| 一种基于回归分析与时序分析的降水预报模型 |
| |
| 引用本文： | 唐亚松,张鑫,蔡焕杰,王健.一种基于回归分析与时序分析的降水预报模型[J].水土保持通报,2009,29(1):88-91. |
| |
| 作者姓名： | 唐亚松 张鑫 蔡焕杰 王健 |
| |
| 作者单位： | 唐亚松,TANG Ya-Song(西北农林科技大学水利与建筑工程学院,陕西,杨凌,712100;武汉大学水利水电学院,湖北武汉430072);张鑫,蔡焕杰,王健,ZHANG Xin,CAI Huan-jie,WANG Jian(西北农林科技大学水利与建筑工程学院,陕西,杨凌,712100)
|
| |
| 基金项目： | 国家高技术研究发展计划(863计划)，国家重大科技支撑项目，西北农林科技大学博士科研启动基金，西北农林科技大学科研专项基金 |
| |
| 摘 要： | 中长期降水预报存在的问题是资料不足和预报精度较低.运用回归分析与时序分析相结合的方法,将实测降水序列分解成趋势、周期、随机3种成分,并分别构建各分量的子模型;将3个子模型线性叠加,并对拟合的纯随机序列进行修正,得到降水预报的第三类模型;进而给出了模型精度评价方法.实例分析表明,运用修正后的模型进行降水预报,可以缩小峰值处的误差,预报精度比其余两类模型有所提高.该方法可仅根据历史资料进行降水预报,对资料要求较低,精度可靠,是一种实用的方法.
|
| 关 键 词： | 降水预报 回归分析 时序分析 模型修正 精度评价 |
| 收稿时间： | 2008/5/22 0:00:00 |
| 修稿时间： | 2008/9/29 0:00:00 |
A Precipitation Forecast Model Based on Regression Analysis and Time Series Analysis |
| |
| TANG Ya-song,ZHANG Xin,CAI Huan-jie and WANG Jian.A Precipitation Forecast Model Based on Regression Analysis and Time Series Analysis[J].Bulletin of Soil and Water Conservation,2009,29(1):88-91. |
| |
| Authors: | TANG Ya-song ZHANG Xin CAI Huan-jie and WANG Jian |
| |
| Institution: | College of Water Resources and Architectural Engineering, Northwest A & F University, Yangling, Shaanxi 712100, China;College of Water Resources and Hydropower, Wuhan University, Wuhan, Hubei 430072, China;College of Water Resources and Architectural Engineering, Northwest A & F University, Yangling, Shaanxi 712100, China;College of Water Resources and Architectural Engineering, Northwest A & F University, Yangling, Shaanxi 712100, China;College of Water Resources and Architectural Engineering, Northwest A & F University, Yangling, Shaanxi 712100, China |
| |
| Abstract: | In this paper,a composite model based on regression analysis and time series analysis is constructed to improve forecast accuracy and deal with problems of data deficiencies in mid-to-long-term precipitation forecast.Generally speaking,hydrologic time series may be split into the trend,cyclic,and random components.Under this guideline,firstly,precipitation sequence is split into the three components mentioned above by using regression analysis and time series analysis methods;three submodels are respectively constructed for the three components;the independent random sequence is stabilized to modify traditional model structure;and the three submodels are added by linear superposition to establish a new precipitation forecast model(the 3rd type).Later,the paper gives accuracy assessment indexes.Results show that errors of the modified model are lower in rain spell,i.e.,its accuracy is higher than other two models.So,using historical data,the model can work well in precipitation forecast and get a proper accuracy.It is proved to be a practical model for forecast. |
| |
| Keywords: | precipitation forecast regression analysis time series analysis model modifying accuracy assessment |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《水土保持通报》浏览原始摘要信息 |
| 点击此处可从《水土保持通报》下载免费的PDF全文 |
|
