内蒙古电力技术

2021, (04) 37-41

[打印本页] [关闭]
本期目录(Current Issue) | 过刊浏览(Archive) | 高级检索(Advanced Search)

基于主成分分析和正则化神经网络的电力负荷预测方法
Power Load Forecasting Method Based on Principal Component Analysis and Regularized Neural Network

徐江,冯雪,付兆庆,马宝明,苗耀庭
XU Jiang,FENG Xue,FU Zhaoqing,MA Baoming,MIAO Yaoting

摘要(Abstract):

为提高用电负荷预测的准确率,提出一种基于主成分分析(PCA)和正则化人工神经网络负荷预测方法。通过分析多个影响负荷变化的变量和因子,利用主成分分析法对变量进行线性降维,减少各变量之间的关联性,用得到的主成分作为BP神经网络的输入变量,同时增加正则化约束项优化神经网络训练过程,提高网络模型泛化能力。最后利用某售电公司所代理用户的实测用电数据进行网络建模和预测,结果表明,与其他预测方法进行对比,该预测方法有较高的准确性。
In order to improve the accuracy of electrical load forecasting, this paper proposes a load forecasting method based on principal component analysis(PCA) and regularized artificial neural network. By analyzing multiple variables and factors that affect the power load change, the principal component analysis method is used to linearly reduce the dimension of the variables to reduce the correlation among the variables. The obtained principal components are used as the input variables of the BP neural network. At the same time, the regularization constraints are added to optimize the training process of the neural network and improve the generalization ability of the network model. Finally, the network modeling and prediction are carried out by using the measured electricity consumption data of a power sales company. Compared with other prediction methods, the results show that the prediction method is of higher accuracy.

关键词(KeyWords): 负荷预测;主成分分析;正则化;神经网络;电力市场交易
load forecasting;principal component analysis;regularization;neural network;electricity market trading

Abstract:

Keywords:

基金项目(Foundation): 内蒙古电力(集团)有限责任公司科技项目“电力多边市场大数据挖掘研究及市场信息可视化展示平台开发”(2018-84)

作者(Author): 徐江,冯雪,付兆庆,马宝明,苗耀庭
XU Jiang,FENG Xue,FU Zhaoqing,MA Baoming,MIAO Yaoting

参考文献(References):

扩展功能
本文信息
服务与反馈
本文关键词相关文章
本文作者相关文章
中国知网
分享