{rfName}
AI

Indexed in

Altmetrics

Analysis of institutional authors

Queral C.AuthorFernández-Cosials K.Author

Share

November 21, 2025
Publications
>
Proceedings Paper
No

AI-based Project CERO: deep learning for operating experience in NPPs

Publicated to: Transactions of the American Nuclear Society. 132 (1): 744-747 - 2025-01-01 132(1), DOI: 10.13182/T140-48321

Authors:

Queral C; Castejón F; Durán Batalla JL; Fernandez-Cosials K; García A; Iñiguez D; Isasia R; Tarancon A; Yllanes D
[+]

Affiliations

Instituto de Biocomputación y Física de Sistemas Complejos (BIFI); Zaragoza; 50018; Spain; Departamento de Física Teórica; Universidad de Zaragoza; Zaragoza; 50009; Spain - Author
Kampal Data Solutions; WTCZ; Avda. Maria Zambrano 31; Zaragoza; 50018; Spain - Author
Kampal Data Solutions; WTCZ; Avda. Maria Zambrano 31; Zaragoza; 50018; Spain; Fundación ARAID; Diputación General de Aragón; Zaragoza; 50018; Spain; Instituto de Biocomputación y Física de Sistemas Complejos (BIFI); Zaragoza; 50018; Spain; Departamento de - Author
Kampal Data Solutions; WTCZ; Avda. Maria Zambrano 31; Zaragoza; 50018; Spain; Fundación ARAID; Diputación General de Aragón; Zaragoza; 50018; Spain; Instituto de Biocomputación y Física de Sistemas Complejos (BIFI); Zaragoza; 50018; Spain; Zaragoza Scient - Author
Nuclear Safety Council; Pedro Justo Dorado 11. 28040; Madrid; Spain - Author
Nuclear Safety Council; Pedro Justo Dorado 11. 28040; Madrid; Spain; Nuclear Fusion Laboratory. CIEMAT.; Av Complutense 40. 28040; Madrid; Spain - Author
Universidad Politécnica de Madrid; Alenza Street; Madrid; 28003; Spain - Author
See more

Abstract

In the field of operating experience, the ability to efficiently retrieve relevant incident reports in response to a given query is essential for learning from past incidents and preventing future occurrences. Current search engines rely on literal matches on whole documents or on preexisting manual classification or annotation. In Project CERO, supported by the Spanish Nuclear Safety Council and based on the NRC ADAMS database, we have developed a state-of-the-art search engine leveraging the latest advances in deep learning. The system builds upon a unified metadata schema, keyphrase extraction and a strong semantic-search model, giving rise to a powerful engine. © 2025, American Nuclear Society. All rights reserved.
[+]

Keywords

'currentDeep learningIncident reportsKey-phrases extractionsLiteralsManual annotationManual classificationMetadata schemaNuclear safetyOperating experienceQuery processingSearch enginesSemanticsState of the art

Quality index

Bibliometric impact. Analysis of the contribution and dissemination channel

The work has been published in the journal Transactions of the American Nuclear Society, Q3 Agency Scopus (SJR), its regional focus and specialization in Industrial and Manufacturing Engineering, give it significant recognition in a specific niche of scientific knowledge at an international level.

[+]

Impact and social visibility

From the perspective of influence or social adoption, and based on metrics associated with mentions and interactions provided by agencies specializing in calculating the so-called "Alternative or Social Metrics," we can highlight as of 2026-04-06:

  • The use of this contribution in bookmarks, code forks, additions to favorite lists for recurrent reading, as well as general views, indicates that someone is using the publication as a basis for their current work. This may be a notable indicator of future more formal and academic citations. This claim is supported by the result of the "Capture" indicator, which yields a total of: 1 (PlumX).
[+]

Leadership analysis of institutional authors

There is a significant leadership presence as some of the institution’s authors appear as the first or last signer, detailed as follows: First Author (QUERAL SALAZAR, JOSE CESAR) .

[+]

Project objectives

Los objetivos perseguidos en esta aportación se centran en mejorar la gestión de la experiencia operativa en centrales nucleares mediante técnicas avanzadas de inteligencia artificial. Se pretende analizar la capacidad de recuperación eficiente de informes de incidentes relevantes ante consultas específicas, evaluar las limitaciones de los motores de búsqueda actuales basados en coincidencias literales y clasificaciones manuales, desarrollar un motor de búsqueda avanzado basado en aprendizaje profundo, caracterizar un esquema unificado de metadatos y extracción de frases clave, y validar un modelo semántico robusto que potencie la búsqueda de información en la base de datos NRC ADAMS, con el fin de facilitar el aprendizaje y la prevención de incidentes futuros en el sector nuclear.
[+]

Most relevant results

Los resultados más relevantes del estudio demuestran avances significativos en la recuperación eficiente de informes de incidentes en centrales nucleares mediante aprendizaje profundo. En primer lugar, se desarrolló un motor de búsqueda avanzado basado en deep learning que supera los métodos tradicionales basados en coincidencias literales o clasificaciones manuales. En segundo lugar, se implementó un esquema unificado de metadatos que mejora la organización y accesibilidad de la información. En tercer lugar, se integró un modelo semántico robusto junto con la extracción automática de palabras clave, optimizando la precisión y relevancia de las consultas. Finalmente, el sistema se validó utilizando la base de datos NRC ADAMS, garantizando su aplicabilidad en entornos reales de experiencia operativa.
[+]