(Digital Presentation) Dynamic Thermal Detection of Water Electrolyzer with Dual-Layer Characteristic Temperature Based on IR

Abstract

As the prevalence of water electrolysis systems continues to expand, increased attention has been placed on the dynamic thermal detection of the electrolyzer. Presently, the scope of dynamic thermal detection for water electrolysis systems is limited to solely the temperatures at the inlet and outlet. This study propounds a novel dual-layer characteristics temperature (DLCT) model for the purpose of water electrolyzer temperature monitoring. The DLCT model addresses two significant issues of the dynamic thermal detection of the WEE: characteristic temperature extraction difficulty and signal perturbation caused by auxiliary factors. The first layer of multi-gaussian distribution (MGD) regression in the DLCT model quantifies the sectional temperature of the electrolyzer surface and produces numerous potential characteristic temperatures (CT). These temperatures are subsequently assigned to each section and are then conveyed to the second layer of linear regression, which provides an incisive quantized temperature variation pattern for the electrolyzer surface temperature distribution. Importantly, the DLCT model requires no supplementary modifications to the water electrolysis system or the installation of temperature sensors within or on the surface. By employing the DLCT model during dynamic operation, the ability to monitor the temperature of the electrolyzer comprehensively is improved, the temperature distribution of the electrolyzer can be quantized and richer insights regarding the surface temperature distribution can be gained, promoting better thermal uniformity, as well as pin-pointing the potential hot spot. Hence this method can also be used for structure design to produce better uniformity. The dynamic temperature detection results under different operating conditions are analyzed, and it can be concluded that the operating parameters, especially the lye circulation flow, can greatly influence the temperature distribution uniformity of the electrolyzer significantly. In conclusion, the DLCT model effectively handles the two primary challenges in dynamic thermal detection of water electrolysis systems by providing global and sectional characteristic temperature readings, enabling electrolyzer to be monitored holistically. So that future improvement regarding both structural and operational optimization can be conducted in improving the thermal uniformity. Figure 1