skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Evaluation of regression and neural network models for solar forecasting over different short-term horizons

Abstract

Forecasting solar irradiation has acquired immense importance in view of the exponential increase in the number of solar pho- tovoltaic (PV) system installations. In this article, analyses results involving statistical and machine-learning techniques to predict solar irradiation for di?erent forecasting horizons are reported. Yearlong typical meteorological year 3 (TMY3) datasets from three cities in the United States with di?erent climatic conditions have been used in this analysis. A simple forecast approach that assumes consecutive days to be identical serves as a baseline model to compare forecasting alternatives. To account for seasonal variability and to capture short-term ?uctuations, di?erent variants of the lagged moving average (LMX) model with cloud cover as the input variable are evaluated. Finally, the proposed LMX model is evaluated against an arti?cial neural network (ANN) model. How the one-hour and 24-hour models can be used in conjunction to predict di?erent short-term rolling horizons is discussed, and this joint application is illustrated for a four-hour rolling horizon forecast scheme. Finally, the e?ect of using predicted cloud cover values, instead of measured ones, on the accuracy of the models is assessed. Results show that LMX models do not degrade in forecast accuracy if models are trained with the forecastmore » cloud cover data.« less

Authors:
 [1];  [2]; ORCiD logo [3]
  1. Arizona State University
  2. Drexel University
  3. BATTELLE (PACIFIC NW LAB)
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1511147
Report Number(s):
PNNL-SA-140536
DOE Contract Number:  
AC05-76RL01830
Resource Type:
Journal Article
Journal Name:
Science and Technology for the Built Environment
Additional Journal Information:
Journal Volume: 24; Journal Issue: 9
Country of Publication:
United States
Language:
English

Citation Formats

Inanlouganji, Alireza, Reddy, Agami, and Katipamula, Srinivas. Evaluation of regression and neural network models for solar forecasting over different short-term horizons. United States: N. p., 2018. Web. doi:10.1080/23744731.2018.1464348.
Inanlouganji, Alireza, Reddy, Agami, & Katipamula, Srinivas. Evaluation of regression and neural network models for solar forecasting over different short-term horizons. United States. doi:10.1080/23744731.2018.1464348.
Inanlouganji, Alireza, Reddy, Agami, and Katipamula, Srinivas. Thu . "Evaluation of regression and neural network models for solar forecasting over different short-term horizons". United States. doi:10.1080/23744731.2018.1464348.
@article{osti_1511147,
title = {Evaluation of regression and neural network models for solar forecasting over different short-term horizons},
author = {Inanlouganji, Alireza and Reddy, Agami and Katipamula, Srinivas},
abstractNote = {Forecasting solar irradiation has acquired immense importance in view of the exponential increase in the number of solar pho- tovoltaic (PV) system installations. In this article, analyses results involving statistical and machine-learning techniques to predict solar irradiation for di?erent forecasting horizons are reported. Yearlong typical meteorological year 3 (TMY3) datasets from three cities in the United States with di?erent climatic conditions have been used in this analysis. A simple forecast approach that assumes consecutive days to be identical serves as a baseline model to compare forecasting alternatives. To account for seasonal variability and to capture short-term ?uctuations, di?erent variants of the lagged moving average (LMX) model with cloud cover as the input variable are evaluated. Finally, the proposed LMX model is evaluated against an arti?cial neural network (ANN) model. How the one-hour and 24-hour models can be used in conjunction to predict di?erent short-term rolling horizons is discussed, and this joint application is illustrated for a four-hour rolling horizon forecast scheme. Finally, the e?ect of using predicted cloud cover values, instead of measured ones, on the accuracy of the models is assessed. Results show that LMX models do not degrade in forecast accuracy if models are trained with the forecast cloud cover data.},
doi = {10.1080/23744731.2018.1464348},
journal = {Science and Technology for the Built Environment},
number = 9,
volume = 24,
place = {United States},
year = {2018},
month = {5}
}