In the face of global warming, many countries are transitioning towards solar energy. To effectively deploy photovoltaic panels, ground-based stations provide solar energy professionals with Global Horizontal Irradiance (GHI) data, enabling solar energy output predictions. Despite the maturity of solar technologies, however, the high operational costs of these monitoring stations limit their adaptation. This is especially true in developing countries, which benefit the most from utilizing sustainable energies.
In response to the low availability of solar radiation estimates, researchers are increasingly exploring the use of data from weather satellites. Ruiz-Munoz and Hoyos-Gómez used machine learning models to improve the accuracy of satellite data estimates. Specifically, they sourced satellite data from the National Solar Radiation Database (NSRDB), a public dataset, and calibrated it against GHI data derived from Automatic Weather Stations.
“By demonstrating the effectiveness of machine learning models in site-adaptation procedures, our work could lead to more reliable solar energy assessments in underrepresented regions with complex climates,” said author Laura Hoyos-Gómez. “This has implications not only for improving the design and deployment of solar energy systems but also for promoting energy equity in developing regions.”
Results of the machine learning model indicate a successful integration of satellite and ground-measured data, with enhanced accuracy of solar radiation estimates and a reduction of bias caused by climate conditions.
Building on this work, the authors intend to improve the accessibility of combined renewable energy sources.
“Our next step is to use this data to create solar radiation potential maps and conduct energy complementarity analyses with other renewable energy sources, such as wind,” said Hoyos-Gómez.
Source: “Accurate solar radiation site adaptation: Harnessing satellite data and in situ measurements,” by Jose F. Ruiz-Munoz and Laura S. Hoyos-Gómez, Journal of Renewable and Sustainable Energy (2024). The article can be accessed at https://doi.org/10.1063/5.0226782.