摘要
郑州市是我国严重缺水的城市之一。为掌握该市降水资源变化状况,提供区域水资源合理开发利用的依据,运用差分自回归滑动平均模型(ARIMA),在对郑州市1951-2008年降水量数据进行预处理、模式识别、参数估计与模型检验的基础上,建立了郑州市年降水量的ARIMA(2,1,2)预测模型。结果表明:该模型预测相对误差范围为-6.36%～0.49%,预测精度较高,可以用于该区未来降水量的短期预测。
Zhengzhou City is one of serious water shortage cities in China.In order to realize the change state of precipitation,and to provide the basis for the rational development and utilization of water resources,we applied ARIMA model to establish the ARIMA（2,1,2） model for the annual precipitation forecasting of Zhengzhou City.Some works including the data preprocessing,pattern recognition,parameter estimation and model checking have been done before the establishment of this model.The results showed that the forecast of ARIMA（2,1,2） model is exact,the relative error ranges from-6.36% to 0.49%.The model can be used for short-term forecast of annual precipitation.
出处
《水土保持研究》
CSCD
北大核心
2011年第6期249-251,共3页
Research of Soil and Water Conservation
基金
河南省教育厅自然科学基金(2009A170003)
河南理工大学博士基金(B2011-033)
河南省政府决策研究招标课题(2011B231)
关键词
郑州市
年降水量
ARIMA
预测
Zhengzhou City
annual precipitation
ARIMA
forecasting
