| Work Detail |
A team of researchers in the United States has created a novel method to integrate raw sky images and measurements of global solar irradiance, current solar forecast, and intra-hour forecast. The methodology uses low-cost radiometric IR cameras instead of expensive ceilometers.
A US research group has created a solar forecasting methodology that uses a combination of infrared (IR) images and measurements of global solar irradiance.
Scientists say the novel approach is capable of improving real-time solar forecasting and intra-hour forecasting, while being applicable to real-time photovoltaic markets and optimizing power dispatch in microgrids.
“The sky imager is more expensive than regular visible-light all-sky imagers, but it can also approximate cloud heights, making it a low-cost alternative to a ceilometer,” he explained. to pv magazine Guillermo Terrén-Serrano, corresponding author of the research, pointing out that the method is suitable for photovoltaic systems of any size. “Ceilometers cost about $20,000, while our method costs less than $2,000. Our system includes a radiometric infrared camera, data logger, high-resolution sun tracker, pyranometer, outdoor computer, weatherproof housing, visible light fisheye, weather sensors and camera lenses. ”
Visible light cameras are often used to image the sky from the ground, helping PV models react to cloudy conditions. However, the sun saturates the pixels of those cameras, destroying information that could increase the performance of a solar forecast. Therefore, IR cameras are used alternatively, which reduce the saturation of the sun.
However, IR-based forecasts have their own problems, such as lower signal-to-noise ratio, among others. This is partly due to solar irradiance, which can distort images in some conditions. “This research introduces effective data processing methods to eliminate the deterministic component of global solar irradiance in pyranometer measurements and infrared images,” the article explains.
To eliminate the effect of irradiance, the novel method first uses machine learning to identify biases that could affect the clear sky index (CSI). Because CSI quantifies the effects of clouds on global solar irradiance (GSI), more accurate results in the first measurement translate into more accurate results in the second.
Another algorithm is then used to classify the sky conditions of the IR images into four: clear sky, cumulus, stratus, and nimbus. From this classification, the algorithm further interacts with the GSI data and calculates the effect of irradiance on the image, effectively clearing it for forecasting.
Additionally, the algorithm eliminates the effect of dirt on the camera. “This research assumes that a sky imager will not be cleaned daily during operation,” the research group explains. “For this situation, a method based on image processing is proposed to remove the radiation emitted by dirt on the outer window of the germanium chamber from the IR images.”
The algorithms were trained and tested with data from Albuquerque (New Mexico, United States), which has an arid semi-continental climate, with minimal rainfall. “Future research is required to develop a global model valid for any location,” they emphasize.
The researchers conclude that the proposed method is efficient and state that low-cost radiometric IR cameras can potentially be a substitute for expensive ceilometers in the future.
“Proper data processing reduces the complexity of the learning algorithm when implemented in the solar prediction application,” they say. “Reducing complexity increases the accuracy of the prediction and reduces the calculation time needed to make it. “This is especially important in real-time applications, such as real-time forecasting and intra-hour forecasting of solar energy.”
The results are presented in the study “ Processing of global solar irradiance and ground-based infrared sky images for solar nowcasting and intra-hour forecasting applications ” and intra-hourly), recently published in Solar Energy . The researchers are from the University of California at Santa Barbara and the University of New Mexico. |
