Nanchang, China May 22 - 24, 2009

Nanchang, China May 22 - 24, 2009

WISA 2009

WISA 2009

Second International Symposium on

Web Information Systems and Applications

Second International Symposium on

Web Information Systems and Applications

Home > Table of Contents

 

Proceedings of the 2nd International Symposium on Web Information Systems and Applications (WISA 2009)

Nanchang, China, May 22-24, 2009

Editors: Fei Yu, Jiexian Zeng, and Guangxue Yue

AP Catalog Number: AP-PROC-CS-09CN001

ISBN: 978-952-5726-00-8 (Print), 978-952-5726-01-5 (CD-ROM)

Page(s): 367-370

Multi-Model Sifting Methodology for Medium and Long Term Load Forecasting Based on Evaluation of Trend Comparability

Dongxiao Niu, Lidong Duan, Zhihong Gu, Haitao Lv

Full text: PDF

Abstract

The diversity of models is the key point in load forecasting. To improve the precision of forecasting, we should focus on how to choose the more suitable ones among many models. However, for medium and long term load forecasting, we pay more attention to veracity of trend forecasting. With reference to gray relation analysis, which evaluates the trend comparability between each forecasting curve and actual curve through calculation of relationship degree, a multi-model automatic sifting methodology is presented in this paper based on the procedure for improving the gray relation analysis. Based on that, the integrated optimum model is built by MV method in this paper. The calculation results prove that this method describes medium and long term load trend more exactly. The method has same suitability to short term load forecasting.

Index Terms

load forecasting; trend; gray relation analysis; automatic sifting

Copyright @ 2009 ACADEMY PUBLISHER All rights reserved