FusionWiki - User contributions [en]

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2026-10-27T14:45:17Z

Gratta: Add Alejandro González Ganzábal doctoral thesis in progress to associated training; remove personal CVN references.

LNF: AID-DAEDALUS. : AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics (2025-2028)

2026-10-27T14:45:08Z

Gratta: Añadir tesis doctoral en curso de Alejandro González Ganzábal; retirar menciones al CVN personal; corregir referencia a la sección de reconocimientos.

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2026-10-27T14:34:58Z

Gratta: Rectification: remove all publications and conference contributions predating the network start on 1 May 2025; retain dissemination within the project period.

LNF: AID-DAEDALUS. : AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics (2025-2028)

2026-10-27T14:33:11Z

Gratta: Rectificación: retirar trabajos anteriores al 01/09/2025 y trabajos JET/W7-X sin vinculación acreditada con AID-DAEDALUS; limitar ficha a TJ-II/OLMAT durante el proyecto.

LNF: AID-DAEDALUS. : AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics (2025-2028)

2026-10-27T14:29:19Z

Gratta: Actualización a 27/10/2026: corregir desglose económico AEI; precisar periodo y alcance; añadir publicaciones, preprint OLMAT, congresos, formación y colaboración, distinguiendo antecedentes y trabajos complementarios.

== LNF - Nationally funded project ==

[[File:Logos_MinFederAEI.jpg|thumb|right|300px|Ayuda PID2024-157659OB-I00 financiada por MICIU/AEI/10.13039/501100011033 y por FEDER Una manera de hacer Europa.]]

'''Title''': '''AID-DAEDALUS. AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics'''

'''Reference''': PID2024-157659OB-I00

'''Programme and date''': Plan Estatal de Investigación Científica, Técnica y de Innovación 2024-2027

'''Programme type (Modalidad de proyecto)''': Proyectos de Generación de Conocimiento 2024

'''Area/subarea (Área temática / subárea)''': Energía y transporte / Energía

'''Principal Investigator''': [https://orcid.org/0000-0002-5676-9631 Giuseppe Rattá Gutiérrez]

'''Beneficiary institution''': CIEMAT, Laboratorio Nacional de Fusión, Madrid, España

'''Project type''': Proyecto individual

'''Start date and duration''': 01/09/2025; 3 años (36 meses), septiembre de 2025-agosto de 2028.<ref name="AEI">Agencia Estatal de Investigación. [https://www.aei.gob.es/sites/default/files/convocatory_info/file/2025-12/PID2024-RC-Texto%2BAnexos-firmado.pdf Resolución de concesión de Proyectos de Generación de Conocimiento 2024], 5 de diciembre de 2025. Anexo I, entrada PID2024-157659OB-I00; anexo II, desglose económico.</ref>

'''Financing granted''': '''65.000 € en total''': 52.000 € de costes directos y 13.000 € de costes indirectos.<ref name="AEI" />

== Description of the project ==

AID-DAEDALUS desarrolla prototipos de gemelos digitales basados en inteligencia artificial para el stellarator TJ-II y la instalación OLMAT del CIEMAT. Su objetivo es construir modelos que permitan explorar escenarios experimentales, estimar señales de diagnóstico y apoyar la preparación de experimentos. El trabajo conecta el modelado del plasma de TJ-II con el análisis de los resultados de exposición de materiales en OLMAT.

El programa combina preparación de datos, aprendizaje automático y validación frente a medidas experimentales. El uso de computación de altas prestaciones, incluido Turgalium, está previsto para el entrenamiento y la evaluación de modelos. La aplicación en tiempo real es un objetivo de investigación que requiere validación específica; las publicaciones relacionadas aportan componentes y pruebas de principio hacia los prototipos.<ref name="AEIabstract">Agencia Estatal de Investigación. [https://www.aei.gob.es/gl/axudas-concedidas/buscador-axudas-concedidas?area=&ccaa=All&cif=&code=&convocatory=All&gender=All&granted%5Bmax%5D=&granted%5Bmin%5D=&keywords=&name=&order=amount&page=3055&province=All&sort=asc&subarea=&summary=&tit=&title=&year=All Ficha y resumen del proyecto PID2024-157659OB-I00].</ref>

=== Ámbitos de trabajo ===
* '''Datos de TJ-II:''' preparación, sincronización y curación de señales y parámetros para entrenar y contrastar modelos.
* '''Modelos predictivos y sustitutos:''' estimación de densidad, emisión de impurezas y corriente toroidal; desarrollo hacia la evaluación de escenarios antes de ejecutar la descarga.
* '''OLMAT:''' métodos de IA y análisis de imágenes para caracterizar los resultados experimentales de exposición de materiales, incluidos cráteres y grietas en tungsteno.
* '''Validación y formación:''' evaluación de los dominios de validez de los modelos, formación de investigadores y cooperación con grupos de física, diagnóstico e inteligencia artificial.

== Main results and activities: progress to 27 October 2026 ==

El proyecto está en ejecución. Este apartado recoge avances y trabajos relacionados documentados hasta el 27 de octubre de 2026; el balance final corresponderá al informe de cierre.

=== Datos y modelos para TJ-II ===
* '''Base de datos curada.''' La publicación sobre señales y diagnósticos de TJ-II presenta una selección de 4.800 descargas, con señales sincronizadas a 1 ms. Es la base de datos descrita en ese artículo, y debe distinguirse del archivo histórico completo de TJ-II.<ref name="DB">A. González Ganzábal et al. [https://doi.org/10.1016/j.fusengdes.2025.115613 A comprehensive database of TJ-II signals and diagnostics for statistically based models]. Fusion Engineering and Design 224, 115613 (2026).</ref>
* '''Densidad del plasma.''' Un trabajo conjunto con P. Aguayo, G. Farías, A. González Ganzábal y otros colaboradores estudia la estimación de densidad en TJ-II a partir de datos experimentales; se incluye la referencia completa en la bibliografía.
* '''Estimación de señales.''' El estudio con A. Cureño-Ramírez, A. González-Ganzábal, A. Pardo y J. Calle-Cancho compara SVR, ESN, LSTM y AE-LSTM para estimar la evolución de la señal de emisión de impurezas ACTON275. El artículo reconoce explícitamente la ayuda PID2024-157659OB-I00. Constituye una prueba de principio para componentes predictivos de futuros gemelos digitales.<ref name="Signals">A. Cureño-Ramírez et al. [https://doi.org/10.1016/j.fusengdes.2026.115847 Data-Driven estimation of experimental signals in TJ-II: A proof of principle]. Fusion Engineering and Design 230, 115847 (2026), incluidos sus agradecimientos.</ref>
* '''Corriente toroidal.''' El estudio de 2026 sobre descargas TJ-II programadas con la bobina OH compara estimadores basados en programación genética y redes recurrentes. La contribución SOFT 2026 extiende la difusión hacia modelos sustitutos para la preparación del pulso. Las referencias se detallan abajo.

