Tutorial

Tutorial Title — Active Gate Driver Enabled Smart Power Electronics Solution

Presenter: Dr. Zhengyang(Jenny) Feng, Bristol Research & Innovation Laboratory, Toshiba Europe Limited

Dr. Sheng Wang, Cardiff University

Dr Manish Kumar, Cardiff University & Bristol Research & Innovation Laboratory, Toshiba Europe Limited


Abstract: The widespread adoption of SiC MOSFETs is driven by their superior switching performance and matured fabrication processes. To unlock the full potential of SiC MOSFETs, active gate drivers offer optimal driving patterns, minimizing switching losses and overshoots for enhanced efficiency and EMC performance in various applications. SiC MOSFETs exhibit inherent variations in key parameters due to semiconductor fabrication processes and temperature sensitivity. From an application perspective, slew rate control and related noises can either expedite or impede the design process, becoming critical factors in converter prototyping and time-to-market. Addressing these considerations, active gate drivers must possess precise controllability and adaptability to varying application requirements. This presentation introduces a fully integrated digital active gate driver augmented with an artificial neural network for optimizing the drain current overshoot and energy loss during switching transients. The on-chip ADC and 2-ns gate current control enable surge current cancellation under various load current conditions. The programmable capabilities of the active gate driver allow adaptability to different application scenarios. An on-chip lookup table ensures easy implementation, contributing to efficient and flexible SiC MOSFET-driven converter designs. This tutorial will also discuss the topic of AI-driven closed-loop design of power electronics converters, focusing on multiple-objective optimization of PE converters featuring high accuracy, fast computational speed, and transparency.