Sub-minute probabilistic solar forecasting for real-time stochastic simulations

Dazhi Yang, Gokhan Mert Yagli and Dipti Srinivasan

Renewable and Sustainable Energy Reviews, 2022, vol. 153, issue C

Abstract: Simulation needs to reflect reality, otherwise, it yields misleading results that are potentially harmful to the operators and decision makers who rely on that simulation. Current stochastic simulation methods for solar energy systems in smart grids mostly consider scenarios generated from a single low-frequency time series, which is unable to reflect the high-frequency fluctuation and changing uncertainty in the actual solar power output. To that end, this paper introduces a state-of-the-art probabilistic solar forecasting method, that provides remedies to the drawbacks, and thus benefits real-time stochastic simulations at large. Lasso-penalized quantile regression is paired with an analog-based preselection algorithm, and is used to forecast the irradiance over a 1 km by 1 km area, in sub-minute timescales. To capture the changing weather condition, the forecasting method uses online training, and updates its parameters and forecasts every few seconds. Despite the heavy computation required, the forecasting method is fast; it takes less than 1 s to complete a forecasting cycle. Through five benchmarking methods, ranging from the naïve climatology and persistence to the state-of-the-art analog ensemble (AnEn), the proposed method is shown to be able to attain an exceptionally high forecast skill in terms of pinball loss (up to a skill score of 55%, with AnEn being the reference model), which is unparalleled by all previous works that used the same dataset. To promote the future uptake of the method, the R code and the final forecast datasets are released on Github.

Keywords: Solar forecasting; Stochastic simulations; Lasso; Analog ensemble; Probabilistic forecasting (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (6)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1364032121010078
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:rensus:v:153:y:2022:i:c:s1364032121010078

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/600126/bibliographic
http://www.elsevier. ... 600126/bibliographic

DOI: 10.1016/j.rser.2021.111736

Access Statistics for this article

Renewable and Sustainable Energy Reviews is currently edited by L. Kazmerski

More articles in Renewable and Sustainable Energy Reviews from Elsevier
Bibliographic data for series maintained by Catherine Liu ().