=== OLMAT: cuantificación del daño en tungsteno ===
El manuscrito de A. Orbañanos Fernandez, G. Rattá y colaboradores desarrolla un flujo automatizado de análisis de imágenes de microscopía electrónica de muestras expuestas en OLMAT. Combina técnicas de visión artificial y segmentación mediante aprendizaje profundo para cuantificar daño superficial, cráteres y grietas. Es una contribución a la caracterización experimental; no demuestra por sí misma un gemelo termomecánico completo ni un sistema de control en tiempo real. El texto está disponible como preprint en SSRN desde el 29 de septiembre de 2026 y se identifica como manuscrito enviado, sin aceptación de revista documentada a esta fecha.<ref name="OLMAT">A. Orbañanos Fernandez et al. [https://doi.org/10.2139/ssrn.7540157 A first deep learning framework for the post-exposure quantification of high heat flux damage in tungsten at OLMAT]. SSRN, preprint depositado el 29 de septiembre de 2026.</ref>

=== Formación asociada ===
El CVN del investigador principal recoge los siguientes trabajos de fin de máster de la Universidad Autónoma de Madrid, Máster en Energías y Combustibles para el Futuro, curso 2025/2026:
* ''Desarrollo de un gemelo digital basado en inteligencia artificial para el stellarator TJ-II'', dirigido por G. Rattá.
* ''Modelado basado en inteligencia artificial para reproducir los resultados experimentales de OLMAT'', codirigido con Daniel Alegre Castro.
* ''Modelado basado en inteligencia artificial para reproducir los resultados experimentales de OLMAT: cráteres y grietas'', codirigido con Daniel Alegre Castro.

Como antecedente formativo, el CVN registra también el TFM de Víctor Ángel Fuentes Blas, ''Estimación de Señales del Plasma Mediante Técnicas de Inteligencia Artificial'', UAM, curso 2024/2025, y la investigación de máster de Paulo Alejandro Aguayo Vallejos con Gonzalo Farías y G. Rattá en PUCV, ''Estimación de señales de fusión nuclear, aplicación en TJ-II''.

== Dissemination: publications and conference contributions ==

Las referencias siguientes documentan los trabajos sobre TJ-II y OLMAT y las colaboraciones metodológicas indicadas en el CVN con A. González Ganzábal, A. Cureño-Ramírez, G. Farías, A. Pardo y J. M. Calle Cancho. Se distinguen los trabajos directamente relacionados con los dispositivos del proyecto, los complementarios sobre JET/W7-X y los antecedentes anteriores al inicio de la ayuda. La coautoría o la fecha de publicación no acreditan por sí solas financiación por AID-DAEDALUS; cuando se ha comprobado el reconocimiento explícito de la ayuda se indica en el texto.

=== Publicaciones sobre TJ-II ===
# G. A. Rattá, B. Ph. van Milligen, V. Á. Fuentes Blas, M. Samper, M. Jurado, A. González-Ganzábal and The TJ-II Curated Database Team. '''Data-Driven Estimation of Net Toroidal Plasma Current Waveforms in OH-Programmed TJ-II Discharges.''' ''Plasma'' '''9'''(3), 33 (2026). [https://doi.org/10.3390/plasma9030033 DOI].
# A. Cureño-Ramírez, L. G. de la Fraga, G. A. Rattá, A. González-Ganzábal, A. Baciero, B. López-Miranda, B. van Milligen, A. Pardo, J. Calle-Cancho and TJ-II Team. '''Data-Driven estimation of experimental signals in TJ-II: A proof of principle.''' ''Fusion Engineering and Design'' '''230''', 115847 (2026). [https://doi.org/10.1016/j.fusengdes.2026.115847 DOI].
# A. González Ganzábal, G. A. Rattá, T. Estrada, J. Martínez-Fernández, N. Panadero, Á. Cappa, B. López-Miranda, A. Baciero, F. Martín, D. Tafalla, B. P. van Milligen, F. Medina, Á. de la Peña and S. Dormido-Canto. '''A comprehensive database of TJ-II signals and diagnostics for statistically based models.''' ''Fusion Engineering and Design'' '''224''', 115613 (March 2026). [https://doi.org/10.1016/j.fusengdes.2025.115613 DOI].
# P. A. Aguayo Vallejos, G. Farías, A. González Ganzábal, E. Fabregas, T. Estrada, B. P. van Milligen, A. Baciero, B. López-Miranda, F. Medina and G. A. Rattá. '''A data-driven approach to estimate plasma density in TJ-II stellarator.''' ''Fusion Engineering and Design'' '''224''', 115596 (March 2026). [https://doi.org/10.1016/j.fusengdes.2025.115596 DOI].

=== Preprint y manuscrito enviado sobre OLMAT ===
* A. Orbañanos Fernandez, G. Rattá, D. Alegre, I. Rivera, P. Fernández-Mayo, D. Tafalla, A. de Castro, L. Serrador, R. Dejarnac, J. Matejicek, E. Oyarzábal y OLMAT Team. '''A first deep learning framework for the post-exposure quantification of high heat flux damage in tungsten at OLMAT.''' SSRN, 29 de septiembre de 2026, 22 páginas. [https://doi.org/10.2139/ssrn.7540157 DOI del preprint]. Manuscrito enviado; no se presenta como artículo de revista aceptado o publicado.

=== Contribuciones a congresos sobre TJ-II ===
# G. A. Rattá and collaborators, including J. M. Calle Cancho, G. Farías and P. Aguayo in the institutional dissemination record. '''Towards AI-based surrogate models for pulse design. Pre-pulse prediction on the TJ-II stellarator.''' Oral contribution, 34th Symposium on Fusion Technology (SOFT 2026), Aix-en-Provence, France, 21-25 September 2026. [https://www.fusion.ciemat.es/lnf/el-laboratorio-nacional-de-fusion-del-ciemat-muestra-su-actividad-cientifica-en-el-34-o-congreso-de-tecnologia-de-fusion/ CIEMAT report].
# A. González Ganzábal, G. A. Rattá, T. Estrada, J. Martínez-Fernández, N. Panadero, Á. Cappa, B. López-Miranda, A. Baciero, F. Martín, D. Tafalla, B. P. van Milligen, F. Medina, Á. de la Peña, S. Dormido-Canto and TJ-II Team. '''Development of a standardised TJ-II database for Machine Learning models and applications.''' Poster, 25th International Stellarator-Heliotron Workshop (ISHW 2026), Córdoba, Spain, 20-24 April 2026.

=== Trabajos metodológicos complementarios: JET y W7-X ===
Los siguientes trabajos comparten investigadores y métodos de análisis de datos, pero sus dispositivos y objetivos se distinguen de los prototipos TJ-II/OLMAT.
# G. A. Rattá, A. González, WPTE Team and JET Contributors. '''Post-alarm time-to-disruption classification in disruptive JET discharges: comparison of SVM, LSTM, and transformer models.''' ''Plasma Physics and Controlled Fusion'' '''68''', 055035 (2026). [https://doi.org/10.1088/1361-6587/ae6945 DOI].
# A. González Ganzábal, G. A. Rattá, S. Dormido-Canto, P. Carvalho, S. Silburn, I. Coffey and I. Balboa; JET/WPTE collaboration signatures recorded in the CVN. '''A methodology for TIE detection and tracking for JET's experimental cameras.''' ''Fusion Engineering and Design'' '''218''', 115164 (2025). [https://doi.org/10.1016/j.fusengdes.2025.115164 DOI].

==== Manuscrito de DIONE ====
* A. González Ganzábal, E. Maragkoudakis, D. Carralero, T. Estrada, T. Windisch, S. Dormido-Canto, G. A. Rattá and W7-X Team. '''DIONE: Automating the Analysis of W7-X Doppler Reflectometry Data with neural networks.''' Submitted to ''Nuclear Fusion''; acceptance pending according to the CVN dated 27 October 2026. This manuscript is distinct from the registered DIONE software.

