Modeling and Evaluation of Gas Diffusion in High-Voltage XLPE Cables Considering Buffer Layer Microstructure

Gas detection for diagnosing high-voltage (HV) cross-linked polyethylene (XLPE) cable degradation has emerged as a promising approach. However, a comprehensive understanding of gas diffusion along the cable—especially its contribution to non-uniform axial concentration distributions—is crucial for accurate diagnostics. In this study, we investigate the diffusion behavior of gases within the primary diffusion medium of HV cables, the buffer layer. Focusing on seven key gases commonly used in gas-based diagnostics, we develop finite element method (FEM) simulation models based on microscopic observations of the buffer layer’s porous structure to determine the effective diffusion coefficients of these gases. Furthermore, by incorporating the compression on the buffer layer, we characterize the axial diffusion and distribution patterns of gases in actual cables. Based on these findings, we propose an inverse modeling method to localize gas-emitting defects and predict gas generation characteristics from the spatiotemporal evolution of gas concentrations. Finally, gas diffusion measurements conducted on the experimental platform with pre-designed defect validate our FEM simulations, demonstrating acceptable agreement with experimental data and confirming the accuracy of the proposed models.