利用统计分析软件建立相关图洪水预报方案的多元回归方程  

Establish multiple regression equation of correlation map flood forecasting scheme by statistical analysis software

作　　者：姜文 JIANG Wen(Hydrology Center of Liuzhou City,Liuzhou 545000,China)

机构地区：[1]柳州水文中心,广西柳州545000

出　　处：《广西水利水电》2021年第6期88-91,共4页Guangxi Water Resources & Hydropower Engineering

摘　　要：相关图是一种常用的洪水预报方案,使用起来简便且直观性强,但是人工制作比较繁琐。为了使这一传统的技术向现代化及自动化转变,本文介绍利用统计分析软件建立相关图洪水预报方案的多元回归方程,进行相关图转换为多元回归方程的分析研究。采用1997—2006年柳州水文站洪峰大于9000 m^(3)/s的洪水资料评定多元回归方程预报方案的精度,预报合格率为86.7%;采用2007年的资料进行检验,检验合格率为66.7%。方案等级评定为甲级,可用于正式的洪水作业预报。As a commonly used flood forecasting scheme,correlation map is easy and intuitive to be used,but re⁃quires cumbersome manual drawing.In order to realize modernization and automation transformation of this tradi⁃tional technique,an introduction was made on the establishment of multiple regression equation of correlation map flood forecasting scheme by statistical analysis software.The correlation map is transformed into multiple regression equation.The data of floods with the peak flow exceeding 9000m^(3)/s measured by Liuzhou Hydrological Station from 1997 to 2006 was used to evaluate the accuracy of multiple regression equation forecasting scheme,with the fore⁃cast qualified rate being 86.7%.The flood data of 2007 was also used to inspect with a qualified rate of 66.7%.This scheme is evaluated as Grade I and can be adopted for formal flood forecasting.

关 键 词：相关图 洪水预报 多元回归方程 SPSS 柳州水文站 

分 类 号：P338[天文地球—水文科学]