==== Difusión de DIONE en congreso ====
# A. González Ganzábal, E. Maragkoudakis, D. Carralero, T. Estrada, T. Windisch, S. Dormido-Canto, G. A. Rattá and W7-X Team. '''DIONE: Automating the analysis of W7-X Doppler Reflectometry data with neural networks.''' FuseNet PhD Event 2025, ITER, Saint-Paul-lès-Durance, France, 4-6 November 2025; presented by A. González Ganzábal. [https://indico.fusenet.eu/event/57/overview Event website].

==== Software DIONE ====
DIONE (''Doppler peak IdentificatiOn using NEural networks'') fue desarrollado por Daniel Carralero, Alejandro González Ganzábal y Giuseppe Rattá para automatizar el análisis de reflectometría Doppler en W7-X. El CVN registra el software con el número 16/2026/3141, de 27 de abril de 2026; CIEMAT difundió su registro el 21 de mayo. Se incluye como transferencia metodológica complementaria de los colaboradores, separada del manuscrito y del trabajo específico sobre TJ-II/OLMAT.<ref name="DIONE">CIEMAT. [https://www.ciemat.es/w/el-ciemat-registra-dione-un-software-basado-en-inteligencia-artificial-para-acelerar-el-an%C3%A1lisis-de-plasmas-de-fusi%C3%B3n El CIEMAT registra DIONE, un software basado en inteligencia artificial para acelerar el análisis de plasmas de fusión], 21 de mayo de 2026.</ref>

=== Antecedentes: publicaciones anteriores al proyecto ===
# A. González Ganzábal, G. A. Rattá, D. Gadariya, S. Dormido-Canto and JET Contributors. '''Advancing MARFE detection in JET's operational camera videos through Machine Learning techniques.''' ''Fusion Engineering and Design'' '''205''', 114534 (2024). [https://doi.org/10.1016/j.fusengdes.2024.114534 DOI].
# G. A. Rattá, J. Vega, A. Murari, D. Gadariya, C. Stuart and G. Farías. '''Characterization of physics events in JET preceding disruptions.''' ''Fusion Engineering and Design'' '''189''', 113468 (2023). [https://doi.org/10.1016/j.fusengdes.2023.113468 DOI].
# G. Farías, S. Dormido-Canto, J. Vega, G. Rattá and collaborators. '''Automatic feature extraction in large fusion databases by using deep learning approach.''' ''Fusion Engineering and Design'' '''112''', 979-983 (2016). [https://doi.org/10.1016/j.fusengdes.2016.06.016 DOI].
# G. A. Rattá, J. Vega, A. Pereira, A. Portas, E. de la Luna, S. Dormido-Canto, G. Farías, R. Dormido, J. Sánchez, N. Duro, H. Vargas, M. Santos, G. Pajares and A. Murari; JET-EFDA Contributors. '''First applications of structural pattern recognition methods to the investigation of specific physical phenomena at JET.''' ''Fusion Engineering and Design'' '''83'''(2-3), 467-470 (April 2008). [https://doi.org/10.1016/j.fusengdes.2007.09.008 DOI].
# S. Dormido-Canto, G. Farías, R. Dormido, J. Vega, J. Sánchez, N. Duro, H. Vargas, G. Rattá, A. Pereira and A. Portas. '''Structural pattern recognition methods based on string comparison for fusion databases.''' ''Fusion Engineering and Design'' '''83'''(2-3), 421-424 (2008). [https://doi.org/10.1016/j.fusengdes.2007.11.009 DOI].
# J. Vega, G. Rattá, A. Murari, P. Castro, S. Dormido-Canto, R. Dormido, G. Farías, A. Pereira, A. Portas, E. de la Luna, I. Pastor, J. Sánchez, N. Duro, R. Castro, M. Santos and H. Vargas. '''Recent results on structural pattern recognition for Fusion massive databases.''' Proceedings of IEEE WISP 2007 (conference proceedings, not a journal article). [https://doi.org/10.1109/WISP.2007.4447569 DOI].

==== Informe técnico ====
* A. González-Ganzábal, G. A. Rattá, D. Gadariya and S. Dormido. '''Herramientas para la detección de MARFE en JET con cámaras operacionales.''' ''Informes Técnicos CIEMAT'' 1533, 1-32 (2024). This is a technical report, listed separately from peer-reviewed articles.

==== Congresos y reuniones anteriores al proyecto ====
Estos trabajos documentan la trayectoria de colaboración y los antecedentes metodológicos, con anterioridad al inicio de AID-DAEDALUS el 1 de septiembre de 2025.
# R. S. Granetz, G. A. Rattá, A. González and collaborators. '''Update on the MDC-JA-4 High-Z UFO disruptions.''' 44th ITPA MHD, Disruptions and Control meeting, 22-25 October 2024.
# G. A. Rattá and A. González. '''Comparison of “Classic” and “State-of-the-Art” Machine Learning Techniques to Estimate the Time to Disruption.''' 44th ITPA MDC meeting, 22-25 October 2024.
# G. A. Rattá and A. González. '''A data-driven disruption classifier based on signal anomalies.''' SOFT 2024, Dublin, Ireland, 22-27 September 2024.
# A. González and G. A. Rattá. '''A methodology for TIE detection and tracking for JET's experimental cameras.''' SOFT 2024, Dublin, Ireland, 22-27 September 2024.
# G. A. Rattá and A. González. '''Identification of physics events for the characterization of disruptions in JET.''' SOFT 2024, Dublin, Ireland, 22-27 September 2024.
# G. A. Rattá and A. González. '''Estimating Time to Disruption Using Transformer Models.''' Third IAEA Technical Meeting on Plasma Disruptions and their Mitigation, ITER, Saint-Paul-lès-Durance, France, 3-6 September 2024. The CVN distinguishes this slide title from the programme title, ''Time-to-Disruption Estimation Using LSTM Networks''. [https://conferences.iaea.org/event/380/contributions/32299/ Contribution record].
# R. S. Granetz, G. A. Rattá, A. González and collaborators. '''MDC-JA-4 High-Z UFO disruptions.''' 43rd ITPA MDC meeting, 22-26 April 2024.
# G. Farías, S. Dormido-Canto, J. Vega and G. Rattá. '''Automatic feature extraction in large fusion databases by using deep learning approach.''' 10th IAEA Technical Meeting on Control, Data Acquisition and Remote Participation for Fusion Research, Ahmedabad, India, 20-24 April 2015.
# J. Vega, A. Murari, S. Dormido-Canto, D. Alves, G. Farías, J. M. López, R. Moreno, A. Pereira, J. M. Martínez and G. Rattá. '''Overview of Real-Time Disruption Prediction in JET: Applicability to ITER.''' 41st EPS Conference on Plasma Physics, Berlin, Germany, 23-27 June 2014.
# J. Vega, A. Murari, G. A. Rattá, G. Farías and collaborators. '''An overview about recent developments on advanced data analysis techniques in fusion.''' 6th Workshop on Fusion Data Processing, Validation and Analysis, Madrid, Spain, January 2010.
# J. Vega, G. Rattá, A. Murari, G. Farías and collaborators. '''Recent results on structural pattern recognition for Fusion massive databases.''' IEEE WISP, Alcalá de Henares, Spain, 3-5 October 2007. This is the presentation associated with the proceedings paper listed above.
# S. Dormido-Canto, G. Farías, R. Dormido, J. Sánchez, N. Duro, H. Vargas, J. Vega, G. Rattá, A. Pereira and A. Portas. '''Comparison of structural pattern recognition methods for fusion databases.''' 6th IAEA Technical Meeting on Control, Data Acquisition and Remote Participation for Fusion Research, Inuyama, Japan, June 2007.
# G. A. Rattá, J. Vega, A. Pereira, A. Portas, E. de la Luna, S. Dormido-Canto, G. Farías and collaborators. '''First applications of structural pattern recognition methods to the investigation of specific physical phenomena at JET.''' 6th IAEA Technical Meeting on Control, Data Acquisition and Remote Participation for Fusion Research, Inuyama, Japan, June 2007.

