MULTIPLE REGRESSION ANALYSIS BETWEEN THE MECHANICAL AND PHYSICAL PROPERTIES OF COHESIVE SOILS

概要

論文の詳細を見る
In order to accurately determine the relationships between a mechanical property and some physical properties in both natural and artificially mixed cohesive soils, multiple regression analysis was performed. In this paper, two kinds of regression models representing the shear strength of soils are examined. The first model is based on the Atterberg limit and the second is based on the Cam-Clay model. The concept of this study is that the mechanical properties of the soils are characterized by soil types and soil states. The following conclusions were obtained : (1) Both regression models established for the shear strength of soils are not only found to be valid for the unconfined compressive strength, but also for the modulus of deformation and consolidation yield stress. (2) The validity of the concept employed here can be quantitatively proved by multiple regression analysis. (3) In a fully saturated soil, the correlation between the two regression models can be found both by statistical analysis and by theoretical examination. (4) The regression equations are obtained in good predictive accuracy for unconfined compressive strength, the modulus of deformation, consolidation yield stress and compression index.
社団法人地盤工学会の論文
1990-09-15