Sensitivity Analysis Method of Temperature-Dependent Parameters During Turn-on Process of SiC Power MOSFETs

Elevated temperature and thermal cycling are one of the biggest limiting aspects in the service lifetime of power modules in power electronic systems. Thus, it is of major importance to accurately measure high-bandwidth junction temperature of power semiconductors for condition monitoring as well as for remaining useful life estimation procedures. The extraction of temperature sensitive electrical parameters (TSEP) has been identified as an appealing approach for junction temperature sensing. However, such parameters are usually affected not only by temperature but also by other factors, such as load current, DC voltage, control strategy and aging effects. In addition, the overall temperature sensitivity of the TSEP is affected by several individual sensitivities of separate device parameters. A strong understanding of these impacts is crucial for highly accurate temperature determination. As of yet, sensitivity analysis of TSEPs to these influences is rarely performed in the literature. Therefore, this paper introduces a sensitivity analysis method of turn-on process TSEPs derived from fitting a simulation model to measurement results through a multi-criteria genetic algorithm. To verify the method, the sensitivity of SiC power MOSFET turn-on TSEPs to threshold voltage, internal gate resistance and transconduction gain are extracted and evaluated.