[关键词]
[摘要]
时间序列分析方法在降水量预报中有着广泛的使用价值和应用前景。根据1971年-2013年郑州市的降水量资料,采用二次多项式拟合提取降水量的趋势分量,采用谐波分析法提取降水量的周期成分,利用自回归模型求解随机成分,最后将三者叠加,构建了郑州市降水量的预报模型。模型计算结果与实测数据对比可知,应用预报模型对降水量进行预报精度较高。因此利用建立的预报模型对2014年-2016三年的降水量进行了预测,为该区水资源的管理提供依据。
[Key word]
[Abstract]
Time series analysis method has comprehensive importance and prospects in the precipitation forecast. Based on the precipitation data from 1971 to 2009 in Zhengzhou city, the trend component of precipitation was determined by the quadratic polynomial calibration, the periodic component of precipitation was extracted by the spectrum analysis, and the stochastic component was simulated using the auto-regression model. The precipitation prediction model in Zhengzhou City was established through the superposition of these components. In comparison with the simulated and measured data, the precipitation prediction model has a high accuracy. The model was applied to forecast the precipitation in Zhengzhou city from 2014 to 2016, which can provide reference for the water resources management in this area.
[中图分类号]
[基金项目]
河南平原地区地下水污染调查评价(淮河流域):国土资源大调查项目(1212010634502)