Vertical GaN Diode BV Maximization through Rapid TCAD Simulation and ML-enabled Surrogate Model
This work addresses the challenge of optimizing semiconductor device performance for electronics engineers, though it is incremental as it builds on existing TCAD and ML techniques.
The paper tackled the problem of maximizing the breakdown voltage (BV) of vertical GaN diodes by developing a faster simulation method and a machine learning surrogate model, achieving a 5X speedup in simulation and designing a structure with 1887V BV (89% of ideal).
In this paper, two methodologies are used to speed up the maximization of the breakdown volt-age (BV) of a vertical GaN diode that has a theoretical maximum BV of ~2100V. Firstly, we demonstrated a 5X faster accurate simulation method in Technology Computer-Aided-Design (TCAD). This allows us to find 50% more numbers of high BV (>1400V) designs at a given simulation time. Secondly, a machine learning (ML) model is developed using TCAD-generated data and used as a surrogate model for differential evolution optimization. It can inversely design an out-of-the-training-range structure with BV as high as 1887V (89% of the ideal case) compared to ~1100V designed with human domain expertise.