We published a new paper on estimating clear sky energy across the entire skydome. Wide spectrum sky energy is useful for many downstream applications, including PV building energy modeling, panel alignment, rendering, etc. This is a data-driven technique, using regression models on measured sky data. We validated the method with holdout data and with libRadtran, a popular radiative transfer software package. If interested in the project, sky data and processing tools can be found here.
Del Rocco, J., Bourke, P. D., Patterson, C. B., & Kider, J. T. (2020). Real-time spectral radiance estimation of hemispherical clear skies with machine learned regression models. In Solar Energy (Vol. 204, p. 48-63). International Solar Energy Society (ISES). (DOI: 10.1016/j.solener.2020.04.006)