摘要
节假日电力系统负荷打破了正常日电力负荷周期性的规律,且节假日负荷样本数据较少,用常规正常日电力负荷模型进行短期预测时,往往效果不佳。为此,提出一种基于卡尔曼滤波预测节假日逐点增长率的电力系统短期负荷预测模型,改善了由于样本数据缺少、预测时间跨度大以及与正常日负荷特性差异较大等原因导致的预测精度不理想的现象。通过对节假日负荷特性进行分析,针对不同类型的节假日建立卡尔曼滤波预测模型,在考虑各类影响负荷变化的外部因素的基础上选择节前相关日,通过预测节假日逐点增长率提高预测精度。将提出的预测模型应用于某市节假日短期负荷预测,得到的结果显示预测精度能够满足实际需要,可为相关电力部门对节假日负荷预测提供一定的参考价值。
The load of power system in holiday breaks the rule of the periodicity of normal daily power load,and the load sample data in holidays are scare,the short-term prediction with conventional models is often ineffective.This paper proposes a short-term power load forecasting model based on Kalman filtering to predict the holiday growth rate by points,which improves the non-ideal prediction accuracy caused by lack of sample data,large prediction time span,large differences from normal day load characteristics and other reasons.This model analyzes holiday load characteristics and builds a Kalman filtering prediction model for different types of holidays.It selects the pre-holiday related days based on various external factors that affect load changes and achieves high accuracy prediction by predicting holiday point-by-point growth rates.The proposed forecasting model is applied to short-term holiday load forecasting in a city,and the results show that the prediction accuracy can meet the actual needs.The study can provide a certain reference value for the related power sector’s holiday load prediction.
作者
陈培垠
方彦军
CHEN Peiyin;FANG Yanjun(School of Power and Mechanical Engineering,Wuhan University,Wuhan 430072,China)
出处
《武汉大学学报（工学版）》
CAS
CSCD
北大核心
2020年第2期139-144,共6页
Engineering Journal of Wuhan University
基金
国家自然科学基金面上项目(编号:61170024)
国家自然科学基金资助项目(编号:51707135)。
关键词
短期负荷预测
卡尔曼滤波
节假日
相关因素
节假日逐点增长率
short-term load forecasting
Kalman filtering
holiday
related factors
holiday point-by-point growth rate
