摘要
经实例预测分析发现,利用累积式自回归动平均法(autoregressive integrated moving average,ARIMA)进行电力短期负荷预测时所得误差序列有较明显的周期规律性,针对此现象及其原因,为提高预测精度,提出采用最小二乘支持向量机(least squares support vector machine,LS-SVM)对ARIMA预测误差进行修正的ARIMA-LS-SVM组合模型;利用该改进模型对哈尔滨电网负荷进行实例预测,结果表明:该方法能够提高短期负荷的预测精度,并且具有较强的推广性和应用能力。
It is found through forecasting case study that the error sequence has obvious cycle regularity while using autoregressive integrated moving average（ARIMA） in short-term power load forecasting.For this phenomenon and its reason,an ARIMA-LS-SVM model,which uses least squares support vector machine（LS-SVM） to correct the error of ARIMA forecasting,is proposed to improve forecasting accuracy.Using this improved model to the load forecasting of Harbin power grid,the results indicate that the model can improve the accuracy of short-term load forecasting and is quite applicable.
出处
《广东电力》
2010年第11期14-17,共4页
Guangdong Electric Power
