Using AI to Identify Optimal Locations for Floating Solar Panels

Background

With the growing threat of global warming, there is a pressing need to adopt renewable energy solutions. Floating solar panels offer an innovative way to harness solar power by utilizing inland water bodies such as drinking water reservoirs and hydropower reservoirs. However, identifying optimal locations for floating solar installations is challenging due to the need for precise data on water depth and conditions. This project aimed to address that challenge using AI and satellite imagery.

Objective

The goal of the project was to apply remote sensing techniques to satellite imagery to infer the depth of inland water bodies. This information would then be integrated into Glint Solar‘s solar site assessment tool, helping to identify the most suitable locations for floating solar panels and accelerating the transition to green energy.

Approach

To solve this challenge, Omdena’s team of over 30 AI engineers collaborated with Glint Solar. The team focused on three main tasks:

1. Preprocessing the satellite data: The team developed steps to denoise satellite images, including a general pipeline for processing raster data, cloud cover removal using Sentinel 2 Level 2 images, and algal blooms detection with MODIS data.
2. AI Modeling: AI models were built to predict the depth of inland water bodies based on multispectral satellite imagery. The team carefully selected and tested different models to ensure high performance.
3. Integration: The successful AI model and preprocessing steps were integrated into Glint Solar’s solar site assessment tool, enabling better site identification for floating solar installations.

Results and Impact

The project successfully achieved its primary objectives:

• The preprocessing steps significantly improved model performance by removing noise from the satellite data.
• The AI model for predicting the depth of inland water bodies was identified and refined.
• The integration of the model into Glint Solar’s tool provides a powerful new feature for assessing potential floating solar sites.

This tool can now assist in faster identification of viable locations for floating solar panels, optimizing land and water usage for renewable energy production. The solution not only enhances the accuracy of site assessments but also contributes to the broader goal of advancing the green energy revolution.

Future Implications

This project lays the foundation for future developments in the field of floating solar panel installations. The AI models and preprocessing techniques can be expanded to accommodate multiple water bodies simultaneously, improving site assessment accuracy at a larger scale. Additionally, as more satellite data becomes available, the models could be refined and adapted for different environmental conditions, potentially influencing solar energy policies and adoption rates worldwide. Further research and development of the technology could pave the way for more widespread implementation of floating solar panels, especially in regions with limited land availability.