently limited in that they only consider the gas purity at locations where composition measurements are available. As these locations are generally positioned downstream of the disturbance origin, this incurs considerable delays and can lead to undetected critical conditions. In this work, we propose the use of an Extended Kalman Filter (EKF) in combination with a simple process model to estimate and control the gas composition at locations where measurements are not available. The model uses noise-driven states for the gas impurity and is hence agnostic towards any mechanistic disturbance model. We show in simulations that this simple approach performs well under various disturbance types and can reduce the time spent in potentially hazardous conditions by up to one order of magnitude.

Figures (5)

• Figure 1: Sketch of the full plant model with magnified pipe and cell sections.
• Figure 2: Simplified model used for the state estimator.
• Figure 3: Block diagram of the concentration control structure. The logic switch indicates that either state, or measurement feedback is taken into account.
• Figure 4: Upper panel: Normalized disturbance sequence. Lower pannel: True gas purity values in the pipe ($HTO_{\textit{n}}$) and separator ($HTO_{\textit{n}+1}$), and their estimate ($\widehat{HTO_{\textit{n}}}$) and noisy measurement ($HTO_{\textit{n}+1}^{\mathbf{v}}$), respectively, together with the alarm limit (AL).
• Figure 5: Simulation results for, (a), the state feedback scenario, and (b), the measurement feedback scenario. Upper panels: True gas purity values in the pipe ($HTO_{\textit{n}}$) and separator ($HTO_{\textit{n}+1}$), and their respective estimate ($\widehat{HTO_{\textit{n}}}$) and noisy measurement ($HTO_{\textit{n}+1}^{\mathbf{v}}$) together with the alarm limit (AL). Lower panels: Pressure as manipulated variable.