摘要
超短期配变负荷预测结果是配电网状态估计中伪量测的主要来源之一,其利用的历史数据主要来源于自动计量系统。伪量测的精度会显著影响状态估计的结果。给出了一种基于鲁棒Holt-Winter模型的超短期负荷预测方法。该方法综合考虑线性趋势、季节变动和随机波动的时间序列特性,并与指数平滑法相结合,具有良好的预测能力。此外,该方法可以自动识别和修正坏数据。以某配电网的实际数据进行了算例分析,结果表明该方法具有良好的预测精度,且具有自动压缩坏数据影响的特性。
The ultra-short term load forecasting results can be used as the pseudo measurements for state estimation in distribution network, in which the historical load data are collected by automatic meter reading system. And the prediction precision significantly influences the state estimation. This paper presents a robust ultra-short term load forecasting method which is based on Holt-winter model. The method considers a linear trend, seasonal change and random fluctuations in the time sequence, and the exponential smoothing method is also combined. The proposed method can automatically identify and correct bad data in historical data. The proposed method has been tested for a real distribution networks and the results show this method has good prediction precision and has the ability of automatic compression influence of bad data.
出处
《电网技术》
EI
CSCD
北大核心
2014年第10期2810-2815,共6页
Power System Technology
基金
国家科技支撑计划项目(2013BAA01B01)
黔科重大专项(字[2012]6022)~~