最近邻点法预测模型及其在河川径流预测中的应用

Nearest Adjacent Forecasting Model and Its Application to Stream Flow Prediction

针对径流非线性特征,建立了最近邻点法预测模型,对黄河、长江干流径流过程进行了预测,得到了满足精度要求的预测结果。该模型能够有效跟踪径流系统中相空间里的吸引子和充分扑捉历史数据中所隐含的有用信息。通过增加邻近点数来观察预测均方误差的变化,对预测效果进行了检验,结果证实该模型可真实地反映河川径流变化的总体趋势,并为判断时间序列数据的非线性提供了一种新方法。

In view of the stream flow nonlinearity, the nearest adjacent forecast model has been established to analyze the flow process in the Yellow River and the Yangtse River main streams and the satisfying precision is achieved. The model can efficiently track the attracted dot in phase-spaces efficiently and useful information in the original dates well, the change of average square errors change is found through adding adjacent dots to further test the forecast effects. The results indicate that the method can reflect the general trend of the stream flow truly, which provides a new method to estimate the nonlinearity of time series.