Spatial distribution of CMA stations. A total of 2,453 blue circles represent routine weather stations, which have sunshine duration measurements. Seventeen red rhombuses represent radiation stations that have Rdir and Rdif observations.
Newswise — In a new study (DOI: 10.34133/remotesensing.0111) published in the Journal of Remote Sensing in February 2024, researchers utilized data augmentation alongside the LightGBM machine learning model for the estimation of both diffuse and direct solar radiation. By leveraging sunshine duration data collected from over 2,453 weather stations throughout China, this research overcomes the limitations posed by sparse and unevenly distributed ground-based observations.
This approach ingeniously utilizes sunshine duration data gathered from over 2,453 weather stations, effectively bypassing the traditional obstacles of sparse and irregularly distributed ground-based observations. The core of this research lies in its novel application of machine learning algorithms, which are trained on augmented datasets to predict solar radiation components with unprecedented accuracy. The methodology is particularly groundbreaking because it does not rely on local ground truth data for calibration, making it a universally applicable solution. The validation of this model against independent datasets not only confirmed its effectiveness within China but also indicated its potential for global application. Moreover, the creation of a new satellite-based dataset as a result of this study stands out for its superior accuracy over existing datasets, providing a detailed spatial distribution of solar radiation components. This dataset is instrumental for advancing solar energy research and deployment, offering insights that can lead to more efficient and optimized solar energy production.
Professor Kun Yang, the lead researcher from Tsinghua University, stated, "Our method significantly enhances the accuracy and applicability of solar radiation component estimates, paving the way for optimized solar energy utilization across China and potentially worldwide."
This innovative approach not only establishes a new standard for estimating solar radiation but also presents a globally scalable solution, signaling a groundbreaking shift in solar energy research and implementation. The newly developed satellite-based dataset excels in precision over prior datasets and delivers an exhaustive spatial analysis of solar radiation components. This advancement is vital for the solar energy sector, enabling more strategic site selection and system optimization, especially in areas with high solar energy potential.
###
References
DOI
Original Source URL
https://doi.org/10.34133/ remotesensing.0111
Funding information
This work was supported by the Sustainable Development International Cooperation Program of National Science Foundation of China (Grant No. 42361144875) and the National Natural Science Foundation of China (Grant No. 42171360).
About Journal of Remote Sensing
The Journal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.
RSS