摘要
为了解决传统短期用电负荷预测系统存在响应时间慢、预测精度差的问题,设计了一种基于梯度提升树的短期用电负荷预测系统。该系统框架采用C/S架构模式搭建,根据预测需求选择系统的组成硬件,并以梯度提升树为核心,建立预测模型,完成系统软件及短期用电负荷预测系统的设计。实验结果表明,与基于神经网络、数据挖掘、支持向量机的三种传统用电负荷预测系统相比,本系统运行下,响应时间缩短,预测精度提高,为电力企业电量生产和供应提供了可靠的依据。
In order to solve the problems of slow response time and poor prediction accuracy in traditional short⁃term load forecasting system,a short⁃term load forecasting system based on gradient lifting tree is designed.The system uses C/S architecture to build the system framework,selects the system hardware according to the prediction demand,builds the prediction model,completes the system software design,and completes the design of the short⁃term power load prediction system with the gradient lifting tree as the core.The experimental results show that compared with the three traditional power load forecasting systems based on neural network,data mining and support vector machine,the response time of the system is shortened and the prediction accuracy is improved,which provides a reliable basis for power production and supply of power enterprises.
作者
曹敏
CAO Min(State Grid Shaanxi Electric Power Company,Xi’an 710048,China)
出处
《电子设计工程》
2020年第22期16-19,24,共5页
Electronic Design Engineering
基金
国家自然科学基金重点项目(61433004)。
关键词
梯度提升树
短期用电负荷
用电负荷预测
系统设计
gradient lifting tree
short⁃term power load
power load prediction
system design
