母线负荷预测中样本数据预处理的新方法.pdfVIP

第 34 卷 第 2 期 电 网 技 术 Vol. 34 No. 2 2010 年 2 月 Power System Technology Feb. 2010 文章编号：1000-3673 （2010 ）02-0149-06 中图分类号：TM 715 文献标志码：A 学科代码：470·4051 母线负荷预测中样本数据预处理的新方法 李光珍，刘文颖，云会周，高玉华 （华北电力大学电气与电子工程学院，北京市 昌平区 102206 ） A New Data Preprocessing Method for Bus Load Forecasting LI Guang-zhen, LIU Wen-ying, YUN Hui-zhou, GAO Yu-hua (School of Electrical and Electronic Engineering, North China Electric Power University, Changping District, Beijing 102206, China) ABSTRACT: Selecting the measured data from SCADA as 要手段，对提高电网负荷预测水平和电网经济运行 original data source, a data pre-processing method for sample 水平具有重要意义。 data is proposed to deal with three kinds of typical abnormal 母线负荷短期预测主要有 2 类方法：一类是基 data in bus load data. The improved horizontal comparison is 于系统负荷分配的预测方法[1-2] 。此类方法先进行系 used to recognize and modify the points where data is lost and 统短期负荷预测，然后根据各母线负荷的分配系数 the abnormal catastrophe points of data caused by sudden event; and then the data fluctuation due to channel noise and other 生成母线预测负荷。预测精度不仅取决于分配系 reasons are dealt with by db4 wavelet threshold denoising 数，还与系统预测负荷紧密相关。受分配系数误差 method to smooth the load curve. The proposed method can 和系统负荷预测误差的双重影响，这类方法难以达 effectively recognize continuous points where data is lost and 到理想的预测精度。另一类是采用母线负荷的历史 abnormal catastrophe points of data, and on the basis of 数据、负荷特性以及相应的影响因素直接进行母线 keeping variation trend of original load curve the abnormal 负荷预测[3-12] 。此类方法的样本数据来源主要有 fluctuate data is rejected, thus the smoothing processing is 2 种：SCADA 系统采集的量测数据和状态估计结 implemented and