基于季节性AR(P)模型的水质预测

自回归模型的建立是基于序列平稳性的假设,只能描述平稳序列的统计特性,而水质的月监测数据序列往往具有季节性变化的现象.文章介绍了平稳过程的相关理论及其检验方法并应用到黄河潼关、三门峡断面的水质序列的检验中,检验结果为非平穗序列,且序列具有明显季节性(月份)变化的特性。为此尝试建立季节性AR (P)模型来捕捉黄河水质的季节性变化规律,实践表明该模型预测总体效果是较为满意的。

Prediction of Water Quality Based on Seasonal AR(P)Model

The regression model is based on a series of assumptions smooth, steady sequence can only describe the statistical characteristics, and water quality monitoring data from the test sequence are often seasonal change phenomenon. This paper introduces the smooth process of testing methods and the theory and application of the Yellow River at Tongguan and Sanmenxia section of the test sequence quality, the test results of non-stationary sequence, and sequences have obvious seasonal(month)in character.This attempt to establish seasonal AR(1) model to capture the seasonal changes in water quality in the Yellow River law practice shows that the model predicted the overall effect is more satisfied.