Doctor of Philosophy Inverse Analysis to Estimate Hydraulic Properties for Unsaturated Sand The Graduate School of the University of Ulsan Department of Civil and Environmental Engineering To Viet Nam 1 Doctor of Philosophy Inverse Analysis to Estimate Hydraulic Properties for Unsaturated Sand The Graduate School of the University of Ulsan Department of Civil and Environmental Engineering To Viet Nam 2 Inverse Analysis to Estimate Hydraulic Properties for Unsaturated Sand Supervisor: Professor Min Tuk-ki A Dissertation Submited to The Graduate School of the University of Ulsan in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy By To Viet Nam Department of Civil and Environmental Engineering Ulsan, Korea November 2013 3 4 ACKNOWLEDGMENTS This is my great pleasure to have this opportunity to thank people who have helped me during my dissertation and my study at Laboratory of Geotechnics, Department of Civil and Environmental Engineering, University of Ulsan, Korea. First of all, I would like to sincerely thank my supervisor Professor Min Tuk-Ki for his brilliant supervising, guidance, patience, consistent supports and encouragements throughout my Ph. I would also like to thank and gratitude to Professor Shin Ho-sung for accepting me into his group, giving me freedom in my work, supporting my attendance at various conferences, engaging me in new ideas, and demanding a high quality of work in all my attemps. Additionally, I would like to acknowledge the committee members Professor Min Tuk-Ki, Professor Lee Jong Seok, Professor Yoo Ji Hyeong, Professor Kang Suk-Bong and Professor Shin Ho-Sung for their comments, advises and suggestions in my study.
I also express my appreciation to all members of the Geotechnical Lab. Every result described in this thesis was accomplished with the help and support of fellow labmates and collaborators. Thank are also due to professors and staff members of Department of civil and Environmental Engineering for their kindly helps. I would like to thank my parents, my elder sister and my elder brother, who provided endless help, patience and good advice to have me through my program.
Special thanks and many loves to my wife, Nguyen Thi Thuy Linh and my daughter, To Nguyen Quynh Huong for endless love, great cares, encouragements and sufferings of a separation during the time I am away from home. Finally, I would like to thank for the support from all my friends from Union of Vietnamese Students in University of Ulsan. With warmest regards Ulsan 2013 To Viet Nam 5 AN INVERSE ANALYSIS TO ESTIMATE HYDRAULIC PROPERTIES FOR UNSATURATED SAND By To Viet Nam Department of civil and Environmental Engineering University of Ulsan ABSTRACT Since the term “unsaturated soils” was defined, and relation of unsaturated soils to many engineering problems was proven, the determination of unsaturated hydraulic properties, which is one of most important parameters in unsaturated soils, is a hot topic in geotechnical engineering. Realizing importance of unsaturated soil hydraulic properties, for a long time, many methods for determining the unsaturated hydraulic properties of soils have been proposed by many researchers.
However, these methods usually encounter the conditions of tediousness, time and money consumption, complexity, inaccuracy, narrow applicability, etc. Therefore, up to now, understanding and discovering a simple, appropriate and accurate method for determination of unsaturated hydraulic properties of soils is still received many interests from the scientists. In recent decades, the inverse analysis method becomes a useful technique in determining the unsaturated hydraulic properties of soils. This method, in many cases, has been shown to be an exciting new tool that allows for soil hydraulic characterization using a wide range of transient laboratory and field experiments.
The objective of this dissertation is to evaluate the applicability of inverse analysis technique in estimating the unsaturated hydraulic properties for grain soils using the input data measured from the one-dimension (1-D) outflow experiments. The applicability of inverse analysis method in this dissertation is evaluated through two following parts. 6 In the first part, the 1-D vertical draining test for homogeneous medium sand was carried out in one-step and multi-step outflow conditions. An inverse analysis method was conducted to determine the hydraulic properties of unsaturated soil.
