Online Monitoring for Aging Detection and Remaining Useful Life Estimation of IGBTs, Considering Main Aging Mechanisms

Online aging monitoring for power converters is an intricate problem due to various aging mechanisms in their components. Among the converters’ components, online aging monitoring of the switches and prediction of their remaining useful life (RUL) are crucial. In this paper, an online aging monitoring approach is presented for IGBTs based on changes in their on-state resistance (R<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">on</inf>), taking into consideration main aging mechanisms, including solder aging, bond wire aging, and mixed aging. A reference model with the same parameters of the converter is run in real time for this purpose. The proposed approach is composed of three main stages. First, the suspected event is detected, when a considerable difference occurs between the estimated switch junction temperature from the measurement and reference model. In the second stage, Ron is calculated for each aging mechanism. Thanks to the predictable behavior of changing Ron during various aging mechanisms, the aging progress level can be determined precisely. Finally, using a Kalman Filter estimator, the RUL of the switch is estimated at each aging progress level. The proposed approach is evaluated by emulating the aging mechanisms on the switch of a boost chopper in different experiments.