Search for: All records

Silicon-carbide (SiC) MOSFET devices are increasing in popularity in high-power converter applications. Device on-resistance (Rdson) is an important indicator for SiC MOSFET health status. Increments in Rdson over device lifetime indicate imminent device failure and result in decreased system efficiency. Direct and accurate measurement of SiC MOSFET device Rdson in high-power applications is difficult. Another approach is to estimate/predict the device Rdson from other, more easily measurable quantities, however, little work has been done on this approach in the literature. This leaves a significant technical gap in measuring/predicting device Rdson and slows down the device health status monitoring and power converter reliability research. To address the technical gaps, this work proposes a novel approach to predicting device Rdson from thermal cycle count and instantaneous temperature using machine learning regression models. The actual hardware data collected from accelerated lifetime tests of high-power SiC MOSFETs are used to train, test, and validate the proposed machine-learning regression models. The developed models, coupled with cycle counting algorithms, and device case thermal measurements, provide accurate live estimates of Rdson and can be used to predict changes in Rdson over expected mission profiles during power converter design. 

As transportation electrification keeps accelerating across a wide range of vehicle classes from light-duty cars to heavy-duty trucks, the need for high-power electric vehicle (EV) charging equipment continues to grow rapidly. Even though the advancements in power electronics are enabling higher efficiency for EV chargers, thermal management continues to be a significant challenge in high-power charger development Liquid cooling with cold plates is commonly used for dissipating the heat generated by semiconductor devices m high-power chargers To design an effective and optimized thermal management system, accurate thermal modeling and analysis are critical, especially m the preliminary design phases. Complex fluid dynamics (CFD) software such as Ansys has been widely used for thermal modeling and analysis in the literature; however, using CFD analysis tools can be expensive, time-consuming, and computationally intense. To address the technical needs for a rapid, accurate preliminary thermal analysis tool, this paper presents a novel and accurate thermal modeling and analysis approach for high- power EV chargers with liquid cooling and Silicon Carbide (SiC) MOSFETs mounted on cold plates. The proposed modeling and analysis approach utilizes a lumped element model for each of the many pieces within the system to mathematically represent the physical system and form thermal networks. The effectiveness, accuracy, and light computational load of the proposed approach have been validated through experimental results conducted on a 21 kW power converter module hardware from a 1 MW EV wireless charge developed by the team for Class 8 semi-trucks.