Best Researcher Award
Oumayma Mabrouk
University of Angers, France
Oumayma Mabrouk is a researcher affiliated with the University of Angers in France whose documented research contribution concerns photovoltaic-system performance diagnosis, renewable-energy monitoring, and data-driven anomaly detection. Her recent article in Solar Energy presents a regime-aware framework for generating robust reference curves for photovoltaic power output under changing meteorological conditions. The publication identifies an approach based on unsupervised learning, statistical reference corridors, and constrained time-series alignment. [1]
Abstract
Mabrouk’s documented publication addresses the challenge of detecting abnormal photovoltaic-system behavior without relying on extensive labeled fault datasets. The study develops reference corridors using median and interquartile-range statistics after identifying representative meteorological regimes through k-means clustering. Limited Dynamic Time Warping is then used to compare daily photovoltaic power profiles with regime-specific reference curves. The reported validation uses receiver operating characteristic analysis and synthetic fault scenarios, with the published study reporting area-under-the-curve values above 0.80. [1]
Keywords
- Photovoltaic systems
- Renewable energy technologies
- Anomaly detection
- Unsupervised learning
- Data-driven diagnosis
Introduction
Reliable monitoring is an important component of photovoltaic-system operation because environmental variability can make it difficult to distinguish normal changes in power production from genuine performance anomalies. The article by Mabrouk and co-authors approaches this problem by separating meteorological regime identification from photovoltaic power anomaly detection. This design seeks to reduce the possibility that changes in weather conditions are interpreted directly as equipment faults. [1]
Research Profile
The available publication record places Mabrouk’s research within renewable energy technologies, with a specific focus on photovoltaic performance monitoring and intelligent diagnostic methods. The 2026 Solar Energy article develops an unsupervised framework that uses environmental variables to identify weather regimes and then constructs statistical reference models for daily photovoltaic power behavior. The research combines machine-learning clustering, robust statistics, and time-series analysis in a single diagnostic workflow. [1]
Research Contributions
The principal contribution is a regime-aware methodology for establishing expected photovoltaic power behavior. Environmental measurements are used to distinguish representative meteorological conditions, while normalized daily power profiles are summarized through median curves and interquartile-range corridors. Limited Dynamic Time Warping provides constrained temporal alignment, and complementary indicators are used for anomaly classification. The study reports that this combination can identify deviations without requiring labeled fault examples. [1]
Publications
The principal documented publication is Robust reference curves generation for photovoltaic systems performance diagnosis using a data-driven approach, authored by Oumayma Mabrouk, Abdérafi Charki, Nizar Chatti, and Xavier Sidambarompoulé. It was published in Solar Energy, volume 317, article 114958, in 2026. The article is indexed in the supplied Scopus profile information and has a reported DOI of 10.1016/j.solener.2026.114958. [1] [2]
Research Impact
The supplied bibliometric record lists one document, one citation, and an h-index of 1. These indicators represent the currently supplied record rather than a comprehensive assessment of research influence. The publication itself addresses a practical problem in photovoltaic operations by combining interpretable statistical reference modeling with unsupervised anomaly detection. Its reported evaluation indicates that the proposed indicators can discriminate injected abnormal behavior from normal operational variation. [1]
Award Suitability
For the Best Researcher Award context, the documented evidence includes a peer-reviewed 2026 journal publication addressing renewable-energy system diagnosis and a methodological contribution combining k-means clustering, robust statistical modeling, and constrained time-series analysis. These documented elements provide a research basis for consideration within an award process focused on renewable energy technologies. Any final award determination remains subject to the applicable evaluation criteria and review procedure of the International Forensic Scientist Awards.
Conclusion
Oumayma Mabrouk’s documented research contribution centers on data-driven photovoltaic performance diagnosis. Her 2026 Solar Energy publication presents a structured approach for modeling expected photovoltaic behavior across meteorological regimes and detecting deviations without labeled fault data. The available bibliometric information records one document, one citation, and an h-index of 1, while the publication provides the principal evidence for assessing her current research profile.
External Links
References
- Mabrouk, O., Charki, A., Chatti, N., & Sidambarompoulé, X. (2026). Robust reference curves generation for photovoltaic systems performance diagnosis using a data-driven approach. Solar Energy, 317, 114958.
https://doi.org/10.1016/j.solener.2026.114958
- Elsevier. (2026). Robust reference curves generation for photovoltaic systems performance diagnosis using a data-driven approach. Solar Energy, Volume 317, Article 114958.
- Scopus. (n.d.). Scopus author details: Oumayma Mabrouk, Author ID 60883647200. Elsevier.
https://www.scopus.com/authid/detail.uri?authorId=60883647200
- HAL. (2026). Publication record for Robust reference curves generation for photovoltaic systems performance diagnosis using a data-driven approach. Université d’Angers research record.
https://cv.hal.science/xavier-sidambarompoule
- Chatti, N. (2026). Selected publications and research outputs: Solar Energy, 2026. University of Angers / LARIS.
https://perso-laris.univ-angers.fr/~nizar.chatti/
