| Diego Rivelino Espinoza Trejo | Universidad Autónoma de San Luis Potosí |
| Luis Ángel López Vélez | Benebión de PHYTOSAN S.A. de C.V. |
| Vladimir Ilich Ruiz Salinas | Fronius Mexico S.A. DE C.V. |
| Isaac Compeán Martínez | Universidad Autónoma de San Luis Potosí |
| Daniel Ulises Campos Delgado | Universidad Autónoma de San Luis Potosí |
| Cristian H. De Angelo | Universidad Nacional de Río Cuarto |
https://doi.org/10.58571/CNCA.AMCA.2025.013
Resumen: Small- and medium-scale photovoltaic (PV) plants commonly rely on monitoring systems limited to electrical parameters, such as generated power. Incorporating sensors to measure plane-of-array irradiance and module operating temperature is often not cost-effective in these installations, which restricts the use of traditional strategies for anomaly detection and performance evaluation. This paper proposes an advanced monitoring scheme based on unsupervised learning, specifically using hierarchical clustering, to identify atypical behaviors and assess the relative performance of neighboring PV inverters. The methodology is grounded in fault diagnosis principles, such as physical redundancy and parity relations, leveraging the expected similarity among power generation profiles. The proposed approach is validated using real data from a 240 kW PV plant consisting of 16 inverters rated at 15 kW, 304 modules of 450 Wp, and 238 modules of 545 Wp.
¿Cómo citar?
Espinoza Trejo, D., López Vélez, L., Ruiz Salinas, V., Compeán Martínez, I., Campos Delgado, D. & De Angelo, C. (2025). Advanced Monitoring of Photovoltaic Plants without Irradiance Sensors Using Unsupervised Learning. Memorias del Congreso Nacional de Control Automático 2025, pp. 74-79. https://doi.org/10.58571/CNCA.AMCA.2025.013
Palabras clave
Anomaly detection, Performance evaluation, Photovoltaic plants, Unsupervised learning, Fault diagnosis.
Referencias
- Bilal Taghezouit, Fouzi Harrou, Ying Sun, and Walid Merrouche. Model-based fault detection in photovoltaic systems: A comprehensive review and avenues for enhancement. Results in Engineering, Volume 21, 2024, Article 101835. ISSN: 2590-1230.
- International Energy Agency (IEA). Photovoltaic Power Systems Programme – Performance, Operation and Reliability of Photovoltaic Systems. IEA PVPS Task 13, Report IEA-PVPS T13-25:2022, October 2022. ISBN 978-3-907281-13-0.
- Joakim Wid´en and Joakim Munkhammar. Solar Radiation Theory. Uppsala University, 2019. ISBN 978-91-506-2760-2. DOI: 10.33063/diva-381852.
- International Energy Agency (IEA). Photovoltaic Power Systems Programme – The Use of Advanced Algorithms in PV Failure Monitoring. IEA PVPS Task 13, Report IEA-PVPS T13-19:2021, September 2021. ISBN 978-3-907281-07-9.
- Secretaría de Energía (SENER). NORMA Oficial Mexicana NOM-001-SEDE-2012, Instalaciones Eléctricas (utilización). Diario Oficial de la Federación, México, 2012. Artículo 690 [690-11, 690-35 c)].
- Tara Doyle, Ryan Desharnais, and Michael Mills-Price. 2019 PV Inverter Scorecard. PV Evolution Labs (PVEL), May 2019. https://www.pvel.com/inverter-scorecard/.
- Thushara Gunda and Rachel Homan. Evaluation of Component Reliability in Photovoltaic Systems Using Field Failure Statistics. Sandia National Laboratories, Albuquerque, NM, Report SAND2020-9231, 2020. DOI: 10.2172/1660804.
- H. Zhu, Z. Wang, and R. S. Balog. Real time arc fault detection in PV systems using wavelet decomposition. In 2016 IEEE 43rd Photovoltaic Specialists Conference (PVSC), Portland, OR, USA, pages 1761–1766, 2016. DOI: 10.1109/PVSC.2016.7749926.
- Trevor Hastie, Robert Tibshirani, and Jerome Friedman. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Second Edition, Springer, 2009. ISBN 978-0-387-84857-0.
- Posit team. RStudio: Integrated Development Environment for R. Posit Software, PBC, Boston, MA, 2025. URL: http://www.posit.co/.