== International collaboration, seminars and outreach ==

=== Actividades durante el proyecto ===
* '''3 de septiembre de 2025, IEE-UNSJ/CONICET, San Juan, Argentina.''' Seminario de G. Rattá: ''STRANAI y AID-DAEDALUS. Investigación colaborativa en IA y fusión nuclear para impulsar la energía del futuro''. [https://www.unsj.edu.ar/home/noticias_detalles/9609/2 Noticia de la UNSJ].
* '''16 de septiembre de 2025, CSC-CONICET, Buenos Aires, Argentina.''' Seminario ''Inteligencia artificial, fusión nuclear y supercomputación'', durante la visita de investigación del 12 al 16 de septiembre. Título y fechas según CVN.
* '''1 de febrero-30 de abril de 2026, CINVESTAV, México.''' Estancia de investigación en el Departamento de Computación, financiada por una ayuda de movilidad Senior, sobre IA para diagnósticos de plasma y estimación de señales. Se distingue la financiación de movilidad de la ayuda de proyecto.
* '''23 de marzo de 2026, CINVESTAV.''' Seminario de G. Rattá sobre IA para acelerar la investigación en fusión y la colaboración CIEMAT-CINVESTAV mediante AID-DAEDALUS y STRANAI. [https://www.fusion.ciemat.es/lnf/ia-para-acelerar-la-investigacion-en-fusion-nuclear-colaboracion-entre-ciemat-y-cinvestav-en-los-proyectos-aid-daedalus-y-stranai/ Noticia de CIEMAT, publicada el 28 de abril de 2026] y [https://www.youtube.com/watch?v=3EMjasIEkwA vídeo del seminario].
* '''11 de agosto-2 de septiembre de 2026, IEE-UNSJ/CONICET.''' Estancia de colaboración científica relacionada con AID-DAEDALUS y STRANAI, según CVN.
* '''3-15 de septiembre de 2026, PUCV, Chile.''' Colaboración y supervisión de investigación sobre predicción de saltos de franja en señales de interferometría, relacionada con AID-DAEDALUS y STRANAI, según CVN.

=== Actividad preparatoria y cobertura institucional ===
* '''19 de agosto de 2025, PUCV, Valparaíso, Chile.''' Seminario de G. Rattá: ''La red STRANAI y el proyecto AID-DAEDALUS: estableciendo sinergias entre fusión nuclear e Inteligencia Artificial''. Esta actividad es anterior al inicio formal de AID-DAEDALUS. [https://www.instagram.com/p/DNYx-GyPAfi/ Anuncio de PUCV].
* '''26 de septiembre de 2025.''' CIEMAT publicó una noticia de conjunto sobre las visitas a Chile y Argentina, que describe la cooperación con PUCV, IEE-UNSJ y CSC-CONICET. [https://www.fusion.ciemat.es/lnf/la-red-stranai-impulsa-nuevas-colaboraciones-en-latinoamerica-en-inteligencia-artificial-aplicada-a-fusion-nuclear/ Noticia institucional].
* La actividad de San Juan tuvo cobertura en [https://www.diariodecuyo.com.ar/columnasdeopinion/con-un-disertante-espanol-se-realizara-un-seminario-sobre-inteligencia-artificial-y-fusion-nuclear-en-la-unsj-1765893.html Diario de Cuyo], [https://www.canal13sanjuan.com/tecnologia/inteligencia-artificial-y-energia---un-futuro-comun-en-la-fusion-nuclear-_a68ba462895043f2db4888f4b Canal 13 San Juan] y [https://www.diarioelzondasj.com.ar/338855-un-seminario-en-san-juan-abordara-la-fusion-nuclear-y-la-inteligencia-artificial Diario El Zonda]. Son noticias sobre una misma actividad, no contribuciones científicas adicionales.

== Related projects and sources ==
* [https://www.ciemat.es/en/proyectos/-/actividades/proyecto/3587?_es_ciemat_publicaciones_web_ActividadesPortlet_project=3587 Ficha institucional de AID-DAEDALUS en CIEMAT].
* [[LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques|STRANAI (RED2024-153593-E)]], red de colaboración complementaria a la ayuda de proyecto AID-DAEDALUS.
* [https://sites.fusion.ciemat.es/stranai/articles/ Noticias de STRANAI], incluidas las actividades conjuntas de difusión.
* CVN de Giuseppe Rattá Gutiérrez, actualizado el 27 de octubre de 2026: proyectos, publicaciones, congresos, formación, estancias y propiedad intelectual.

== References ==
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2026-10-27T14:21:51Z

Gratta: Update STRANAI project data, publications, conferences and outreach to 27 October 2026; correct funding reference and end date; distinguish earlier work and submitted manuscript.

== LNF - Nationally funded project ==
* '''Title:''' STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques.
* '''Reference:''' RED2024-153593-E.
* '''Funding programme:''' Redes de Investigación, convocatoria 2024; strategic network (infrastructure / international organisation).
* '''Funding body:''' Agencia Estatal de Investigación (AEI).
* '''Coordinating institution:''' Laboratorio Nacional de Fusión, CIEMAT, Madrid, Spain.
* '''Coordinator:''' [https://orcid.org/0000-0002-5676-9631 Giuseppe Rattá Gutiérrez].
* '''Duration:''' 1 May 2025 - 30 April 2027 (24 months).
* '''Funding granted:''' EUR 65,000.
* '''Website:''' [https://sites.fusion.ciemat.es/stranai/ STRANAI].
* '''News:''' [https://sites.fusion.ciemat.es/stranai/articles/ STRANAI news and dissemination].

The AEI award resolution identifies STRANAI as RED2024-153593-E and awards EUR 65,000 to CIEMAT. [https://www.aei.gob.es/sites/default/files/convocatory_info/assistants/2025-09/RED2024-RC-Texto%2BAnexo-firmado.pdf Award resolution].

== Description of the project ==
STRANAI brings together Spanish teams working on artificial intelligence, experimental and theoretical plasma physics, and fusion technology. Coordinated by CIEMAT, the network promotes exchange of methods, data and expertise, strengthens cooperation with international partners, and supports Spanish participation in fusion research, including EUROfusion activities.

