Fault Detection in Solar Power Plants Based on Energy Production Data

Authors

Dominykas Vilčinskas
Vilnius University, Institute of Applied Mathematics
Lukas Voveris
Vilnius University, Institute of Applied Mathematics
Jolita Bernatavičienė
Vilnius University, Institute of Data Science and Digital Technologies

Synopsis

This research addresses the critical need for the timely identification of faults in solar power plants to minimize electricity loss. The study analyses energy production data from a Lithuanian solar power plant, comprising 143 strings distributed across 12 inverters, over a 19-month period. During data preprocessing, 16 key features were extracted from each string's time series data to represent the global structure of the data. The extraction process resulted in a transformed dataset, where each time series is represented as an object with 16 features, enabling for more effective analysis. Statistical and machine learning techniques — including PCA + $\alpha$-HULL, Isolation Forest (iForest), and Local Outlier Factor (LOF) — were employed to identify systems exhibiting abnormal behavior. The results demonstrate that a combination of these methods can help effectively identify outliers, with a combined anomaly score providing a comprehensive assessment of string performance. Additionally, RANSAC and DBSCAN methods were used to construct fault profiles, which enabled a more in-depth analysis of each system's performance and provided further confirmation of previously identified systems exhibiting abnormal behavior.

Author Biographies

Dominykas Vilčinskas, Vilnius University, Institute of Applied Mathematics

Dominykas Vilčinskas is Data Science student at Vilnius University, currently pursu-ing a Master’s degree after completing his Bachelor’s studies in Data Science in 2025. During his studies, he has been actively involved in academic events and research ac-tivities focused on data analysis and machine learning. He has presented his work at several conferences, including a poster presentation titled ”Fault Detection in Solar Power Plants Using Energy Production Data” at the 15th Conference on Data Analysis Methods for Software Systems in 2024. He also co-authored a publication on ”Wage Prediction for Salaried Employees Using Machine Learning Methods,” presented at the national conference Lietuvos magistrantų informatikos ir IT tyrimai in 2025. His main research interests include applied machine learning and data-driven decision making.

Vilnius, Lithuania. E-mail: dominykas.vilcinskas@mif.stud.vu.lt

Lukas Voveris, Vilnius University, Institute of Applied Mathematics

Lukas Voveris is a Master’s student in Data Science at Vilnius University, Faculty of Mathematics and Informatics. He earned his Bachelor’s degree in Data Science at Vil-nius University in 2025. His research focuses on solar PV monitoring and energy ana-lytics. He co-authored and presented the poster ”Fault Detection in Solar Power Plants Using Energy Production Data” at the 15th Conference on Data Analysis Methods for Software Systems in 2024. He also authored ”Implementation of Machine Learning and Statistical Techniques in Solar Energy Generation Monitoring Systems”. This paper was presented at the national conference Lietuvos magistrantų informatikos ir IT tyrimai in 2025 and at the AI2SEP project conference in Varaždin, Croatia. His research interests include applied machine learning, time series forecasting, solar irradiation modeling, and decision support for energy systems.

Vilnius, Lithuania. E-mail: lukas.voveris@mif.stud.vu.lt

Jolita Bernatavičienė, Vilnius University, Institute of Data Science and Digital Technologies

Jolita Bernatavičienė graduated from Vilnius Pedagogical University in 2004 and re-ceived a master’s degree in informatics. In 2008, she received a doctoral degree in computer science (PhD) from the Institute of Mathematics and Informatics jointly with Vilnius Gediminas Technical University. She is a senior researcher at the Cognitive Computing Group of Vilnius University’s Institute of Data Science and Digital Tech-nologies. Her research interests include databases, data mining, neural networks, image analysis, visualisation, decision support systems and internet technologies, and high-performance computing. She supervises 3 PhD students and has written more than 60 articles, 18 of which are in CA WoS database.

Vilnius, Lithuania. E-mail: jolita.bernataviciene@mif.vu.lt

Downloads

Published

June 18, 2026

License

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Vilčinskas, D., Voveris, L., & Bernatavičienė, J. (2026). Fault Detection in Solar Power Plants Based on Energy Production Data. In R. Leskovar (Ed.), Artificial Intelligence and Environmental Challenges: Research Insights and Emerging Solutions (pp. 87-106). University of Maribor Press. https://doi.org/10.18690/um.fov.5.2026.5