Research on Machine Learning-Assisted Numerical Simulation Methods for IGBT Module Bonded Interface Fatigue Damage
- 21 April 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Power Electronics
- Vol. 13 (3) , 3259-3272
Abstract
Bonded interface cracking is the primary failure mode of insulated gate bipolar transistor (IGBT) modules during prolonged operation. A numerical simulation method that effectively characterizes fatigue crack propagation at the bonded interface of IGBT modules under power cycling conditions was developed in this study. First, power cycling tests were performed for different numbers of cycles, followed by pull tests on the Al bond wires, obtaining load-displacement (F-δ) curves for the bonded interface after different power cycles. Then, a finite element-fatigue cohesive zone model (FE-FCZM) was established, considering the cumulative effects of interface damage during the power cycling. To determine the FCZM parameters, the influence of these parameters on the F-δ curves was investigated using sensitivity analysis, and the F-δ curves for different degrees of damage at the bonded interface were calculated after various numbers of power cycles. A machine learning (ML) model was subsequently established utilizing the simulated F-δ curve as input and the FCZM parameters as outputs, effectively identifying the FCZM parameters. Finally, the F-δ curves obtained from numerical simulations and experiments, as well as the degree of damage at the bonded interface, were compared to verify the accuracy of the ML method and the numerical simulation method for bonded interface fatigue damage proposed in this study. In addition, the proposed method was used to predict the lifetime of IGBT modules, and compared with several commonly used lifetime models. The results show that the proposed method has the smallest error among all the models that were compared. This article provides a numerical method for describing the damage failure of the bonded interface in IGBT modules, which is significant for lifetime prediction and reliability assessment.Keywords
Funding Information
- National Natural Science Foundation of China (12272013)
This publication has 48 references indexed in Scilit:
- A CNN-LSTM framework for flight delay predictionExpert Systems with Applications, 2023
- Use of a machine learning-based framework to approximate the input features of an intrinsic cohesive zone model of recycled asphalt mixes tested at low temperaturesConstruction and Building Materials, 2023
- A robust fatigue parameter determination method for a local fatigue cohesive zone modelInternational Journal of Fatigue, 2023
- Identification of the parameters contained in a cyclic cohesive zone model for fatigue crack propagationEngineering Fracture Mechanics, 2023
- An EBSD Study of Fatigue Crack Propagation in Bonded Aluminum Wires Cycled from 55°C to 85°CJournal of Electronic Materials, 2022
- Cohesive zone modeling of fatigue crack propagation in slab track interface under cyclic temperature loadEngineering Failure Analysis, 2022
- Identification of the interfacial cohesive law parameters of FRP strips externally bonded to concrete using machine learning techniquesEngineering Fracture Mechanics, 2021
- On damage accumulations in the cyclic cohesive zone model for XFEM analysis of mixed-mode fatigue crack growthComputational Materials Science, 2009
- Computational modelling of impact damage in brittle materialsInternational Journal of Solids and Structures, 1996
- Yielding of steel sheets containing slitsJournal of the Mechanics and Physics of Solids, 1960