Physics-Informed Condition Monitoring of SiC Power Modules

2026-08-08

AI Summary

This study addresses the lack of efficient health monitoring methods for silicon carbide (SiC) power modules in embedded applications, where physics-based models are difficult to deploy online and purely data-driven approaches suffer from poor generalization and heavy reliance on large labeled datasets. To overcome these limitations, this work proposes a lightweight hybrid framework that integrates physical priors with data-driven learning. The approach incorporates cumulative damage features derived from Miner’s rule, enforces monotonic degradation through gradient penalty constraints, and models outputs using a heavy-tailed distribution to enhance robustness against anomalous disturbances—such as wire bond lift-off—and improve uncertainty calibration. Evaluated on Infineon’s industrial power cycling dataset, the method reduces mean absolute error by approximately 70% compared to purely data-driven baselines, demonstrates stable performance under rigorous cross-validation, and meets the computational constraints required for embedded deployment.

Abstract

Silicon carbide (SiC) power modules are increasingly deployed in automotive traction inverters, where condition monitoring is essential to prevent in-service failures. Despite extensive qualification under AQG 324, no consolidated approach exists for in-field health state estimation: physics-of-failure lifetime models lack real-time applicability, purely data-driven architectures require large labeled datasets and generalize poorly, and physics-informed frameworks remain too demanding for embedded deployment. We address SiC MOSFET modules assembled with sintered packaging, which suppresses solder degradation and produces aging behavior distinct from previously studied devices. Instead of the smooth quasi-exponential drift of solder-based modules, the forward voltage drop $V_{DS}$ exhibits multi-regime profiles, with wirebond liftoff events introducing abrupt, non-monotonic perturbations. We propose a condition monitoring framework combining three elements. First, physics-informed features replace raw sensor signals with cumulative damage indicators derived from junction temperature swing, mean junction temperature and a Miner rule accumulator, encoding degradation history in an interpretable form. Second, a monotonicity constraint enforced by gradient penalty regularization embeds the expected degradation direction as a physics-guided prior. Third, a heavy-tailed output distribution replaces the point estimate, giving calibrated uncertainty robust to the out-of-distribution variance introduced by liftoff. On an industrial power cycling dataset from Infineon Technologies, several neural architectures are compared under a strict cross-validation protocol. The full configuration reduces mean absolute error by approximately 70% over purely data-driven baselines and stays stable across all folds, while remaining lightweight enough for embedded deployment.

Problem

• condition monitoring
• SiC power modules
• health state estimation
• physics-informed
• degradation modeling

Innovation

• physics-informed monitoring
• SiC power modules
• monotonicity constraint
• cumulative damage indicators
• heavy-tailed uncertainty