Solar Park Detection Based On Machine Learning
- 1Indian Institute of Technology Bombay, Civil Engineering, Mumbai, India (shivamb3@gmail.com)
- 2Indian Institute of Technology Bombay, Civil Engineering, Mumbai, India (indus.j@gmail.com)
- 3Indian Institute of Technology Bombay, Center of Studies in Resources Engineering (CSRE), Mumbai, India (getbiplab@gmail.com)
Solar energy shall be an indispensable part in India’s clean energy transition. As renewable energy requires large amount of space considerations, policy makers often question the land based targets for deploying solar parks. A robust geospatial information on existing solar parks shall be crucial for both the governments and policy makers.
This study presents a novel method to detect solar parks using a synergy of satellite imagery from Sentinel-2 and convolutional neural networks (CNN). For the work, a total of nearly 2000 satellite images from Sentinel-2 were chosen over ten number of solar parks situated in India. Case study results are presented for the solar parks in India namely Bhadla Solar Park, Rajasthan, and Pavagada Solar Park, Karnataka. This dataset measures solar footprint over India and examines environmental impacts of solar parks over nearby ecosystem.
How to cite: Basant, S., Indu, J., and Banerjee, B.: Solar Park Detection Based On Machine Learning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4584, https://doi.org/10.5194/egusphere-egu24-4584, 2024.