The eight participating groups are:
* Laboratorio Nacional de Fusión (LNF), CIEMAT, Madrid.
* Centro Extremeño de Tecnologías Avanzadas (CETA), CIEMAT, Cáceres.
* Plasma Science and Fusion Technology (PSFT), Universidad de Sevilla.
* Fusion Research Group, Barcelona Supercomputing Center (BSC).
* Departamento de Informática y Automática, Universidad Nacional de Educación a Distancia (UNED), Madrid.
* Consorcio IFMIF-DONES España, Granada.
* Service Robotics Lab, Universidad Pablo de Olavide (UPO), Seville.
* Centre for Automation and Robotics (CAR, CSIC-UPM), Group of Advanced Automation of Machines, Highly Complex Processes and Environments, Madrid.

=== Research lines ===
# [https://sites.fusion.ciemat.es/stranai/research-lines/lr1 Identification of physical events from images and videos].
# [https://sites.fusion.ciemat.es/stranai/research-lines/lr2 Automation of turbulence analysis using Doppler reflectometry].
# [https://sites.fusion.ciemat.es/stranai/research-lines/lr3 Explainable machine learning, physics-informed AI and equation extraction from AI models].
# [https://sites.fusion.ciemat.es/stranai/research-lines/lr4 Shared multi-machine databases and code repositories].
# [https://sites.fusion.ciemat.es/stranai/research-lines/lr5 Plasma control and operational limits].
# [https://sites.fusion.ciemat.es/stranai/research-lines/lr6 Predictive models for disruption mitigation, avoidance and prevention].

== Main results and activities: progress to 27 October 2026 ==
The network is ongoing. The following progress summary covers documented scientific outputs and collaboration activities; it is not the final project report.

* '''Diagnostic data and modelling in TJ-II.''' Collaborative publications address a curated diagnostic database, plasma-density estimation, experimental-signal estimation and net toroidal-current waveform estimation. These outputs provide methods and data for statistical modelling and experimental preparation (references below).
* '''Image and video analysis in JET.''' Work with Alejandro González Ganzábal addresses detection and tracking of transient impurity events (TIEs), building on earlier MARFE-detection methods.
* '''Disruption timing.''' A 2026 study compares SVM, LSTM and transformer models for post-alarm time-to-disruption classification in JET.
* '''DIONE software.''' Daniel Carralero, Alejandro González Ganzábal and Giuseppe Rattá developed DIONE (Doppler peak IdentificatiOn using NEural networks) to automate W7-X Doppler-reflectometry analysis. The CVN records intellectual-property registration 16/2026/3141, dated 27 April 2026. CIEMAT announced the registration on 21 May 2026. [https://www.ciemat.es/w/el-ciemat-registra-dione-un-software-basado-en-inteligencia-artificial-para-acelerar-el-an%C3%A1lisis-de-plasmas-de-fusi%C3%B3n CIEMAT announcement]. The associated Nuclear Fusion manuscript is listed separately as submitted, with acceptance pending in the CVN dated 27 October 2026.
* '''International cooperation.''' Visits and seminars in Chile and Argentina in August-September 2025 strengthened cooperation with PUCV and IEE-UNSJ/CONICET and opened exchanges with CSC-CONICET. A research stay at CINVESTAV, Mexico, from 1 February to 30 April 2026 included a seminar on STRANAI and AID-DAEDALUS. These activities connect AI methods with fusion diagnostics and predictive modelling.

== Dissemination: publications and conference contributions ==
This list documents collaborations involving G. A. Rattá and A. González Ganzábal, A. Cureño-Ramírez, G. Farías, A. Pardo or J. M. Calle Cancho, as recorded in the coordinator's CVN. Works predating 1 May 2025 are explicitly listed as scientific background. Inclusion here identifies scientific links with the network; it does not by itself establish that each work was funded by STRANAI.

=== Peer-reviewed publications during the network period ===
# G. A. Rattá, B. Ph. van Milligen, V. Á. Fuentes Blas, M. Samper, M. Jurado, A. González-Ganzábal and The TJ-II Curated Database Team. '''Data-Driven Estimation of Net Toroidal Plasma Current Waveforms in OH-Programmed TJ-II Discharges.''' ''Plasma'' '''9'''(3), 33 (2026). [https://doi.org/10.3390/plasma9030033 DOI].
# A. Cureño-Ramírez, L. G. de la Fraga, G. A. Rattá, A. González-Ganzábal, A. Baciero, B. López-Miranda, B. van Milligen, A. Pardo, J. Calle-Cancho and TJ-II Team. '''Data-Driven estimation of experimental signals in TJ-II: A proof of principle.''' ''Fusion Engineering and Design'' '''230''', 115847 (2026). [https://doi.org/10.1016/j.fusengdes.2026.115847 DOI].
# G. A. Rattá, A. González, WPTE Team and JET Contributors. '''Post-alarm time-to-disruption classification in disruptive JET discharges: comparison of SVM, LSTM, and transformer models.''' ''Plasma Physics and Controlled Fusion'' '''68''', 055035 (2026). [https://doi.org/10.1088/1361-6587/ae6945 DOI].
# A. González Ganzábal, G. A. Rattá, T. Estrada, J. Martínez-Fernández, N. Panadero, Á. Cappa, B. López-Miranda, A. Baciero, F. Martín, D. Tafalla, B. P. van Milligen, F. Medina, Á. de la Peña and S. Dormido-Canto. '''A comprehensive database of TJ-II signals and diagnostics for statistically based models.''' ''Fusion Engineering and Design'' '''224''', 115613 (March 2026). [https://doi.org/10.1016/j.fusengdes.2025.115613 DOI].
# P. A. Aguayo Vallejos, G. Farías, A. González Ganzábal, E. Fabregas, T. Estrada, B. P. van Milligen, A. Baciero, B. López-Miranda, F. Medina and G. A. Rattá. '''A data-driven approach to estimate plasma density in TJ-II stellarator.''' ''Fusion Engineering and Design'' '''224''', 115596 (March 2026). [https://doi.org/10.1016/j.fusengdes.2025.115596 DOI].
# A. González Ganzábal, G. A. Rattá, S. Dormido-Canto, P. Carvalho, S. Silburn, I. Coffey and I. Balboa; JET/WPTE collaboration signatures recorded in the CVN. '''A methodology for TIE detection and tracking for JET's experimental cameras.''' ''Fusion Engineering and Design'' '''218''', 115164 (2025). [https://doi.org/10.1016/j.fusengdes.2025.115164 DOI].

=== Submitted manuscript ===
* A. González Ganzábal, E. Maragkoudakis, D. Carralero, T. Estrada, T. Windisch, S. Dormido-Canto, G. A. Rattá and W7-X Team. '''DIONE: Automating the Analysis of W7-X Doppler Reflectometry Data with neural networks.''' Submitted to ''Nuclear Fusion''; acceptance pending according to the CVN dated 27 October 2026. This manuscript is distinct from the registered DIONE software.

