You need JavaScript to view this

Unsupervised/supervised learning concept for 24-hour load forecasting

Abstract

An application of artificial neural networks in short-term load forecasting is described. An algorithm using an unsupervised/supervised learning concept and historical relationship between the load and temperature for a given season, day type and hour of the day to forecast hourly electric load with a lead time of 24 hours is proposed. An additional approach using functional link net, temperature variables, average load and last one-hour load of previous day is introduced and compared with the ANN model with one hidden layer load forecast. In spite of limited available weather variables (maximum, minimum and average temperature for the day) quite acceptable results have been achieved. The 24-hour-ahead forecast errors (absolute average) ranged from 2.78% for Saturdays and 3.12% for working days to 3.54% for Sundays. (Author)
Authors:
Djukanovic, M; [1]  Babic, B; [2]  Sobajic, D J; Pao, Y -H [3] 
  1. Electrical Engineering Inst. 'Nikola Tesla', Belgrade (Yugoslavia)
  2. Electrical Power Industry of Serbia, Belgrade (Yugoslavia)
  3. Case Western Reserve Univ., Cleveland, OH (United States). Dept. of Electrical Engineering and Computer Science
Publication Date:
Jul 01, 1993
Product Type:
Journal Article
Reference Number:
GB-94-050108; EDB-94-044805
Resource Relation:
Journal Name: IEE Proceedings, Part C: Generation, Transmission and Distribution (Institution of Electrical Engineers); (United Kingdom); Journal Volume: 140:4
Subject:
24 POWER TRANSMISSION AND DISTRIBUTION; 29 ENERGY PLANNING, POLICY AND ECONOMY; LOAD MANAGEMENT; NEURAL NETWORKS; POWER SYSTEMS; ALGORITHMS; EXPERT SYSTEMS; FORECASTING; WEATHER; ENERGY SYSTEMS; MANAGEMENT; MATHEMATICAL LOGIC; 240100* - Power Systems- (1990-); 296000 - Energy Planning & Policy- Electric Power
OSTI ID:
5445635
Country of Origin:
United Kingdom
Language:
English
Other Identifying Numbers:
Journal ID: ISSN 0143-7046; CODEN: IPPDDA
Submitting Site:
GB
Size:
Pages: 311-318
Announcement Date:
Apr 01, 1994

Citation Formats

Djukanovic, M, Babic, B, Sobajic, D J, and Pao, Y -H. Unsupervised/supervised learning concept for 24-hour load forecasting. United Kingdom: N. p., 1993. Web.
Djukanovic, M, Babic, B, Sobajic, D J, & Pao, Y -H. Unsupervised/supervised learning concept for 24-hour load forecasting. United Kingdom.
Djukanovic, M, Babic, B, Sobajic, D J, and Pao, Y -H. 1993. "Unsupervised/supervised learning concept for 24-hour load forecasting." United Kingdom.
@misc{etde_5445635,
title = {Unsupervised/supervised learning concept for 24-hour load forecasting}
author = {Djukanovic, M, Babic, B, Sobajic, D J, and Pao, Y -H}
abstractNote = {An application of artificial neural networks in short-term load forecasting is described. An algorithm using an unsupervised/supervised learning concept and historical relationship between the load and temperature for a given season, day type and hour of the day to forecast hourly electric load with a lead time of 24 hours is proposed. An additional approach using functional link net, temperature variables, average load and last one-hour load of previous day is introduced and compared with the ANN model with one hidden layer load forecast. In spite of limited available weather variables (maximum, minimum and average temperature for the day) quite acceptable results have been achieved. The 24-hour-ahead forecast errors (absolute average) ranged from 2.78% for Saturdays and 3.12% for working days to 3.54% for Sundays. (Author)}
journal = []
volume = {140:4}
journal type = {AC}
place = {United Kingdom}
year = {1993}
month = {Jul}
}