• Manshi Hooda
• Dr. Neha Khurana
• Dr. Gopal Krishan
• Dr. Smita Pareek
• Ankush Tyagi

Solar panels, also known as photovoltaic (PV) modules play a role, in the world of energy by transmuting sunlight into electricity and providing a green alternative to traditional fossil fuels. As the need for energy continues to rise there is a growing emphasis, on solar PV systems. Yet the efficiency of these systems can be greatly affected by conditions with partial shadowing standing out as a major challenge. When photovoltaic (PV) modules are partially shaded their performance tends to decrease leading to energy production and system inefficiencies. To assess how 4x4, 5x5, and 6x4 solar PV modules perform under shading conditions this study compares three intelligence (AI) methods; neural networks (NN), Regression, and Support vector machines (SVM). Data, from shadowing scenarios in solar PV systems were collected for analysis. The models created using each AI technique were evaluated using accuracy metrics such as Mean Absolute Error (MAE) Mean Squared Error (MSE) Root Mean Squared Error (RMSE) and the Coefficient of Determination (R²). Our findings indicate that neural networks effectively predict module performance across all configurations and capture the effects of shading. Support vector machines also demonstrate performance, with balancing and computational efficiency. On the other hand regression models exhibit prediction accuracy expressly for complex shading patterns despite being simpler and quicker to train. This research underscores how AI methods can enhance solar PV module performance predictions, aiding in optimizing energy systems with miscellaneous module setups.

series-parallel, total-cross-tied, photovoltaic panel, hybrid reconfiguration, artificial intelligence, neural network, regression, support vector machine

"Performance Prediction of Shaded Solar PV Modules Using Neural Networks, Regression, and SVM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 5, page no.p650-p659, May-2023, Available :http://www.jetir.org/papers/JETIR2305G86.pdf

"Performance Prediction of Shaded Solar PV Modules Using Neural Networks, Regression, and SVM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 5, page no. ppp650-p659, May-2023, Available at : http://www.jetir.org/papers/JETIR2305G86.pdf