Remote Sensing for Monitoring Photovoltaic Solar Plants in Brazil Using Deep Semantic Segmentation

• Published in: Energies (Basel) Vol. 14; no. 10; p. 2960
• Main Authors: Costa, Marcus Vinícius Coelho Vieira da, Carvalho, Osmar Luiz Ferreira de, Orlandi, Alex Gois, Hirata, Issao, Albuquerque, Anesmar Olino de, Silva, Felipe Vilarinho e, Guimarães, Renato Fontes, Gomes, Roberto Arnaldo Trancoso, Júnior, Osmar Abílio de Carvalho
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 20.05.2021
• Subjects: Accuracy; Alternative energy sources; Computer applications; Computer vision; Costs; deep learning; Electricity; Emissions; Energy; Energy management; Energy resources; Greenhouse gases; Image classification; Industrial plant emissions; Inspection; Nuclear energy; Nuclear power plants; Photovoltaic cells; Photovoltaics; Remote monitoring; Remote sensing; Renewable resources; Semantic segmentation; Semantics; Solar energy; solar panel; Brazil
• ISSN: 1996-1073; 1996-1073
• DOI: 10.3390/en14102960

Brazil is a tropical country with continental dimensions and abundant solar resources that are still underutilized. However, solar energy is one of the most promising renewable sources in the country. The proper inspection of Photovoltaic (PV) solar plants is an issue of great interest for the Brazilian territory’s energy management agency, and advances in computer vision and deep learning allow automatic, periodic, and low-cost monitoring. The present research aims to identify PV solar plants in Brazil using semantic segmentation and a mosaicking approach for large image classification. We compared four architectures (U-net, DeepLabv3+, Pyramid Scene Parsing Network, and Feature Pyramid Network) with four backbones (Efficient-net-b0, Efficient-net-b7, ResNet-50, and ResNet-101). For mosaicking, we evaluated a sliding window with overlapping pixels using different stride values (8, 16, 32, 64, 128, and 256). We found that: (1) the models presented similar results, showing that the most relevant approach is to acquire high-quality labels rather than models in many scenarios; (2) U-net presented slightly better metrics, and the best configuration was U-net with the Efficient-net-b7 encoder (98% overall accuracy, 91% IoU, and 95% F-score); (3) mosaicking progressively increases results (precision-recall and receiver operating characteristic area under the curve) when decreasing the stride value, at the cost of a higher computational cost. The high trends of solar energy growth in Brazil require rapid mapping, and the proposed study provides a promising approach.