Obtaining soil-water characteristic curves by numerical modeling of drainage in particulate media


Sattari A. S., TOKER N. K.

COMPUTERS AND GEOTECHNICS, cilt.74, ss.196-210, 2016 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 74
  • Basım Tarihi: 2016
  • Doi Numarası: 10.1016/j.compgeo.2016.01.006
  • Dergi Adı: COMPUTERS AND GEOTECHNICS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.196-210
  • Anahtar Kelimeler: Unsaturated soils, Soil-water characteristic curve, Packing algorithm, Pore network, Numerical modeling, LIQUID BRIDGE, BULK-DENSITY, NETWORKS, SPHERE, RETENTION, TEXTURE, PREDICT, FORCES, VOLUME
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

Soil-water characteristic curve (SWCC) is widely used for obtaining mechanical and hydraulic properties of unsaturated soils, such as shear strength, deformation, permeability, and flow. An innovative approach, where a meso-scale medium is generated based on particle size distribution and void ratio of non-plastic soils, for estimating the drying SWCC is developed. With application of the finite difference and Newton-Raphson (Jacobian) approximations, the air-entry pressures of pore bodies in inter-particle medium (micro-scale) are determined and implemented in drainage simulation of the medium. The volumes of drained pore bodies and subsequently developed liquid bridges after each suction iteration are calculated and plotted. Eventually, homogeneity of the developed packing algorithm, parametric study, comparison of the simulated drying SWCC to experimental and estimation results as well as computational performance of the developed MATALB code is presented and discussed. It is shown that, the proposed method results in much superior approximation of SWCC in comparison to the two major estimation methods (Arya and Paris, 1981; Fredlund and Wilson, 1997) and due to its accuracy and time efficiency, the algorithm can even be a viable option for replacing the tedious experimental methods. (C) 2016 Elsevier Ltd. All rights reserved.