Insulation Life Prediction for Silicon Carbide (SiC) Motor-Drive Systems
Abstract
Emerging SiC devices switch much faster than Silicon devices and the voltage stress, thermal stress and frequency effects of these devices significantly increase the stress on motor drive insulation. These devices have also enabled motor drive applications to be designed for higher dc link voltages than current silicon devices. The combined effect of higher dc link voltages, higher voltage rise times (dv/dt) and higher switching frequencies of SiC motor drives has made it imperative to predict insulation life for SiC inverter applications. Much as some knowledge exist on the impact of various stresses, such as voltage, temperature, voltage rise time, partial discharge, vibration, humidity, and pressure on the aging and failure of conventional motor insulation systems, there is sparse knowledge regarding these issues in SiC applications. Furthermore, when the stress factors combine, they can lead to an even shorter lifespan for insulation. However, the synergy between stress factors and the complex mechanism by which degradation is accelerated is currently not well understood. Considering that insulation life can be exacerbated by the increased level of stresses from SiC devices, a SiC insulation life prediction model that accounts for the combined effects of multiple stress factors and their interactions is needed. Multi-factor models developed in the literature to account for the impact of multiple stresses contained at mosttwo stress factors (temperature and voltage stress). This is because, the number of variables and tests that are needed to determine the parameters of the models dramatically increases with an additional stress factor. As evident in the literature, some of these models produced conflicting results for conventional insulation systems. Design of Experiments (DOE) are statistical methods that can enable to conduct and analyze test results and help to reduce the number of experimental tests conducted to determine the parameters of the model. Coupled with analytical, finite element models to properly account for the physics of failure mechanisms and statistical modeling to process DOE based accelerated life testing results, it is possible to develop a multi-factor model to predict the life of SiC insulation systems. Therefore, the objective of this project is to develop a physics-based insulation life prediction model for SiC motor-drive systems, considering multiple stress factors including environmental and operational (mechanical) factors, and conduct experimental validation of the model.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 15, 2023
- Source ID
- N000142312424
Entities
People
- Emmanuel Agamloh
Organizations
- Baylor University
- Office of Naval Research
- United States Navy