A non-linear optimization method combining a finite element code and inversion analysis was used to minimize the objective function, defined by the difference between observed and predicted data. Unsaturated hydraulic parameters in van Genuchten model were estimated using soil suction measurements in 10 cm intervals and an outflow rate at the bottom of a sand column. The predicted hydraulic properties and the experimental results were in close agreement when the measurements were compared from the one-step and multi-step outflow experiments. The results also showed that the multi-step outflow experiment was more appropriate in determining the unsaturated hydraulic properties than the one- step outflow experiment.
The comparison between predicted and measured results concluded that the inverse analysis based on the 1-D outflow experiment was reliable and useful to determine the hydraulic properties of unsaturated soils. The second part is the further development of the first part. In this part, the inverse analysis method was applied to determine the unsaturated hydraulic properties for vertical heterogeneous sand column of two sand layers (fine and medium sand) from 1-D outflow experiment. The one- step outflow experiment was conducted to obtain the profiles of suction, saturation and flow rate with time for two heterogeneous soil columns (case A and case B, with the interchange position of two soil layers).
Similar to the first part, the inverse modeling was also carried out using two computer models, one for forward water flow modeling and the other for nonlinear regression. Finally, compare the predicted hydraulic properties with measured hydraulic properties in two cases to get the conclusion about the applicability of inverse analysis technique for vertical heterogeneous sand. 7 TABLE OF CONTENTS Chapter Title Page Acknowledgement i Abstract ii Table of contents iv List of figures vii List of tables xii Symbols xiii 1.2 Objectives and contributions 3 1.2 Original contribution in this work 4 1.3 Dissertation organization 4 2 UNSATURATED SOIL HYDRAULIC PROPERTIES 6 2.2 Unsaturated hydraulic properties of soil 9 2.1 Typical previous study for unsaturated hydraulic 13 8 properties of soils 2.2 Determination of unsaturated hydraulic properties 19 3 INVERSE ANALYSIS 29 3.2 Inverse analysis for unsaturated soil hydraulic properties 30 3.2 Water flow modeling 33 3.3 Parameter Optimization 35 4 INVERSE ANALYSIS TO ESTIMATE UNSATURATED 39 HYDRAULIC PROPERTIES FOR HOMOGENEOUS SAND 4. INVERSE ANALYSIS TO ESTIMATE UNSATURATED 63 HYDRAULIC PROPERTIES FOR LAYERED SAND 5.2 Materials and Methods 64 5.
CONCLUSIONS 91 REFERENCES 93 10 LIST OF FIGURES Fig.1 The subdivisions of unsaturated soil zone (vadose zone) on local and 2 regional basis.1 The phase diagram in soil.2 The mass and volume of phase in unsaturated soil.3 Conceptual illustration of the unsaturated soil zone: (a) pore water 8 regimes and (b) saturation, total head, and pore pressure profiles.4 The categorization of soil mechanics and stress state variables of 9 saturated soil and unsaturated soil.5 Typical hydraulic properties of differently textured soils (a) soil-water 11 characteristic curves. (b) unsaturated conductivity curves.6 Definition of terms for a typical SWCC for a soil.7 The zones in typical drying SWCC.9 van Genuchten (1980) fits a soil: (a) soil-water characteristic curve, 16 and (b) relative hydraulic conductivity function.10 Fredlund (1994) fits for Touchet silt loam (a) soil-water characteristic 18 curve, and (b) relative hydraulic conductivity function.11 Hydraulic properties of a fluviatile silt loam. (a) water retention curve 20 measurements. (b) unsaturated conductivity measurements, determined by five different methods.12 Experimental setup using Buchner funnel.13 The schematic of centrifuge testing technique.14 The experimental system of constant head method.15 The typical outflow experiment.16 The measured results of one-step outflow experiment (a) cumulative 26 outflow and (b) suction head with time.17 The measured results of multi-step outflow experiment for a soil (a) 27 cumulative outflow and (b) suction head with time.18 The measured results of continuous outflow experiment for a soil (a) 28 cumulative outflow and (b) suction head with time.1 Inverse problems in engineering and science.2 Flow chart of the inverse method illustrating the integration of 32 measurement, modeling, and optimization.3 Least square solution for initial guess of the solution.1 (a) The schematic diagram of column test, (b) tensiometer equipment 41 set.
