摘要
时间序列模型在国际和国内的短期电力负荷预测中得到了广泛应用。然而,这种方法的一个主要缺点是无法将影响负荷预测的主要因素之一即气象因素考虑进去。在此背景下,首先基于负荷和气温数据建立了负荷预测的回归模型,然后构造了回归模型残差累积式自回归—滑动平均模型并对回归模型进行修正。最后,用广东电力系统的实际负荷数据说明了所发展的短期负荷预测模型的实际预测效果。计算结果表明所提出的方法可以弥补现有时间序列模型的缺点,有效地提高负荷预测精度。
The time series forecasting model represents a classical prediction method,and has been widely used for short-term load forecasting in actual power systems around the world.However,this model＇s one main shortage is that it cannot take weather factors into account,which usually play an important role in short-term load forecasting.In this back grourd,the regression model between load and temperature was first developed for load forecasting.Then the Auto-Regressive Integrated Moving Average（ARIMA） model of the regression model＇s residual was built for modifying the regression model.Actual load data from Guangdong power system were employed to demonstrate the developed short-term load forecasting model,and it is shown by simulation results that the developed approach could avoid the shortcoming of the time series model and improve the load forecasting accuracy efficiently.
出处
《华北电力大学学报（自然科学版）》
CAS
北大核心
2011年第3期61-66,共6页
Journal of North China Electric Power University：Natural Science Edition
基金
高等学校博士学科点专项科研基金资助项目(200805610020)
广东省电力公司科技项目
