3D response surface modeling based in-situ assessment of physico-mechanical characteristics of alluvial soils using dynamic cone penetrometer. (September 2022)

• Record Type: Journal Article
• Authors: Khalid, Usama; Rehman, Zia ur; Mujtaba, Hassan; Farooq, Khalid
• Is Part Of: Transportation geotechnics. Volume 36(2022)
• Publication Date: 2022-09
• Subjects: Dynamic cone penetrometer -- In-situ testing -- Alluvial soils -- Physico-mechanical characteristics -- 3D response surface modeling; Engineering geology -- Periodicals; Soil mechanics -- Periodicals; Rock mechanics -- Periodicals; Transportation -- Periodicals; 624.15105
• DOI: 10.1016/j.trgeo.2022.100781
• Languages: English
• ISSNs: 2214-3912

Abstract

Highlights: Assess the ability of in-situ DCP on alluvial soils. Evaluate the behaviors of γd, qu, E50, εf, en, and CBR against DCPI. Develop predictive models using wn integrated DCPI by 3D response surface analyses. The wn has major significant on mechanical behavior of alluvial soils determined by DCP. Abstract: Dynamic cone penetrometer ( DCP ) is the quickest and cheapest in-situ probing apparatus to estimate the soil stiffness; however, insufficient correlations are available to directly assess various physico-mechanical characteristics of natural alluvial soils using DCP . The key purpose of this study is to statistically equate the in-situ DCP test outcomes to different physico-mechanical characteristics of alluvial soils. For this purpose, in-situ DCP tests were performed at 26 different locations in the Indus plane, and corresponding undisturbed specimens were collected for laboratory testing. Various physico-mechanical characteristics of soil i.e., natural water content ( wn ), natural dry unit weight ( γd ), unconfined compressive strength ( qu ), deformation modulus ( E50 ), strain at failure ( εf ), natural void ratio ( en ) and California bearing ratio ( CBR ) were determined in the laboratory. Initially, the regression-based models were established for γd, qu, E50, εf, en, and CBR using dynamic cone penetrometer index ( DCPI ) as a solitary predictor. Afterward, predictive models of the aforementioned parameters were developed based on wn integrated DCPI by quantifying the best-fit trend line and three-dimensional (3D) response surface among pertinent dependent and independent variables. In addition, it is found that the models developed using 3D response surface modeling hold 8.2–42.4% lesser error than the regression-based models developed in the current study. Moreover, results show that the DCP based assessment of soil is influenced by wn and independent of consistency limits and soil type of the alluvial soils.