Physics-Guided Machine Learning Assisted Characteristic Prediction of p-GaN Gate HEMTs

In this study, we demonstrate the feasibility of predicting and optimizing GaN-based high-electron-mobility field-effect transistors (GaN HEMTs) devices using the physics-guided machine learning (PGML) method. This paper illustrates a physics-guided artificial neural network (PG-ANN) consisting of three networks: Para-net, Vol-net, and G-net, which are trained on a dataset produced by Technology Computer-Aided Design (TCAD) simulations. Our method highlight the importance of first-order derivative characteristics \((\partial {\text{I}}/\partial {\text{V}})\) correlated with physical parameters for enhanced accuracy of IV curve predictions and uses a physics-based loss function to guide the PG-ANN towards accurate solutions. The PGML method is a promising numerical methodology to assist the modeling framework in Design Technology Co-Optimization (DTCO) flow.