Lifetime Modeling of IGBT Devices Based on Micro-Defect Topography Inversion

As a key component of traction transmission system, the reliability of power electronic devices is an important guarantee of system safety. However, the existing studies mainly focus on the life assessment of the bathtub curve failure loss period, and lack the state characterization and remaining life prediction methods of the bathtub curve random failure period. In this paper, the solder layer CT scanning results of IGBT devices are combined with the finite element simulation model to invert the fatigue degree of the alive mesh through the stress relation between the dead and alive mesh and the microscopic representation of the Coffin-Manson model. The health state of the device is evaluated according to the void distribution in the solder layer during the random failure period, and the corresponding residual life model of the IGBT device is established to realize the life cycle evaluation of the IGBT device in service. The expected results of this paper enrich the life prediction and reliability evaluation of power devices, provide data basis for the continuous use of replaced IGBT modules, and thus reduce system operation and maintenance costs, which has important strategic value and economic significance.