Review on anodic porous transport layers in PEMWE: Physics-grounded structure–transport–electrical relationships and ML-enabled prediction and optimization

Proton exchange membrane water electrolysis (PEMWE) is a promising pathway for low-carbon hydrogen production, but its efficiency and durability are still limited by coupled transport and resistive losses. On the anode side, the porous transport layer (PTL) controls water supply, oxygen removal, mechanical support, and electronic conduction, while also having a strong influence on interfacial contact resistance and other ohmic losses. This review examines physics-grounded structure–transport–electrical relationships in titanium-based anodic PTLs, with machine learning treated as a supporting tool for characterization, surrogate prediction, and constrained optimization. Key structural descriptors, electrical metrics, and simplified physical relationships linking microstructure, contact mechanics, and two-phase transport to measurable losses are summarized. ML approaches for tabular, image/voxel, and graph-based data are also discussed, with direct PTL-specific evidence distinguished from transferable methods developed in related porous-media systems. The overall aim is to improve cross-study comparability and to support more rigorous, manufacturability-aware ML applications in PEMWE anodic PTL research.