Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{201048,
        author = {MS.J. Jesimaagnasrajamalar and AP Tamizh Kumaran M and Ragu S and Santhosh Kumar S},
        title = {AI Solar Dust Accumulation Predictor Using Operational Data},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {no},
        pages = {253-257},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201048},
        abstract = {Dust accumulation on photovoltaic (PV) solar panels significantly degrades energy output, causing losses of 20–40% in high-irradiance regions. This paper presents SolarSense, a machine learning-powered web application that detects and quantifies dust-induced performance degradation using standard operational sensor data. An XGBoost regression model is trained on 17,401 real-world 15-minute interval records to predict the Performance Ratio (PR) of solar panels under clean conditions. The predicted PR is compared against actual energy generation to compute energy loss, which is mapped to a three-level dust severity classification (Low, Medium, High) with actionable maintenance recommendations. The model achieves an R² score of 0.9412 and MAE of
0.031. A Flask-based web interface supports manual data entry and CSV batch prediction modes with optional email alert dispatch. The framework requires no specialised soiling sensors, relying entirely on standard plant monitoring infrastructure.},
        keywords = {solar panel soiling; dust detection; XGBoost; performance ratio; predictive maintenance; Flask; machine learning; photovoltaic systems},
        month = {May},
        }