Proceedings of the KIEE Conference (대한전기학회:학술대회논문집)
- 1993.11a
- /
- Pages.29-31
- /
- 1993
Short-term Load Forecasting using Neural Network
신경회로망을 이용한 단기부하예측
- Koh, Hee-Soek (Kyungnam University) ;
- Lee, Chung-Sik (Kyungnam University) ;
- Kim, Hyun-Deuk (Kyungnam University) ;
- Lee, Hee-Chul (Kyungnam University)
- Published : 1993.11.26
Abstract
This paper presents Neural Network(NN) approach to short-term load forecasting. Input to the NN are past loads and the output is the predicted load for a given day. The NN is used to learn the relationship among past, current and future temperature and loads. Three different cases are presented. Case 1 divides into weekday and weekendday load pattern. Case 2 forcasts 24-hour ahead load. Case 3 searchs for the same load pattern as present load pattern in past load pattern. From result of forecasting, an average absolute percentage errors of case 1 shows 2.0%. That of case 2 shows 2.2, and That of case 3 shows 1.6%.
Keywords