On condition monitoring of high frequency power GaN converters with adaptive prognostics

Abstract

There is no doubt that in the future, a need for higher switching frequency is inevitable to extract the full benefits of reliable Gallium Nitride (GaN) device characteristics. Along with the reliability enhancement for GaN-based power converters, it is essential to monitor a precursor signature identification for diagnostics/prognostics techniques. With the availability of the most granular information deduced from advanced devices, a new data-driven scheme is proposed for system monitoring and possible lifetime extension of 400W power GaN converters at 100kHz. The approach relies on the real-time R <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ds</sub> ( <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">on</sub> ) data extraction from the power converter, and calibration of an adaptive model using multi-physics co-simulations under thermal cycling. More specifically, the focus is on deploying machine learning algorithms to exploit for the parameter estimation in power electronics engineering reliability.