=== Conference contributions during the network period ===
# G. A. Rattá and collaborators, including J. M. Calle Cancho, G. Farías and P. Aguayo in the institutional dissemination record. '''Towards AI-based surrogate models for pulse design. Pre-pulse prediction on the TJ-II stellarator.''' Oral contribution, 34th Symposium on Fusion Technology (SOFT 2026), Aix-en-Provence, France, 21-25 September 2026. [https://www.fusion.ciemat.es/lnf/el-laboratorio-nacional-de-fusion-del-ciemat-muestra-su-actividad-cientifica-en-el-34-o-congreso-de-tecnologia-de-fusion/ CIEMAT report].
# A. González Ganzábal, G. A. Rattá, T. Estrada, J. Martínez-Fernández, N. Panadero, Á. Cappa, B. López-Miranda, A. Baciero, F. Martín, D. Tafalla, B. P. van Milligen, F. Medina, Á. de la Peña, S. Dormido-Canto and TJ-II Team. '''Development of a standardised TJ-II database for Machine Learning models and applications.''' Poster, 25th International Stellarator-Heliotron Workshop (ISHW 2026), Córdoba, Spain, 20-24 April 2026.
# A. González Ganzábal, E. Maragkoudakis, D. Carralero, T. Estrada, T. Windisch, S. Dormido-Canto, G. A. Rattá and W7-X Team. '''DIONE: Automating the analysis of W7-X Doppler Reflectometry data with neural networks.''' FuseNet PhD Event 2025, ITER, Saint-Paul-lès-Durance, France, 4-6 November 2025; presented by A. González Ganzábal. [https://indico.fusenet.eu/event/57/overview Event website].

=== Scientific background: publications before the network period ===
# A. González Ganzábal, G. A. Rattá, D. Gadariya, S. Dormido-Canto and JET Contributors. '''Advancing MARFE detection in JET's operational camera videos through Machine Learning techniques.''' ''Fusion Engineering and Design'' '''205''', 114534 (2024). [https://doi.org/10.1016/j.fusengdes.2024.114534 DOI].
# G. A. Rattá, J. Vega, A. Murari, D. Gadariya, C. Stuart and G. Farías. '''Characterization of physics events in JET preceding disruptions.''' ''Fusion Engineering and Design'' '''189''', 113468 (2023). [https://doi.org/10.1016/j.fusengdes.2023.113468 DOI].
# G. Farías, S. Dormido-Canto, J. Vega, G. Rattá and collaborators. '''Automatic feature extraction in large fusion databases by using deep learning approach.''' ''Fusion Engineering and Design'' '''112''', 979-983 (2016). [https://doi.org/10.1016/j.fusengdes.2016.06.016 DOI].
# G. A. Rattá, J. Vega, A. Pereira, A. Portas, E. de la Luna, S. Dormido-Canto, G. Farías, R. Dormido, J. Sánchez, N. Duro, H. Vargas, M. Santos, G. Pajares and A. Murari; JET-EFDA Contributors. '''First applications of structural pattern recognition methods to the investigation of specific physical phenomena at JET.''' ''Fusion Engineering and Design'' '''83'''(2-3), 467-470 (April 2008). [https://doi.org/10.1016/j.fusengdes.2007.09.008 DOI].
# S. Dormido-Canto, G. Farías, R. Dormido, J. Vega, J. Sánchez, N. Duro, H. Vargas, G. Rattá, A. Pereira and A. Portas. '''Structural pattern recognition methods based on string comparison for fusion databases.''' ''Fusion Engineering and Design'' '''83'''(2-3), 421-424 (2008). [https://doi.org/10.1016/j.fusengdes.2007.11.009 DOI].
# J. Vega, G. Rattá, A. Murari, P. Castro, S. Dormido-Canto, R. Dormido, G. Farías, A. Pereira, A. Portas, E. de la Luna, I. Pastor, J. Sánchez, N. Duro, R. Castro, M. Santos and H. Vargas. '''Recent results on structural pattern recognition for Fusion massive databases.''' Proceedings of IEEE WISP 2007 (conference proceedings, not a journal article). [https://doi.org/10.1109/WISP.2007.4447569 DOI].

=== Scientific background: technical report ===
* A. González-Ganzábal, G. A. Rattá, D. Gadariya and S. Dormido. '''Herramientas para la detección de MARFE en JET con cámaras operacionales.''' ''Informes Técnicos CIEMAT'' 1533, 1-32 (2024). This is a technical report, listed separately from peer-reviewed articles.

=== Scientific background: conference and meeting presentations ===
The following contributions are recorded in the coordinator's CVN. Dates are reported at the precision supported by the record; unresolved venue inconsistencies are not reproduced.
# R. S. Granetz, G. A. Rattá, A. González and collaborators. '''Update on the MDC-JA-4 High-Z UFO disruptions.''' 44th ITPA MHD, Disruptions and Control meeting, 22-25 October 2024.
# G. A. Rattá and A. González. '''Comparison of “Classic” and “State-of-the-Art” Machine Learning Techniques to Estimate the Time to Disruption.''' 44th ITPA MDC meeting, 22-25 October 2024.
# G. A. Rattá and A. González. '''A data-driven disruption classifier based on signal anomalies.''' SOFT 2024, Dublin, Ireland, 22-27 September 2024.
# A. González and G. A. Rattá. '''A methodology for TIE detection and tracking for JET's experimental cameras.''' SOFT 2024, Dublin, Ireland, 22-27 September 2024.
# G. A. Rattá and A. González. '''Identification of physics events for the characterization of disruptions in JET.''' SOFT 2024, Dublin, Ireland, 22-27 September 2024.
# G. A. Rattá and A. González. '''Estimating Time to Disruption Using Transformer Models.''' Third IAEA Technical Meeting on Plasma Disruptions and their Mitigation, ITER, Saint-Paul-lès-Durance, France, 3-6 September 2024. The CVN distinguishes this slide title from the programme title, ''Time-to-Disruption Estimation Using LSTM Networks''. [https://conferences.iaea.org/event/380/contributions/32299/ Contribution record].
# R. S. Granetz, G. A. Rattá, A. González and collaborators. '''MDC-JA-4 High-Z UFO disruptions.''' 43rd ITPA MDC meeting, 22-26 April 2024.
# G. Farías, S. Dormido-Canto, J. Vega and G. Rattá. '''Automatic feature extraction in large fusion databases by using deep learning approach.''' 10th IAEA Technical Meeting on Control, Data Acquisition and Remote Participation for Fusion Research, Ahmedabad, India, 20-24 April 2015.
# J. Vega, A. Murari, S. Dormido-Canto, D. Alves, G. Farías, J. M. López, R. Moreno, A. Pereira, J. M. Martínez and G. Rattá. '''Overview of Real-Time Disruption Prediction in JET: Applicability to ITER.''' 41st EPS Conference on Plasma Physics, Berlin, Germany, 23-27 June 2014.
# J. Vega, A. Murari, G. A. Rattá, G. Farías and collaborators. '''An overview about recent developments on advanced data analysis techniques in fusion.''' 6th Workshop on Fusion Data Processing, Validation and Analysis, Madrid, Spain, January 2010.
# J. Vega, G. Rattá, A. Murari, G. Farías and collaborators. '''Recent results on structural pattern recognition for Fusion massive databases.''' IEEE WISP, Alcalá de Henares, Spain, 3-5 October 2007. This is the presentation associated with the proceedings paper listed above.
# S. Dormido-Canto, G. Farías, R. Dormido, J. Sánchez, N. Duro, H. Vargas, J. Vega, G. Rattá, A. Pereira and A. Portas. '''Comparison of structural pattern recognition methods for fusion databases.''' 6th IAEA Technical Meeting on Control, Data Acquisition and Remote Participation for Fusion Research, Inuyama, Japan, June 2007.
# G. A. Rattá, J. Vega, A. Pereira, A. Portas, E. de la Luna, S. Dormido-Canto, G. Farías and collaborators. '''First applications of structural pattern recognition methods to the investigation of specific physical phenomena at JET.''' 6th IAEA Technical Meeting on Control, Data Acquisition and Remote Participation for Fusion Research, Inuyama, Japan, June 2007.

