|
|||||
|
|
| 基于粗糙集理论和动态时序模型的日负荷曲线预测新方法 | |
| 引用本文： | 谢宏,程浩忠,张国立,牛东晓,杨文璐.基于粗糙集理论和动态时序模型的日负荷曲线预测新方法[J].电网技术,2004,28(14):10-14. |
| 作者姓名： | 谢宏 程浩忠 张国立 牛东晓 杨文璐 |
| 作者单位： | 1. 上海海运学院电子工程系,上海,200135 2. 上海交通大学电气工程系,上海,200030 3. 华北电力大学信息系,河北省,保定市,071003 |
| 摘 要： | 作者提出了一种短期日负荷曲线预测新方法.该方法首先采用日最小负荷对日负荷曲线进行规范化,再将日负荷曲线预测转化为对日最小负荷的预测和对日规范化负荷曲线的预测.对日最小负荷预测应用动态时序模型;对日规范负荷曲线应用专家系统进行推理预测,专家系统中的推理规则应用粗糙集理论从历史数据中获取.采用上海电网数据对该预测方法进行了测试,结果表明该方法便于对各种影响因素进行分析处理,能够更有效地利用历史数据所包含的信息. |
| 关 键 词： | 日负荷曲线预测 粗糙集 聚类 动态时序模型 |
| 文章编号： | 1000-3673(2004)14-0010-05 |
| 修稿时间： | 2004年3月29日 |
A NEW METHOD TO FORECAST DAILY LOAD CURVE BASED ON ROUGH SET THEORY AND DYNAMIC TIME SERIES MODEL |
|
| XIE Hong,CHENG Hao-zhong,ZHANG Guo-li,NIU Dong-xiao,YANG Wen-lu Dept. of Electronic Engineering,Shanghai Maritime University,Shanghai ,China Dept. of Electrical Engineering,Shanghai Jiaotong University,Shanghai ,China Dept. of Computation Science & Information,North China Electric Power University,Baoding ,Hebei Province,China.A NEW METHOD TO FORECAST DAILY LOAD CURVE BASED ON ROUGH SET THEORY AND DYNAMIC TIME SERIES MODEL[J].Power System Technology,2004,28(14):10-14. | |
| Authors: | XIE Hong CHENG Hao-zhong ZHANG Guo-li NIU Dong-xiao YANG Wen-lu Dept of Electronic Engineering Shanghai Maritime University Shanghai China Dept of Electrical Engineering Shanghai Jiaotong University Shanghai China Dept of Computation Science & Information North China Electric Power University Baoding Hebei Province China |
| Affiliation: | XIE Hong,CHENG Hao-zhong,ZHANG Guo-li,NIU Dong-xiao,YANG Wen-lu Dept. of Electronic Engineering,Shanghai Maritime University,Shanghai 200135,China Dept. of Electrical Engineering,Shanghai Jiaotong University,Shanghai 200030,China Dept. of Computation Science & Information,North China Electric Power University,Baoding 071003,Hebei Province,China |
| Abstract: | A new forecasting method for short-term daily load curve is presented. In this method, at first the daily load curve is normalized by daily load minima, then the daily load curve forecasting is transformed into daily load minima forecasting and normalized daily load curve forecasting. The dynamic time series model is applied to the forecasting of daily load minima, while an expert system, in which the inference rules are acquired by rough set theory from history data, is applied to the reasoning forecasting of normalized daily loads. The presented method is tested by the data of Shanghai power grid, the results show that the presented method could analyze and deal with the impact factors easily, and utilize the information involved in historic data more effectively. |
| Keywords: | Daily load curv forecasting Rough set Cluster Dynamic time series model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |