Abstract
We introduce a computational framework to forecast cloud index (CI) fields for up to one hour on a spatial domain that covers a city. Such intra-hour CI forecasts are important to produce solar power forecasts of utility scale solar power and distributed rooftop solar. Our method combines a 2D advection model with cloud motion vectors (CMVs) derived from a mesoscale numerical weather prediction (NWP) model and sparse optical flow acting on successive, geostationary satellite images. We use ensemble data assimilation to combine these sources of cloud motion information based on the uncertainty of each data source. Our technique produces forecasts that have similar or lower root mean square error than reference techniques that use only optical flow, NWP CMV fields, or persistence. We describe how the method operates on three representative case studies and present results from 39 cloudy days.
Highlights
Power grid management benefits from accurate predictions of solar power generation
The Root Mean Square Error (RMSE) of the individual ANOC ensemble members are all similar to the RMSE of the forecast based on numerical weather prediction (NWP) winds and the forecast based on dense optical flow for this day
In contrast to Case Study 2, we find that RMSE of the forecast based on NWP winds is higher than RMSE of the forecast based on dense optical flow
Summary
Power grid management benefits from accurate predictions of solar power generation. Load balancing, dispatching reserves, curtailing production, energy storage, and economical trading in energy markets are aided by solar power forecasts on an intra-hour time scale (Kleissl, 2013). We use DA to assimilate CMVs derived from optical flow (Horn and Schunck, 1981; Lucas and Kanade, 1981) applied to successive geostationary satellite images every 15 min and CMV fields derived hourly from a mesoscale NWP model. Advection of satellite-derived cloud properties for intra-hour CI or irradiance field forecasts for solar power applications has been considered in several studies (Kleissl, 2013). These two characteristics of the ANOC model allow us to assimilate CMV data into our ensemble, taking the certainty in each source of data into account This approach is inspired by Lorenzo et al (2017) where DA is used to combine ground sensors with clear-sky index fields derived from geostationary satellite images.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call