Enhancing Production Forecasting of Grid-Connected PV Strings Operating under Semi-Arid Climate Conditions

Integrating photovoltaic (PV) solar energy into the electrical grid poses significant challenges due to the unpredictability of PV systems production, which introduces variability and complicates synchronization between electricity generation and consumption [1], [2]. Thus, accurate forecasting of PV power production is crucial for ensuring grid stability and efficient energy management [3]. This study focuses on modeling and forecasting the performance of grid-connected PV strings coming from different technologies, aiming to enhance the accuracy of output power prediction. The semi-arid climate, characterized by high solar irradiance and significant temperature fluctuations [4], [5], presents unique challenges and opportunities for solar energy generation. To accurately predict the performance of PV strings in these conditions, the study employs predictive approaches based on both implicit and explicit models: Single Diode Model (SDM) and Das Model (DM). New analytical formulas for DM shape parameters are introduced. The reliability of the proposed approaches is assessed by comparing I-V curves, P-V curves and peak power predicted by SDM and BM models to the actual measurements of PV strings coming from different technologies and operating at Green Energy Park (GEP) research facility in Morocco. The comparison reveals that the predicted outcomes align well with the real data, with an average of Normalized Root Mean Square Error (NRMSE) not exceeding 3.34% for SDM and 5% for DM throughout the day. The results demonstrate that the predictive approaches effectively forecast the production of PV strings, accounting for climatic influences and providing insights into optimizing PV system performance, thereby contributing to improved grid stability in semi-arid regions.