摘要
进行了黄土的击实试验,提出了击实率的概念,拓展了CART(classification and regression trees)决策树算法的功能。将其用于相关性挖掘,进行了原状黄土湿陷性与击实率等指标的相关性挖掘,结果表明,击实率与湿陷系数具有相关性,击实率与最优含水率状态的击实率呈显著负相关性。根据试验及分析结果提出黄土在击实过程中的变形特性可反映黄土的湿陷性,黄土的击实效果同其湿陷性具相关性。这一研究为黄土湿陷性的评价提出了一条新的思路,同时成果具有实用性。
Loess collapsibility test and compaction test have been carried out.The concept of compaction rate is suggusted.The classification and regression trees （CART） algorithm is improved;and is used for analyzing the correlative （data） mining.The correlative mining has been performed to undisturbed loess collapsibility and index of compaction test.The results of correlative mining indicates that coefficient of collapsibility is closely correlated to the compaction rate,but significantly in negative correlation with the collapsibility coefficients.In the compaction process,the closer the soil water content is to the optimal water content,the stronger this correlation is.Finally,this paper presents a viewpoint that undisturbed loess collapsibility is correlated with the engineering property of the compaction loess,and undisturbed loess collapsibility can be evaluated with deformation property of disturbed loess samples.This is a new method of evaluating loess collapsibility.
出处
《岩土力学》
EI
CAS
CSCD
北大核心
2011年第2期393-397,共5页
Rock and Soil Mechanics
基金
国家自然科学基金资助项目(No.10572090)
中国中煤能源集团公司重点科技项目(No.08-26)
中央高校基本科研业务费专项资金长安大学基础研究支持计划专项基金(No.CHD2009JC084)
关键词
湿陷性
击实率
决策树
相关性挖掘
负相关
collapsibility
compaction rate
decision trees
correlation mining
negative correlation
