Revolutionizing TCAD Simulations with Universal Device Encoding and Graph Attention Networks

TL;DR

This paper introduces a novel AI-driven approach for semiconductor device simulation using graph attention networks and a universal encoding scheme that captures material, device, and spatial relationships, improving modeling accuracy.

Contribution

It presents a universal device encoding scheme combined with a graph attention network, RelGAT, for more accurate and comprehensive TCAD device simulations.

Findings

Effective surrogate Poisson emulation achieved

Accurate current-voltage prediction demonstrated

Device simulation process enhanced

Abstract

An innovative methodology that leverages artificial intelligence (AI) and graph representation for semiconductor device encoding in TCAD device simulation is proposed. A graph-based universal encoding scheme is presented that not only considers material-level and device-level embeddings, but also introduces a novel spatial relationship embedding inspired by interpolation operations typically used in finite element meshing. Universal physical laws from device simulations are leveraged for comprehensive data-driven modeling, which encompasses surrogate Poisson emulation and current-voltage (IV) prediction based on drift-diffusion model. Both are achieved using a novel graph attention network, referred to as RelGAT. Comprehensive technical details based on the device simulator Sentaurus TCAD are presented, empowering researchers to adopt the proposed AI-driven Electronic Design Automation (EDA) solution at the device level.