Predictive & Prognostic Controller for Wide Band Gap (Silicon Carbide) Power Conversion

This paper presents an approach to predictive and prognostic control, combining prognostics with predictive control to extend life, for the purpose of increasing the confidence levels for power converters used in aerospace applications. One goal of this work is to accelerate the adoption of new technologies, such as advanced power semiconductor devices. While the devices are immature the technology provides significant overall performance advantages. Using prognostication it is hoped to significantly increase the confidence levels for successful mission completion and so overcome reservations about adopting these new technologies. The methodology uses predictive modeling and simple, robust, sensing, attempting to avoid excessive sensing requirements in the prognostics of complex converters. This approach requires detailed knowledge of the dominant aging and failure mechanisms, the physical driving forces behind these mechanisms and the observable precursors of failure. The precise history of the power converter system, combined with measurement and modeling, can then predict whether the system can operate successfully for a given duration, with a high degree of confidence. The information could also be used to guide decisions about the future use or loading of specific converters in larger, parallel, redundant systems.