Twenty-Four-Hour Ahead Probabilistic Global Horizontal Irradiance Forecasting Using Gaussian Process Regression
نویسندگان
چکیده
Probabilistic solar power forecasting has been critical in Southern Africa because of major shortages due to climatic changes and other factors over the past decade. This paper discusses Gaussian process regression (GPR) coupled with core vector for short-term hourly global horizontal irradiance (GHI) forecasting. GPR is a powerful Bayesian non-parametric method that works well small data sets quantifies uncertainty predictions. The choice kernel characterises covariance function crucial issue regression. In this study, we adopt minimum enclosing ball (MEB) technique. MEB improves smaller is, shorter training time, hence performance robust. Forecasting real-time was done on two South African radiometric stations, Stellenbosch University (SUN) coastal area Western Cape Province, Venda (UNV) station Limpopo Province. Variables were selected using least absolute shrinkage selection operator via hierarchical interactions. approach informative priors used parameter estimation. Based root mean square error, error percentage bias results showed model gives most accurate predictions compared those from gradient boosting support models, making study useful tool decision-makers system operators utility companies. main contribution use methodology which GHI data. first application applied data, best our knowledge.
منابع مشابه
Monthly streamflow forecasting using Gaussian Process Regression
Bureau of Economic Geology, Jackson School of Geosciences, University of Texas Austin, Austin, TX 78713, United States Department of Civil, Environmental, and Construction Engineering, University of Central Florida, Orlando, FL 32816, United States Key Laboratory for Agro-Ecological Processes in Subtropical Region, Institute of Subtropical, Agriculture, Chinese Academy of Sciences, Changsha, Ch...
متن کاملProbabilistic Solar Forecasting Using Quantile Regression Models
In this work, we assess the performance of three probabilistic models for intra-day solar forecasting. More precisely, a linear quantile regression method is used to build three models for generating 1 h–6 h-ahead probabilistic forecasts. Our approach is applied to forecasting solar irradiance at a site experiencing highly variable sky conditions using the historical ground observations of sola...
متن کاملSpatial Estimation of Sub-Hour Global Horizontal Irradiance Based on Official Observations and Remote Sensors
This study was motivated by the need to improve densification of Global Horizontal Irradiance (GHI) observations, increasing the number of surface weather stations that observe it, using sensors with a sub-hour periodicity and examining the methods of spatial GHI estimation (by interpolation) with that periodicity in other locations. The aim of the present research project is to analyze the goo...
متن کاملGaussian process regression for forecasting battery state of health
Gaussian process (GP) regression used for forecasting battery state of
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Algorithms
سال: 2021
ISSN: ['1999-4893']
DOI: https://doi.org/10.3390/a14060177