高填路堤沉降的优化组合预测  

Optimized Combination Forecasting Settlement of High Backfill Subgrade

作　　者：曹喜仁[1] 林龙[1] 

机构地区：[1]湖南大学土木工程学院,湖南长沙410082

出　　处：《中南公路工程》2007年第2期9-12,共4页Central South Highway Engineering

摘　　要：通过对高填路堤沉降规律的分析,采用组合预测方法进行沉降预测,以便综合利用不同预测模型所提供的信息并提高预测精度。普遍组合预测模型多为以组合预测的误差平方和或离差绝对值之和最小为精度指标的单目标优化模型,不一定能如实反映预测方法的有效性。采用基于预测有效度为精度指标的多目标组合预测模型,求解优化预测值,并通过具体工程实例说明其优越性。Through analysis of settlement rules of high backfill subgrade,we can make a combination forecasting model with several curves to forecast settlement, in order to utilize the information synthetically provided by different models and improve the precision of forecasting. Ordinary combination forecasting models may not show the availability of the forecasting methods with the only target model of the accuracy norm of the sum of errors＇ square or the deviation absolute sum. We can make another combination forecasting model with the several targets model of the accuracy norm of the forecasting effective measure and solve the problem simply, then illustrate its availability through the analysis of the engineering practice.

关 键 词：高填路堤 沉降预测 组合预测模型 预测有效度 

分 类 号：U416.12[交通运输工程—道路与铁道工程]