Convolutional Machine Learning Method for Accelerating Nonequilibrium Green’s Function Simulations in Nanosheet Transistor
Preslav Aleksandrov, Ali Rezaei, Tapas Dutta, Nikolas Xeni, Asen Asenov, Vihar Georgiev
Abstract
This work describes a novel simulation approach that combines machine learning (ML) and device modeling simulations. The device simulations are based on the quantum mechanical nonequilibrium Green’s function (NEGF) approach, and the ML method is an extension of a convolutional generative network. We have named our new simulation approach ML-NEGF. It is implemented in our in-house simulator called Nano-Electronics Simulation Software (NESS). The reported results demonstrate the improved convergence speed of the ML-NEGF method in comparison to the “standard” NEGF approach. The trained ML model effectively learns the underlying physics of nano-sheet transistor behavior, resulting in faster convergence of the coupled Poisson-NEGF self-consistency simulations. Quantitatively, our ML-NEGF approach achieves an average convergence speedup of 60%, substantially reducing the computational time while maintaining the same accuracy.