(c) electrical resistivity tester.2 Calibration relationship between electrical resistivity and saturation 42 for the Jumunjin sand.3 The one-dimension outflow experiment: a) one-step outflow method; 43 b) multi-step outflow method.4a Cumulative outflow over time in the one-step outflow experiment.4b Discharge rate over time in the one-step outflow experiment.5 Surface displacement over time from the one-step outflow experiment. 4-6 (a) Water saturation over time obtained from the one-step outflow 47 experiment at different heights; (b) water saturation profile from the one-step outflow experiment.7a Pore-water pressure over time from the one-step outflow experiment at 48 different heights.7b Pore-water pressure head profile from the one-step outflow experiment 49 Fig.8 Values of lambda () and P0 vs.9 Comparison between predicted and measured outflow rate in the one- 50 step outflow experiment.10 Comparison between predicted and measured pore water pressures in 52 the one-step outflow experiment at locations L2, L5 and L8.11 (a) Comparison between predicted and measured degree of saturation 54 in the one-step outflow experiment at locations L2, L5 and L8; (b) comparison between predicted and measured compiled SWCC in the one-step outflow experiment at locations L2, L5 and L8.12 Pore-water pressure over time from the multi-step outflow experiment 56 at three locations.13 Compiled SWCC from locations L2, L5 and L8: a) one-step outflow 57 experiment, b) multi-step outflow experiment.14 Values of lambda () and P0 vs.15 Comparison between predicted and measured pore water pressure in 59 the multi-step outflow experiment at locations L2, L5 and L8.16 (a) Comparison between predicted and measured degree of saturation 61 in the one-step outflow experiment at locations L2, L5 and L8; (b) comparison between predicted and measured compiled SWCC in the one-step outflow experiment at locations L2, L5 and L8.1 Particle size distribution of fine sand and medium sand.2 The drying SWCC of medium sand and fine sand determined from 66 independent experiments.3 The schematic diagram of column test.4 Calibration relationship between electrical resistivity and saturation 68 13 for: solid-line for medium sand; dash-line for fine sand.5 The soil column experiment in case A.6 (a) Cumulative outflow with time in case A; (b) discharge rate with 72 time in case A.7 (a) Pore-water pressure with time at different heights in case A; (b) 74 pore-water pressure head profile in case A.8 (a) Water saturation with time obtained at different heights in case A; 75 (b) water saturation profile in case A.9 Comparison between predicted and measured outflow rate in case A.10 Comparison between predicted and measured pore water pressure in 77 one-step outflow experiment at location of L2, L4, L5 and L8 in case A.11 Comparison between predicted and measured hydraulic properties at 80 location of 2, 4, 5 and 8 in case A: (a) change in degree of saturation, (b) compiled SWCC.12 The schematic diagram of column test in case B (the fine sand overlies 81 on medium sand).13 (a) Cumulative outflow with time in case B; (b) discharge rate with 82 time in case B.14 (a) Water saturation with time obtained at different heights in case B; 83 (b) water saturation profile in case B.15 (a) Pore-water pressure with time at different heights in case B; (b) 84 pore-water pressure head profile in case B.16 Comparison between predicted and measured outflow rate in case B.17 Comparison between predicted and measured pore water pressure at 87 location of L2, L5 and L8 in case B.18 Comparison between predicted and measured hydraulic properties at 89 location of 2, 5 and 8 in case B: (a) change in degree of saturation, (b) compiled SWCC. 15 LIST OF TABLES Table 3.1 The common models to describe the SWCC and permeability of soil.1 Properties of Jumunjin sand.2 Root mean square error from predicted and measured data in time 55 domain (one-step outflow experiment).3 Root mean square error from predicted and measured data in time 58 domain (multi-step outflow experiment).