A method for cloud detection and opacity classification based on ground based sky imagery

Abstract

Abstract. Digital images of the sky obtained using a total sky imager (TSI) are classified pixel by pixel into clear sky, optically thin and optically thick clouds. A new classification algorithm was developed that compares the pixel red-blue ratio (RBR) to the RBR of a clear sky library (CSL) generated from images captured on clear days. The difference, rather than the ratio, between pixel RBR and CSL RBR resulted in more accurate cloud classification. High correlation between TSI image RBR and aerosol optical depth (AOD) measured by an AERONET photometer was observed and motivated the addition of a haze correction factor (HCF) to the classification model to account for variations in AOD. Thresholds for clear and thick clouds were chosen based on a training image set and validated with set of manually annotated images. Misclassifications of clear and thick clouds into the opposite category were less than 1 %. Thin clouds were classified with an accuracy of 60 %. Accurate cloud detection and opacity classification techniques will improve the accuracy of short-term solar power forecasting.

Coupling sky images with radiative transfer models: a new method to estimate cloud optical depth

“…These approaches have led to improved accuracy of cloud detection, yet limited progress has been made towards understanding the phenomena that influence the performance of these methods. Although a direct relationship with aerosol optical depth (τ a ) and RBR is observed for small τ a , (τ a < 0.3) (Ghonima et al, 2012) no direct relationship has been found between RBR, or other variables determined from sky imagers, and larger optical depths (τ > 0.3) such as those found typically in clouds. This has limited sky imager cloud detection to a binary classification in which the image is segmented into cloud or clear sky.…”

“…However, Ghonima et al (2012) found minimal differences in performance between RBD and RBR retrieval with RBR outperforming RBD. Gauchet et al (2012) used RBD combined with a different approach to account for the directional effects in cloud detection, in which they segmented images into five zones, solar disk, circumsolar disk, extended circumsolar disk, main zone, sky horizon, and orographic horizon.…”

“…To estimate a cloud's optical depth (τ c ), the most advanced methods separate the image into clear sky, thin cloud, and thick cloud and assign a τ c to each of these groups. To distinguish thin and thick clouds, the red-blue ratio (RBR) (or a function of RBR) has been used as the default method (Koehler et al, 1991;Shields et al, 1993;Chow et al, 2011;Ghonima et al, 2012;Roy et al, 2001). It is defined as the ratio of the signal from the red channel to the signal from the blue channel.…”

Sky camera geometric calibration using solar observations

“…In an operational setting, an unobstructed image of the sun is less likely to occur at large zenith angles because of the longer optical path through the troposphere (Warren et al, 1986). A cloud detection process (e.g., Ghonima et al, 2012) can be used to discard images with significant cloud cover.…”

“…Forecasting of renewable power generation (e.g., Monteiro et al, 2009;Perez et al, 2010;Kleissl, 2013) enables more economical and reliable scheduling and dispatch of all generation resources, including renewables, which in turn accommodates a larger amount of variable supply on the electricity grid. Specifically for solar power forecasting, a number of technologies are being applied: numerical weather prediction (e.g., Lorenz et al, 2009;Mathiesen and Kleissl, 2011;; satellite image-based forecasting (e.g., Hammer et al, 1999;Perez and Hoff, 2013); and stochastic learning methods (e.g., Bacher et al, 2009;Marquez and Coimbra, 2011;Pedro and Coimbra, 2012). For very short term (15 min ahead) solar power forecasting on the kilometer scale, sky imaging from ground stations has demonstrated utility (Chow et al, 2011;Urquhart et al, 2013;Marquez and Coimbra, 2013;Yang et al, 2014).…”