城市用水量组合预测模型及其应用

针对城市用水量时间序列包含逐步增长趋势、季节性趋势及不确定性的非线性波动特点,单一预测模型往往很难充分反映原始数据中全部的有效信息,结合季节性时间序列模型(SARIMA)和BP神经网络二者优点,构建了一种新型的组合预测模型,对上海市用水量进行不同时间尺度的预测。结果表明,在不同时间尺度上组合预测模型均比单一预测模型精度高、预测质量稳定。

Combination Prediction Model of Urban Water Supply and Its Application

The time series of urban water consumption contains the characteristics of gradual growth, seasonal trend and uncertainty. The single prediction model is often difficult to fully reflect all the effective information in the original data. Combining two advantages of seasonal time series model (SARIMA) and BP neural network, a new combined forecasting model is constructed to predict the water consumption of Shanghai at different time scales. The results show that the combination forecasting model has higher accuracy and better prediction quality than the single prediction model at different time scales.