Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight

[EN]Self-heating significantly impacts the performance and reliability of GaN high electron mobility transistors, but capturing these effects with electrothermal Monte Carlo (MC) simulations is computationally intensive. This paper presents the application of a hybrid AI-thermal model, previously te...

Descripción completa

Detalles Bibliográficos
Autores: Sergio, García Sánchez, Íñiguez de la Torre Mulas, Ignacio, Mateos López, Javier, González Sánchez, Tomás
Tipo de recurso: artículo
Estado:Versión enviada para evaluación y publicación
Fecha de publicación:2025
País:España
Institución:Universidad de Salamanca (USAL)
Repositorio:GREDOS. Repositorio Institucional de la Universidad de Salamanca
OAI Identifier:oai:gredos.usal.es:10366/168892
Acceso en línea:http://hdl.handle.net/10366/168892
Access Level:acceso abierto
Palabra clave:Electronic transport
Two-dimensional electron gas
Semiconductors
Field effect transistors
Heterostructures
Deep learning
Artificial neural networks
Monte Carlo methods
id ES_4dbc9b8f09eb0cce45f1cbcd09f0588e
oai_identifier_str oai:gredos.usal.es:10366/168892
network_acronym_str ES
network_name_str España
repository_id_str
spelling Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insightSergio, García SánchezÍñiguez de la Torre Mulas, IgnacioMateos López, JavierGonzález Sánchez, TomásElectronic transportTwo-dimensional electron gasSemiconductorsField effect transistorsHeterostructuresDeep learningArtificial neural networksMonte Carlo methods[EN]Self-heating significantly impacts the performance and reliability of GaN high electron mobility transistors, but capturing these effects with electrothermal Monte Carlo (MC) simulations is computationally intensive. This paper presents the application of a hybrid AI-thermal model, previously tested on another device, to the electrothermal analysis of GaN HEMTs. This first component consists of an artificial neural network (ANN) trained on isothermal MC data to predict drain current and lattice temperature. To extend the framework, a set of ANN-based microscopic models is introduced, composed of three dedicated networks that reconstruct spatially resolved quantities—electric field, carrier velocity, and sheet electron density. The system is coupled with compact thermal resistance models and iterated until convergence. The proposed approach achieves excellent agreement with electrothermal MC simulations while reducing computation time by approximately an order of magnitude. In addition to global performance metrics, it provides detailed internal profiles under electrothermally consistent conditions, making it a practical tool for fast device evaluation, in-depth analysis, and integration into compact modeling flows.This work has been partially supported through Grant Nos. PID2023-147555OB-I00 and PDC2023-145896-I00 funded by MCIN/AEI/10.13039/501100011033 and the Junta de Castilla y León and Fondo Europeo de Desarrollo Regional (FEDER) through Project No. SA136P23. This research has made use of the high performance computing resources of the Castilla y León Supercomputing Center (SCAYLE, www.scayle.es), financed by the European Regional Development Fund (ERDF).AIP Publishing202620262025info:eu-repo/semantics/articleinfo:eu-repo/semantics/submittedVersionapplication/pdfhttp://hdl.handle.net/10366/168892reponame:GREDOS. Repositorio Institucional de la Universidad de Salamancainstname:Universidad de Salamanca (USAL)Inglésinfo:eu-repo/semantics/openAccessoai:gredos.usal.es:10366/1688922026-06-07T06:28:51Z
dc.title.none.fl_str_mv Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
title Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
spellingShingle Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
Sergio, García Sánchez
Electronic transport
Two-dimensional electron gas
Semiconductors
Field effect transistors
Heterostructures
Deep learning
Artificial neural networks
Monte Carlo methods
title_short Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
title_full Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
title_fullStr Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
title_full_unstemmed Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
title_sort Electrothermal modeling of GaN high electron mobility transistors using a Monte Carlo-trained hybrid AI-thermal approach with microscopic physical insight
dc.creator.none.fl_str_mv Sergio, García Sánchez
Íñiguez de la Torre Mulas, Ignacio
Mateos López, Javier
González Sánchez, Tomás
author Sergio, García Sánchez
author_facet Sergio, García Sánchez
Íñiguez de la Torre Mulas, Ignacio
Mateos López, Javier
González Sánchez, Tomás
author_role author
author2 Íñiguez de la Torre Mulas, Ignacio
Mateos López, Javier
González Sánchez, Tomás
author2_role author
author
author
dc.subject.none.fl_str_mv Electronic transport
Two-dimensional electron gas
Semiconductors
Field effect transistors
Heterostructures
Deep learning
Artificial neural networks
Monte Carlo methods
topic Electronic transport
Two-dimensional electron gas
Semiconductors
Field effect transistors
Heterostructures
Deep learning
Artificial neural networks
Monte Carlo methods
description [EN]Self-heating significantly impacts the performance and reliability of GaN high electron mobility transistors, but capturing these effects with electrothermal Monte Carlo (MC) simulations is computationally intensive. This paper presents the application of a hybrid AI-thermal model, previously tested on another device, to the electrothermal analysis of GaN HEMTs. This first component consists of an artificial neural network (ANN) trained on isothermal MC data to predict drain current and lattice temperature. To extend the framework, a set of ANN-based microscopic models is introduced, composed of three dedicated networks that reconstruct spatially resolved quantities—electric field, carrier velocity, and sheet electron density. The system is coupled with compact thermal resistance models and iterated until convergence. The proposed approach achieves excellent agreement with electrothermal MC simulations while reducing computation time by approximately an order of magnitude. In addition to global performance metrics, it provides detailed internal profiles under electrothermally consistent conditions, making it a practical tool for fast device evaluation, in-depth analysis, and integration into compact modeling flows.
publishDate 2025
dc.date.none.fl_str_mv 2025
2026
2026
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/submittedVersion
format article
status_str submittedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10366/168892
url http://hdl.handle.net/10366/168892
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv AIP Publishing
publisher.none.fl_str_mv AIP Publishing
dc.source.none.fl_str_mv reponame:GREDOS. Repositorio Institucional de la Universidad de Salamanca
instname:Universidad de Salamanca (USAL)
instname_str Universidad de Salamanca (USAL)
reponame_str GREDOS. Repositorio Institucional de la Universidad de Salamanca
collection GREDOS. Repositorio Institucional de la Universidad de Salamanca
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869407711366479872
score 15,812455

Similares en LA Referencia