Physics-Informed Neural Network for Predicting Out-of-Training-Range TCAD Solution with Minimized Domain Expertise

TL;DR

This paper introduces a physics-informed neural network that predicts TCAD solutions beyond training data range with minimal domain expertise, demonstrated on silicon nanowires, and does not require internal solver access or human-coded equations.

Contribution

It presents a novel PINN approach that predicts out-of-range TCAD solutions without internal solver access or extensive domain knowledge, and simplifies extension to complex systems.

Findings

Predicts 2.5 times larger range than training data

Successfully predicts inversion regions from subthreshold data

Does not require human-coded equations for physics-informed training

Abstract

Machine learning (ML) is promising in assisting technology computer-aided design (TCAD) simulations to alleviate difficulty in convergence and prolonged simulation time. While ML is widely used in TCAD, they either require access to the internal solver, require extensive domain expertise, are only trained by terminal quantities such as currents and voltages, and/or lack out-of-training-range prediction capability. In this paper, using Si nanowire as an example, we demonstrate that it is possible to use a physics-informed neural network (PINN) to predict out-of-training-range TCAD solutions without accessing the internal solver and with minimal domain expertise. The machine not only can predict a 2.5 times larger range than the training but also can predict the inversion region by only being trained with subthreshold region data. The physics-informed module is also trained with data without the need for human-coded equations making this easier to be extended to more sophisticated systems.