Solar forecasting model focuses on shorter time scales to address irradiance variability
DOI: 10.1063/10.0005503
Solar forecasting model focuses on shorter time scales to address irradiance variability lead image
Solar irradiance is highly variable, and this variability is smoothed out by collector arrays. Adding many point sensors would enable better variability monitoring at solar power plants, but this is cost prohibitive.
Researchers are turning to prediction models, which incorporate a small number of point sensors, to describe variability. Most models focus on irradiance reduction over longer periods of time, averaging out effects of cloud motion.
Joseph Ranalli and Esther Peerlings developed a model to investigate how a cloud’s shadow moving in one direction across a solar power plant affects variability on short time scales.
Based on a Fourier transform method, which turns a time signal into the frequency domain, the model showed the solar facility acts as a low-pass filter, limiting the effects of swiftly moving smaller clouds while reacting to larger ones that affect the plant for longer time periods. High frequencies represent small clouds and low frequencies represent larger ones.
Since the power plant serves as a filter, different panel configurations will have different impacts on variability.
“Because our model is about looking closely at periods where cloud motion is constant, it predicts how the distributed irradiance is sensitive to changes in the plant shape along the direction of that motion,” Ranalli said. “Our results show models that resolve current cloud motion conditions yield better short-term results than those developed to match long-term average conditions.”
The researchers plan to scale up their study by looking at longer time windows and different directions of cloud motion to more fully understand how power plant filtering affects the irradiance for improved forecasting capability.
Source: “Cloud advection model of solar irradiance smoothing by spatial aggregation,” by Joseph Ranalli and Esther E. M. Peerlings, Journal of Renewable and Sustainable Energy (2021). The article can be accessed at https://doi.org/10.1063/5.0050428 .