== Seminars, international collaboration and outreach ==
* '''18 June 2025, Universidad Pablo de Olavide (Spain).''' Seminar: ''Red estratégica STRANAI: acelerando la investigación en fusión nuclear mediante Inteligencia Artificial'', by G. A. Rattá (CVN).
* '''19 August 2025, PUCV, Valparaíso (Chile).''' Seminar: ''La red STRANAI y el proyecto AID-DAEDALUS: estableciendo sinergias entre fusión nuclear e Inteligencia Artificial'', by G. A. Rattá. [https://www.instagram.com/p/DNYx-GyPAfi/ PUCV announcement].
* '''3 September 2025, IEE-UNSJ/CONICET, San Juan (Argentina).''' Seminar: ''STRANAI y AID-DAEDALUS. Investigación colaborativa en IA y fusión nuclear para impulsar la energía del futuro'', by G. A. Rattá. [https://www.unsj.edu.ar/home/noticias_detalles/9609/2 UNSJ report].
* '''16 September 2025, CSC-CONICET, Buenos Aires (Argentina).''' Seminar: ''Inteligencia artificial, fusión nuclear y supercomputación'', by G. A. Rattá (date and title from CVN).
* '''26 September 2025.''' CIEMAT published a report on the Chile and Argentina collaboration visits. [https://www.fusion.ciemat.es/lnf/la-red-stranai-impulsa-nuevas-colaboraciones-en-latinoamerica-en-inteligencia-artificial-aplicada-a-fusion-nuclear/ Institutional report].
* '''23 March 2026, Department of Computer Science, CINVESTAV (Mexico).''' Seminar on AI for accelerating fusion research and CIEMAT-CINVESTAV collaboration through STRANAI and AID-DAEDALUS, delivered by G. A. Rattá during his February-April 2026 research stay. [https://www.fusion.ciemat.es/lnf/ia-para-acelerar-la-investigacion-en-fusion-nuclear-colaboracion-entre-ciemat-y-cinvestav-en-los-proyectos-aid-daedalus-y-stranai/ CIEMAT report] and [https://www.youtube.com/watch?v=3EMjasIEkwA video].
* '''21 May 2026.''' CIEMAT announced the registration of DIONE. [https://www.ciemat.es/w/el-ciemat-registra-dione-un-software-basado-en-inteligencia-artificial-para-acelerar-el-an%C3%A1lisis-de-plasmas-de-fusi%C3%B3n Institutional news].

=== Media coverage of the San Juan activities ===
These links cover the same outreach activity and are not counted as additional scientific presentations:
* [https://www.diariodecuyo.com.ar/columnasdeopinion/con-un-disertante-espanol-se-realizara-un-seminario-sobre-inteligencia-artificial-y-fusion-nuclear-en-la-unsj-1765893.html Diario de Cuyo].
* [https://www.canal13sanjuan.com/tecnologia/inteligencia-artificial-y-energia---un-futuro-comun-en-la-fusion-nuclear-_a68ba462895043f2db4888f4b Canal 13 San Juan].
* [https://www.diarioelzondasj.com.ar/338855-un-seminario-en-san-juan-abordara-la-fusion-nuclear-y-la-inteligencia-artificial Diario El Zonda].

== References ==
* [https://sites.fusion.ciemat.es/stranai/ Official STRANAI website] and [https://sites.fusion.ciemat.es/stranai/articles/ news].
* [https://www.aei.gob.es/sites/default/files/convocatory_info/assistants/2025-09/RED2024-RC-Texto%2BAnexo-firmado.pdf AEI award resolution, RED2024-153593-E].
* Coordinator's CVN, updated 27 October 2026: bibliographic records, conference contributions, seminars and software registration. Individual publication records are linked through their DOIs above.
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2026-04-27T18:39:56Z

Gratta: /* LNF - Nationally funded project */

== LNF - Nationally funded project ==

[[File:Logos_MinFederAEI.jpg|thumb|right|300px|Ayuda PID2024-157659OB-I00 financiada por MICIU/AEI/10.13039/501100011033 y por FEDER Una manera de hacer Europa.]]

'''Title''': '''STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques'''

'''Reference''': RED2024-153593-E

'''Programme and date''': Proyectos de 2 Años

'''Programme type (Modalidad de proyecto)''': Red de investigación

'''Area/subarea (Área temática / subárea)''': Energía / Ciencias físicas

'''Principal Investigator(s)''': [https://orcid.org/0000-0002-5676-9631 Giuseppe Rattá Gutiérrez]

'''Project type''': Proyecto coordinado

'''Start-end dates''': 01/05/2025 - 31/04/2027

'''Financing granted (direct costs)''': 65000 €

== Description of the project ==

STRANAI is a national strategic network coordinated by the Laboratorio Nacional de Fusión (CIEMAT) that connects eight leading Spanish research groups in the fields of artificial intelligence, plasma physics, and fusion technology. The goal is to promote national and international collaboration, particularly within the framework of EUROfusion, through the development and application of AI tools to critical challenges in magnetic confinement fusion.

The network focuses on six research lines: (1) detection of physical events from videos and images, (2) automation of turbulence analysis via Doppler reflectometry, (3) development of explainable and physics-informed AI models, (4) creation of shared data and code repositories, (5) plasma control and operational limits, and (6) disruption prediction and mitigation.

Participating groups include CIEMAT-LNF, BSC, UNED, CETA-CIEMAT, IFMIF-DONES, UPO, CAR (CSIC-UPM), and the PSFT group at the University of Seville. The network benefits from access to major international facilities and HPC resources. STRANAI aims to develop transferable tools and methodologies for devices such as ITER, DEMO, SMART, and IFMIF-DONES.

== Main results ==
To be completed at the end of the project (taken from the final report)

== Dissemination of project results (peer-reviewed publications and conference presentations) ==
Enter text here

== References ==
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: AID-DAEDALUS. : AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics (2025-2028)

2026-04-27T18:39:44Z

Gratta: /* LNF - Nationally funded project */

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2025-10-08T13:03:18Z

Gratta: /* LNF - Nationally funded project */

== LNF - Nationally funded project ==

'''Title''': '''STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques'''

'''Reference''': RED2024-153593-E

'''Programme and date''': Proyectos de 2 Años

'''Programme type (Modalidad de proyecto)''': Red de investigación

'''Area/subarea (Área temática / subárea)''': Energía / Ciencias físicas

'''Principal Investigator(s)''': [https://orcid.org/0000-0002-5676-9631 Giuseppe Rattá Gutiérrez]

'''Project type''': Proyecto coordinado

'''Start-end dates''': 01/05/2025 - 31/04/2027

'''Financing granted (direct costs)''': 65000 €

== Description of the project ==

STRANAI is a national strategic network coordinated by the Laboratorio Nacional de Fusión (CIEMAT) that connects eight leading Spanish research groups in the fields of artificial intelligence, plasma physics, and fusion technology. The goal is to promote national and international collaboration, particularly within the framework of EUROfusion, through the development and application of AI tools to critical challenges in magnetic confinement fusion.

The network focuses on six research lines: (1) detection of physical events from videos and images, (2) automation of turbulence analysis via Doppler reflectometry, (3) development of explainable and physics-informed AI models, (4) creation of shared data and code repositories, (5) plasma control and operational limits, and (6) disruption prediction and mitigation.

