基于大数据技术的多变量短期电力需求预测研究
    点此下载全文
引用本文:袁小凯,李果,黄世平.基于大数据技术的多变量短期电力需求预测研究[J].电网与清洁能源,2020,36(12):30~34
摘要点击次数: 101
全文下载次数: 21
作者单位
袁小凯 南方电网科学研究院 
李果 南方电网科学研究院 
黄世平 南方电网科学研究院 
基金项目:广东经研院咨询科技项目(SGNXJY30GHKL 1902560)
中文摘要:针对传统多变量短期电力需求预测方法没有归一化处理电力数据,导致预测性能较差、精度较低,提出基于大数据技术的多变量短期电力需求预测方法。在电网大数据框架中,连接MONGOOSE数据库引擎与短期电子服务器,完成大数据技术支持下的短期电力环境搭建。基于大数据技术,通过确定神经预测网络层数的方式,实现电力需求数据的归一化处理,根据多变量短期预测误差的计算结果,实现基于大数据技术多变量短期电力需求预测方法的应用。实验结果表明,研究方法的电力需求预测有效性更好,预测精度更高。
中文关键词:大数据技术  多变量需求  短期电力预测  数据库引擎  电子服务器  预测神经网络  电力需求数据  预测误差
 
Multi-Variable Short-Term Power Demand Forecasting Research Based on Big Data Technology
Abstract:In view of the fact that the traditional multivariable short-term power demand forecasting method does not normalize the power data, which results in poor forecasting performance and low accuracy, a multivariable short-term power demand forecasting method based on big data technology is proposed. In the framework of grid big data, MONGOOSE database engine and short-term electronic server are connected to complete the construction of short-term electric environment supported by big data technology. Based on big data technology, the normalized processing of power demand data is realized by determining the number of neural prediction network layers. According to the calculation result of multivariable short-term prediction error, the application of multivariable short-term power demand prediction method based on big data technology is realized. The experimental results show that the proposed method is more effective and accurate in power demand forecasting.
keywords:big data technology  multivariate demand  short-term power forecast  database engine  electronic server  predictive neural network  electricity demand data  prediction error
查看全文  查看/发表评论  下载PDF阅读器
    《电网与清洁能源》杂志

期卷浏览

关键词检索

最新公告栏

您是第2463792位访问者    Email: psce_sn@163.com
版权所有:电网与清洁能源网 陕ICP备16002958号-1
本系统由北京勤云科技发展有限公司设计