摘要
电能表需求量的预测直接影响计量中心采购、检定和配送等计划的制定以及库存的有效利用。介绍基于加法的Holt-Winters预测模型原理,利用该模型对2013—2015年历史数据进行拟合,通过最小残差平方和计算出最优平滑系,对2016年数据进行预测并实际值进行对比,验证了该模型的有效性。
The forecast results of the energy meter demand will have a direct impact on the formulation of the measurement center procurement,verification and distribution plans and the effective use of inventory.Therefore,this paper first introduces the principle of time series prediction model of cumulative seasonal HoltWinters;Secondly,the model is used to fit the historical data from 2013 to 2015,and the optimal smoothing parameters are selected according to the sum of squared residuals;Finally,the demand for electric energy meter in 2016 is predicted and compared with the actual demand.The comparison results verify the validity of the model.
作者
李兵
李翀
吴一敌
张颖琦
文雅
LI Bing;LI Chong;WU Yidi;ZHANG Yingqi;WEN Ya(STATE GRID Hebei Electric Power Research Institute,Shijiazhuang 050021,China;STATE GRID Hebei Electric Power Co.,Ltd.,Shijiazhuang 050021,China;Holley Technology Ltd,Hangzhou 310000,China)
出处
《河北电力技术》
2019年第5期7-9,共3页
Hebei Electric Power
