摘要
用电量预测是开展电力需求侧管理的一项基础性工作。在实际预测中,许多经济变量是可以定量度量的,而某些影响经济变量的因素却无法量化,定性因素对模型预测有很大的影响。采用虚拟变量方法,将定性因素定量化,建立含有多个虚拟变量的回归模型,在模型中反映这些因素的影响。将这一方法用于电量预测,可以有效提高模型的预测精度,取得理想的效果。
Electric consumption forecasting is a basic work in power demand side management（DSM）. In actual forecasting, many economic variables can be measured quantificationally. However, some factors which influence the economic variables can not be quantified while qualitative factors influence the model forecasting greatly. Dummy variables are used in the model to quantify those factors, and the regression model including multi-dummy variables is established. This method is applied in power consumption forecasting which can improve forecasting precision effectively and achieve better results.
出处
《电力需求侧管理》
北大核心
2007年第3期16-18,共3页
Power Demand Side Management
基金
河北省科学技术研究与发展计划项目(054472261D)
关键词
需求侧管理
电量预测
回归模型
虚拟变量
DSM
power consumption forecasting
regression model
dummy variables
