浙江电力

2019, v.38;No.281(09) 100-104

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

基于多源信息的短期负荷混合预测模型应用研究
Study on Application of Short-term Hybrid Load Forecasting Model Based on Multi-source Information

应张驰,陈淑萍,卢旭航
YING Zhangchi,CHEN Shuping,LU Xuhang

摘要(Abstract):

针对单一负荷预测模型在实时预测中的局限性,建立了以ARIMA(自回归移动平均)模型和人工神经网络模型为基础的混合预测模型。除历史负荷数据信息以外,引入气象数据、用户特征、日期信息等多源信息,探究各类特征值选择对预测结果的影响。某市工业用户的实际应用效果表明,综合ARIMA模型对时间序列趋势的拟合能力和人工神经网络对多源信息的捕捉能力后,建立的混合预测模型具有较好的预测效果。
Aiming at the limitation of single load forecasting model in real-time forecasting, this paper establishes a hybrid forecasting model based on ARIMA(autoregressive integrated moving average) model and artificial neural network model. In addition to historical load data information, it introduces multi-source information such as meteorological data, user features and date information to explore the impact of various feature selections on forecasting results. The application of the forecasting model in an industrial user shows that the ARIMA model that integrates the fitting capacity of time series trend and the multi-source information capture capacity of the artificial neural network, the established hybrid forecasting model turns out to be effective.

关键词(KeyWords): 负荷预测;ARIMA模型;人工神经网络;混合预测
load forecasting;ARIMA model;artificial neural network;hybrid forecasting

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 应张驰,陈淑萍,卢旭航
YING Zhangchi,CHEN Shuping,LU Xuhang

DOI: 10.19585/j.zjdl.201909017

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