基于时间序列和灰色模型的短期电力负荷预测

Short-term power load forecasting based on time series and grey model

针对短期电力负荷预测,从不同研究角度出发,分别构建了时间序列ARIMA模型和基于相似日的灰色预测GM(1,1)模型,使用SPSS、MATLAB等软件编程,得出不同模型下所研究地区未来七天内各个时点的电力负荷预测值,并对其精度进行了检验。结果表明两种模型下的预测结果精度较高且各具优点,在实际应用中,可根据短期电力负荷的具体特征和预测要求,灵活的选择模型加以应用。

For short-term power load forecasting, this study respectively constructs ARIMA time series model and based on the similar day grey prediction GM（1,1） model from different point of view, uses SPSS and MATLAB software, obtains the forecast values o f power load at different time points in the next seven days by different models and their accuracy is also tested. The results show that the prediction results o f the two models are o f high accuracy and each has its own advantages. In practical application, according to the specific characteristics of short-term power load and forecast requirements, a flexible choice model can be applied.