Participating groups include CIEMAT-LNF, BSC, UNED, CETA-CIEMAT, IFMIF-DONES, UPO, CAR (CSIC-UPM), and the PSFT group at the University of Seville. The network benefits from access to major international facilities and HPC resources. STRANAI aims to develop transferable tools and methodologies for devices such as ITER, DEMO, SMART, and IFMIF-DONES.



== References ==
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: AID-DAEDALUS. : AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics (2025-2028)

2025-10-08T13:01:32Z

Gratta: Created page with "== LNF - Nationally funded project == '''Title''': '''AID-DAEDALUS. AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics''' '''Reference''': PID2024-157659OB-I00 '''Programme and date''': Plan Estatal de Investigación Científica, Técnica y de Innovación 2024-2027 '''Programme type (Modalidad de proyecto)''': Proyectos de Generación de Conocimiento 2024 '''Area/s..."

== LNF - Nationally funded project ==

'''Title''': '''AID-DAEDALUS. AI Driven Digital twin prototype of the TJ-II and OLMAT for Advanced Experimental Design, Analysis and Leading-edge Understanding of Stellarators physics'''

'''Reference''': PID2024-157659OB-I00

'''Programme and date''': Plan Estatal de Investigación Científica, Técnica y de Innovación 2024-2027

'''Programme type (Modalidad de proyecto)''': Proyectos de Generación de Conocimiento 2024

'''Area/subarea (Área temática / subárea)''': Energía y transporte / Energía

'''Principal Investigator(s)''': [https://orcid.org/0000-0002-5676-9631 Giuseppe Rattá Gutiérrez]

'''Project type''': Proyecto individual

'''Start-end dates''': 01/09/2025 - 30/08/2028

'''Financing granted (direct costs)''': 65000 €

== Description of the project ==

El proyecto AID-DAEDALUS busca crear prototipos de gemelos digitales impulsados por inteligencia artificial de los dispositivos de fusión TJ-II y OLMAT del CIEMAT. Su finalidad es simular, predecir y controlar en tiempo real el comportamiento del plasma y de los materiales sometidos a condiciones extremas, con el fin de optimizar los recursos y reducir los costes de las campañas experimentales. Para ello se aprovechará una amplia base de datos con más de cincuenta mil experimentos del TJ-II y los registros de OLMAT, que servirán para entrenar modelos de redes neuronales capaces de reproducir parámetros como la densidad electrónica, la radiación o los daños en tungsteno, y de generar señales sintéticas cuando las mediciones reales sean incompletas o defectuosas.

Durante tres años se desarrollarán bases de datos estandarizadas, modelos de inteligencia artificial y pruebas de control en tiempo real utilizando el supercomputador Turgalium. El proyecto cuenta con la colaboración de centros nacionales e internacionales y con la participación de varios doctorandos, y se espera que refuerce el liderazgo científico de España en el uso de inteligencia artificial aplicada a la energía de fusión, reduciendo los costes experimentales y promoviendo la transferencia de estas tecnologías a otros dispositivos europeos y a sectores como la ciencia de materiales o los sistemas energéticos avanzados.

== References ==
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2025-07-28T10:15:40Z

Gratta: /* LNF - Nationally funded project */

== LNF - Nationally funded project ==

'''Title''': '''STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques'''

'''Reference''': RED2024-153593-E

'''Programme and date''': Proyectos de 2 Años

'''Programme type (Modalidad de proyecto)''': Red de investigación

'''Area/subarea (Área temática / subárea)''': Energía / Ciencias físicas

'''Principal Investigator(s)''': [https://orcid.org/ 0000-0002-5676-9631 Giuseppe Rattá Gutiérrez]

'''Project type''': Proyecto coordinado

'''Start-end dates''': 01/05/2025 - 31/04/2027

'''Financing granted (direct costs)''': 65000 €

== Description of the project ==

STRANAI is a national strategic network coordinated by the Laboratorio Nacional de Fusión (CIEMAT) that connects eight leading Spanish research groups in the fields of artificial intelligence, plasma physics, and fusion technology. The goal is to promote national and international collaboration, particularly within the framework of EUROfusion, through the development and application of AI tools to critical challenges in magnetic confinement fusion.

The network focuses on six research lines: (1) detection of physical events from videos and images, (2) automation of turbulence analysis via Doppler reflectometry, (3) development of explainable and physics-informed AI models, (4) creation of shared data and code repositories, (5) plasma control and operational limits, and (6) disruption prediction and mitigation.

Participating groups include CIEMAT-LNF, BSC, UNED, CETA-CIEMAT, IFMIF-DONES, UPO, CAR (CSIC-UPM), and the PSFT group at the University of Seville. The network benefits from access to major international facilities and HPC resources. STRANAI aims to develop transferable tools and methodologies for devices such as ITER, DEMO, SMART, and IFMIF-DONES.



== References ==
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2025-05-27T14:56:23Z

Gratta: /* Description of the project */

== LNF - Nationally funded project ==

'''Title''': '''STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques'''

'''Reference''': RED2024-153593-E

'''Programme and date''': Proyectos de 2 Años

'''Programme type (Modalidad de proyecto)''': Red de investigación

'''Area/subarea (Área temática / subárea)''': Energía / Ciencias físicas

'''Principal Investigator(s)''': Giuseppe Rattá Gutiérrez [https://orcid.org/ 0000-0002-5676-9631]

'''Project type''': Proyecto coordinado

'''Start-end dates''': 01/05/2025 - 31/04/2027

'''Financing granted (direct costs)''': 65000 €

== Description of the project ==

STRANAI is a national strategic network coordinated by the Laboratorio Nacional de Fusión (CIEMAT) that connects eight leading Spanish research groups in the fields of artificial intelligence, plasma physics, and fusion technology. The goal is to promote national and international collaboration, particularly within the framework of EUROfusion, through the development and application of AI tools to critical challenges in magnetic confinement fusion.

The network focuses on six research lines: (1) detection of physical events from videos and images, (2) automation of turbulence analysis via Doppler reflectometry, (3) development of explainable and physics-informed AI models, (4) creation of shared data and code repositories, (5) plasma control and operational limits, and (6) disruption prediction and mitigation.

Participating groups include CIEMAT-LNF, BSC, UNED, CETA-CIEMAT, IFMIF-DONES, UPO, CAR (CSIC-UPM), and the PSFT group at the University of Seville. The network benefits from access to major international facilities and HPC resources. STRANAI aims to develop transferable tools and methodologies for devices such as ITER, DEMO, SMART, and IFMIF-DONES.



== References ==
<references />


<hr>
[[LNF:Nationally Funded Projects|Back to list of nationally funded projects]]

[[Category:LNF Nationally Funded Projects]]

LNF: STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques

2025-05-27T14:55:55Z

Gratta: Created page with "== LNF - Nationally funded project == '''Title''': '''STRANAI: STRAtegic network for Nuclear fusion research acceleration through Artificial Intelligence techniques''' '''Reference''': RED2024-153593-E '''Programme and date''': Proyectos de 2 Años '''Programme type (Modalidad de proyecto)''': Red de investigación '''Area/subarea (Área temática / subárea)''': Energía / Ciencias físicas '''Principal Investigator(s)''': Giuseppe Rattá Gutiérrez [https://orcid..."