Machine Learning Models Predict Solar Radiation for Solar Power and Crops
Researchers at the University of Córdoba have developed a suite of machine learning models to predict solar radiation, now freely available on GitHub. The study, published in Applied Energy, demonstrates high precision in estimating daily solar radiation values, crucial for solar power plant construction and crop optimisation.
The models, tested in diverse geo-climatic conditions, can be applied worldwide. They include neural networks, tree-based models, and others like Support Vector Machine. The research team used Bayesian Optimization to find optimal parameters, leading to significant improvements when applied to new locations.
The models predict solar radiation using thermal data, which is more affordable than traditional methods. The study, part of the Smarity project funded by Spain's Ministry of Science, was conducted in nine locations across southern Spain and North Carolina, USA.
The University of Córdoba's models, now freely accessible, can aid in solar power plant planning and crop development worldwide. The research, published in Applied Energy, showcases the potential of machine learning in harnessing solar energy efficiently.
Read also:
- Increase in Electric Vehicle Charging Stations Across U.S., But Is It Sufficient?
- Tesla's Semi-Truck enters partnership with Uber Freight, aiming to accelerate the usage of electric trucks.
- The current status of green hydrogen for developing countries following the wave of hype: Assessment of remains
- Rapid Growth in Bio-based Polypropylene Sector Anticipated at a Compound Annual Growth Rate of 26.5% by 2034
