Data Mining and Artificial Intelligence for the management and Diagnosis of Grid-Connected Photovoltaic Power Plants in Algeria

Abstract

The variability of solar energy presents a major challenge for the expansion of photovoltaic systems, as it affects grid stability and overall energy management. To address this issue, advanced forecasting techniques are required to accurately predict production fluctuations. This study focuses on short-term photovoltaic energy forecasting in Algeria’s High Plains, using real-world data from the Djelfa power plant. Deep learning models are developed to improve forecasting accuracy while capturing monthly variations. The findings contribute to optimizing photovoltaic energy integration and supporting decision-making for grid operators and energy policymakers.