基于ARIMA模型的区域用电量预测方法

Regional electricity consumption forecasting method based on ARIMA model

传统区域用电量预测方法存在预测能力差的问题,为此,提出基于ARIMA模型的区域用电量预测方法。获取区域的历年用电量数据进行预处理,获取统一的用电量数据;再利用ARIMA模型对用电量时间序列进行平稳性检测,利用最小二乘估计方法估值用电量参数;最后结合线性神经网络构建区域用电量预测模型,将统一的用电量数据放入模型中进行计算,以此完成区域用电量的预测。实验结果表明,所提方法可以有效检测出电量负荷及电量同比增速,预测能力强、预测精度高。

The traditional regional electricity consumption prediction method has the problem of poor prediction ability, therefore, a regional electricity consumption prediction method based on ARIMA model is proposed. The regional electricity consumption data over the years is obtained, and preprocessed to obtain the unified power consumption data. Then the stability of electricity consumption time series is detected by ARIMA model, and the electricity consumption parameters are estimated by least square estimation method. Finally, combined with the linear neural network, the regional electricity consumption prediction model is constructed, and the unified electricity consumption data is put into the model for calculation, so as to complete the prediction of regional electricity consumption. The experiment results show that this method can effectively detect the power load and the year-on-year growth rate of power, with excellent prediction ability and high prediction accuracy.