Acceleration of Modelling with Physics Informed Learning: Frameworks and Perspectives for Real-Time Control of Electrochemical Devices

TL;DR

Electrochemical devices require real-time control with ms-scale predictions, but their governing coupled multi-physics PDEs hinder fast simulation. The paper compares three physics-informed learning frameworks—PINN, PI-DeepONet, and PINO—across training effort, inference speed, and extrapolation, mapping their strengths to device geometries and operating regimes. PI-DeepONet excels in irregular, porous geometries, PINO offers rapid inference on regular, layered domains with strong extrapolation, and PINN provides mesh-free, simple deployment for fixed problems. The findings support physics-informed operator learning as a transformative path toward real-time, safe, and efficient control of batteries, fuel cells, and electrolyzers, with offline training shifting the computational burden away from fast decision-making.

Abstract

Electrochemical devices (batteries, fuel cells, and electrolyzers) are in full development, driven by the green energy transition. Their real-time control requires ms predictions in order to take critical decisions during fast transients or faults. The physics behind include coupled multi-physics phenomena that conventional finite element methods cannot solve so fast with the current CPU technology. This paper evaluates the potential of physics-informed machine learning represented by three frameworks: \ac{pinn}, \ac{pideeponet}, and \ac{pino} by evaluating their training effort, inference speed, and extrapolation capacity. Our analysis reveals valuable performance trade-offs. \acp{pinn} offer simplicity for fixed problem instances but require retraining for parameter changes. \ac{pideeponet} enables operator learning across varying conditions with mesh-free geometric flexibility. \ac{pino} delivers superior performance on regular grids, with the strongest extrapolation capabilities due to spectral derivative computation and resolution invariance. \ac{pideeponet} is particularly suited for irregular, unstructured geometries (e.g., porous electrodes or complex flow fields), while \ac{pino} works best for layered, structured-grid problems (e.g., transport across stacked electrochemical layers) requiring fast inference. Possible future applications include real-time lithium concentration prediction for safe fast-charging and micro short circuit detection, water management in fuel cells, and optimal power management in electrolyzers under intermittent renewable inputs. These findings establish physics-informed operator learning as a transformative approach for next-generation electrochemical device controller technology.

Figures (7)

• Figure 1: Architecture of PINN learning framework
• Figure 2: Performance of the PINN for battery thermal monitoring
• Figure 3: The architecture of pideeponet
• Figure 4: Architecture of the PI-SPM. (Adapted from zhuang_physics-informed_2026)
• Figure 5: Training results of the PI-SPM models. Left: loss curves; right: training errors.