计算机工程 ›› 2008, Vol. 34 ›› Issue (18): 226-227.doi: 10.3969/j.issn.1000-3428.2008.18.081

• 人工智能及识别技术 • 上一篇    下一篇

基于BP网的中长期负荷预测因素优化选择

朱继萍,戴 君   

  1. (西安文理学院机械电子工程系,西安 710065)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-09-20 发布日期:2008-09-20

Optimization Selection of Correlative Factors for Medium and Long Term Load Forecasting Based on BP Network

ZHU Ji-ping, DAI Jun   

  1. (Department of Mechanical and Electronic Engineering, Xi’an University of Arts and Science, Xi’an 710065)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-09-20 Published:2008-09-20

摘要: 基于人工神经网络原理,设计一个由输入层、隐含层和输出层组成的三层BP网络模型,利用神经网络高度非线性建模能力,选取影响电力负荷的一些经济因素作为BP人工神经网络的输入变量,采用新定义的方差贡献法对输入变量进行优化选择,对预测精度的影响进行探讨。仿真结果证明,采用方差贡献法对影响中长期电力负荷预测的相关因素进行优化选择是可行有效的。

关键词: BP网络, 中长期负荷预测, 方差贡献法, 优化选择

Abstract: Based on the theory of Artificial Neural Network(ANN), a three-layer Back Propagation(BP) network is proposed. The idea is to forecast medium and long term load by using the ability of ANN to nonlinear system factors. Some economic factors are selected as inputs for the BP ANN model. Variance contribution method is used for the optimization selection of correlative factors, and forecasting accuracy is discussed. Simulation results show that the method is feasible.

Key words: Back Propagation(BP) network, medium and long term load forecasting, variance contribution method, optimization selection

中图分类号: