Technology Category
Machine Learning: Calibration & Uncertainty QuantificationKnowledge Representation and Reasoning: Diagnosis and Abductive ReasoningSearch and Optimization: Adversarial Search
Ga$_2$O$_3$ TCAD Mobility Parameter Calibration using Simulation Augmented Machine Learning with Physics Informed Neural Network
🤖 AI Summary
Conventional TCAD modeling of Ga₂O₃ Schottky barrier diodes suffers from manual calibration of critical parameters (e.g., mobility) and poor fitting in the turn-on region. Method: This work proposes a fully automated, physics-informed calibration framework based on a novel autoencoder–physics-informed neural network (AE-PINN) architecture. Leveraging only TCAD-simulated I–V data, it enables end-to-end inversion of seven key parameters—including PhuMob mobility model coefficients, effective anode work function, and temperature—without human prior knowledge. The PINN explicitly enforces carrier transport physics constraints. Contribution/Results: The method improves turn-on-state modeling accuracy significantly: full-range I–V curve fitting error is reduced by 62%; all inverted parameters exhibit relative errors <8.5%; subthreshold-region accuracy matches expert-level calibration; and measured and simulated I–V characteristics show excellent agreement.
Technology Category
Application Category
📝 Abstract
In this paper, we demonstrate the possibility of performing automatic Technology Computer-Aided-Design (TCAD) parameter calibration using machine learning, verified with experimental data. The machine only needs to be trained by TCAD data. Schottky Barrier Diode (SBD) fabricated with emerging ultra-wide-bandgap material, Gallium Oxide (Ga$_2$O$_3$), is measured and its current-voltage (IV) is used for Ga$_2$O$_3$ Philips Unified Mobility (PhuMob) model parameters, effective anode workfunction, and ambient temperature extraction (7 parameters). A machine comprised of an autoencoder (AE) and a neural network (NN) (AE-NN) is used. Ga$_2$O$_3$ PhuMob parameters are extracted from the noisy experimental curves. TCAD simulation with the extracted parameters shows that the quality of the parameters is as good as an expert's calibration at the pre-turned-on regime but not in the on-state regime. By using a simple physics-informed neural network (PINN) (AE-PINN), the machine performs as well as the human expert in all regimes.
Problem
Research questions and friction points this paper is trying to address.
Calibrate Ga2O3 TCAD mobility parameters using machine learning
Extract Schottky Barrier Diode parameters from noisy experimental data
Improve parameter accuracy in all regimes with physics-informed neural networks
Innovation
Methods, ideas, or system contributions that make the work stand out.
Machine learning for TCAD parameter calibration
Autoencoder and neural network combined
Physics-informed neural network enhances accuracy
🔎 Similar Papers
No similar papers found.
