Algerian Journal of Renewable Energy and Sustainable Development
Volume 7, Numéro 1, Pages 51-57
2025-06-15
Recommender System For Optimal Solar Panel Placement Using Satellite Imagery And Weather Data
Authors : Abderaouf Bahi . Ibtissem Gasmi . Sassi Bentrad .
Abstract
The optimal placement of solar panels plays a critical role in maximizing photovoltaic energy production, especially in regions with diverse geographic, climatic, and urban characteristics. In this work, we propose a novel recommender system that leverages artificial intelligence to suggest ideal solar panel locations using a combination of satellite imagery, weather data, and deep learning techniques. Our method integrates rooftop segmentation via U-Net, image feature extraction using a fine-tuned ResNet50, and contextual weather data from NASA's POWER database. The scoring function for recommendations balances three core factors: sunlight exposure, shading impact, and physical rooftop suitability. We validate the system across three Algerian regions—urban (Algiers), rural (Bouira), and coastal (Oran)—and show that our model outperforms ablated versions, achieving high precision and accurate segmentation. The proposed approach demonstrates significant potential to support sustainable energy planning and smart grid deployment in varied environments.
Keywords
Solar panel placement; Recommender system; Satellite imagery; Deep learning; Renewable energy;
Les articles similaires
Souhaib Remha
.
Saliha Chettih
.
Salem Arif
.
pages 44-49.
Zellagui Mohamed
.
pages 627-639.
Bouaraki Mouloud
.
Recioui Abdelmadjid
.
pages 102-109.
العوفي حكيمة
.
بصدار زوليخة
.
ص 95-104.
Kebaïli S.
.
Benalla H.
.
pages 179-191.
