A Self-Supervised Machine Learning Approach for the Estimation of Open-Circuit Voltage Degradation in Photovoltaic Systems

In this work, we present a machine learning (ML) approach that integrates physics-based knowledge with data-driven techniques to estimate the open-circuit voltage ((Formula presented.)) of photovoltaic (PV) systems. This self-supervised approach allows the detection of anomalies in operating systems, enabling the identification of potential faults and degradation related to (Formula presented.) without requiring labelled data. Deviations in (Formula presented.) values are detected by analysing the measurements recorded in the supervisory control and data acquisition (SCADA) system. This analysis is performed by a combination of clustering and regression algorithms. The proposed approach is validated on three different PV installations, showcasing its ability to detect variations in open-circuit voltage and to predict performance degradation. The proposed method achieves an (Formula presented.) -squared (r2) value larger than 0.9 when trained on experimental (Formula presented.) data from three distinct PV systems. Moreover, it estimates (Formula presented.) from SCADA data with an average error below 5% compared with (Formula presented.) – (Formula presented.) curve measurements. The results of this study demonstrate that physics informed ML techniques can significantly enhance the performance and reliability of PV systems, enabling early fault detection and degradation forecasting.