Deterministic methods have been used in geotechnical engineering for a long period, such as slope stability calculations. However, only applying deterministic methods is subjective and imperfect. There is a demand to develop a systematic methodology to link the assessed slope stability and field measurement data, which is also known as inverse analysis and forward calculation. Based on the Nya Slussen project, this thesis includes the development of a methodology, deterministic calculation for 4 cross sections using finite element program Plaxis 2D and probabilistic calculation for one section. Deterministic analyses showed satisfying results for all the studied cross sections since their factors of safety exceeded the minimum requirement. In probabilistic design, three parameters were found to have the most uncertainties through sensitivity analysis (undrained shear strength of clay, Young’s modulus of clay and friction angle of fill). Inverse analysis was done by testing different values of them in Plaxis and to try to match the displacement components provided by field measurement. After finding the best optimization for all the parameters, forward calculation gave a final factor of safety. It is suggested that both of the methods should be utilized together for better assessment.

The estimation of bedrock level for soil and rock engineering is a challenge associated to many uncertainties. Nowadays, this estimation is performed by geotechnical or geophysics investigations. These methods are expensive techniques, that normally are not fully used because of limited budget. Hence, the bedrock levels in between investigations are roughly estimated and the uncertainty is almost unknown. Machine learning (ML) is an artificial intelligence technique that uses algorithms and statistical models to predict determined tasks. These mathematical models are built dividing the data between training, testing and validation samples so the algorithm improve automatically based on passed experiences. This thesis explores the possibility of applying ML to estimate the bedrock levels and tries to find a suitable algorithm for the prediction and estimation of the uncertainties. Many diferent algorithms were tested during the process and the accuracy level was analysed comparing with the input data and also with interpolation methods, like Kriging. The results show that Kriging method is capable of predicting the bedrock surface with considerably good accuracy. However, when is necessary to estimate the prediction interval (PI), Kriging presents a high standard deviation. The machine learning presents a bedrock surface almost as smooth as Kriging with better results for PI. The Bagging regressor with decision tree was the algorithm more capable of predicting an accurate bedrock surface and narrow PI. / BIG and BeFo project "Rock and ground water including artificial intelligence

Triaxial shear test is the most accurate test for deciding the undrained shear strength of clay. In every test the ratio between the horizontal and vertical stresses, the coefficient of earth pressure at rest (K0′), has to be decided. It’s widely believed that the choice of this parameter will influence the results, but it’s not known to what extent. In this thesis 20 consolidated undrained active triaxial shear tests on clay has been con- ducted with a K0′ at 0.6 and 0.8. The clay was collected with a 54 mm piston sampler at the Norwegian Geo-Test Site in Trondheim, Norway, on depth of 3.0 to 7.8 meters. Besides the triaxial testing, index tests and oedometer tests was conducted on every cylinder to do a gen- eral classification of the clay. The clay has a sensitivity of 9-20, a water content of 35-51 %, a plasticity index of 27-65 % and an over consolidation ratio of 2.6-6.8. The consolidation phase of the triaxial test was conducted in five loading steps with a rest time in-between to monitor the amount of pore water expelled at each stress level. The loading steps was 50 %, 75 % and 100 % of maximum cell pressure and thereafter at 50 % and 100 % of the maximum deviator stress. The shear phase was done at a speed of 1.5 % per hour to a total of 10 % axial strain. It is not possible to reach an unambiguous conclusion from the results, but the maximum shear strength of tests with a K0′ at 0.8 is 17 % higher, while the total amount of pore water extortion is equal between the two values. The amount of creep in the latest steps is on the other hand smaller for a K0′ at 0.8. This indicates that the soil is handling the stress level better than with a K0′ at 0.6.

An  important  task   during  the  construction  process  is  to  validate  the dimensions  and  properties  of  a  given  structure.  The  dimensions  like  for instance the thickness of a construction element should  be measured after finishing  building  it.  The  aim  is  to  compare  the  measured  value  with  the design  value   to  avoid   that   elements  do   not  correspond  to  the  input requirements.  Moreover,  the  measurements  are  helpful  to  analyse  the statistical    distribution    of    the    investigated    geometrical    property    by computing e.g. a histogram, which visualises the dispersion and  enable the calculation of the probability of failure for a specific structure or element.In  this  work,  a  shotcrete  layer  has  been  analysed  in  order  to  provide information  about  the  homogeneity  of  the  shotcrete  thickness  in  a  pre-determined  tunnel  section.  The  calculation  method  is  based  on  two  laser scans,  before  and  after  applying  the  shotcrete.  Due  to  the  construction process, the shotcrete layer will not be totally equal, which might be a safety problem. Especially, when  the shotcrete layer is thinner than  required and hence, the actual variation of the shotcrete must be considered and verified.To determine the statistical distribution, correlograms and histograms have been  computed  for  a  wall  area  in  a  tunnel  in  Southern  Sweden.  The correlogram shows the distance where the values have a correlation to each other  and  usually  this  distance  is  called  scale  of  fluctuation.  For  the  wall section, this scale of fluctuation has been calculated for the length (0.8m) as well  as  the  height  (0.8m).  Compared  to  the  original  sample  distance,  e.g. distance of the rock bolts, the variance for the calculation of the probability of failure might be reduced.

Seismic pressures on displacing or rigid retaining or basement walls have been derived based on the original work of Mononobe and Okabe, who used a shake table to calculate dynamic pressures of displacing retaining walls existing in cohesionless soils. Since this original work was done over eighty years ago, the results of Mononobe and Okabe, colloquially known as M-O theory, have been applied to different conditions, including non-displacing basement walls, as well as changes in soil properties. Since the original work of M-O, there have been numerous studies completed to verify the accuracy of the original calculation, most notably the work of Seed and Whitman (1970), Wood (1973), Sitar (Various), and Ostadan (2005). This has resulted in varying opinions for the accuracy of M-O theory, whether it is grossly unconservative or conservative, as well as its effectiveness for situations where the wall does not displace enough to engage active soil conditions. This study examines (3) different wall cases, a cantilever retaining wall, gravity retaining wall, and rigid basement wall, through an implcit finite element analysis, under simple sinusoidal boundary accelerations. The soil is modeled using the Drucker-Prager model for elastic-plastic properties. The dynamic pressure increment is observed for different driving frequencies, with the anticipation that an in-phase and out of phase response between the soil and structure will be achieved, resulting in both lower and higher than M-O pressure values.

The caving/sloughing of sandy layers into drilled shafts is a common and costly phenomenon in the drilling industry. A prototype soil-testing device known as the Pneumatic In-situ Soil Caving Index Sampler (PISCIS) has been developed to test sandy layers above the water table for their propensity to cave/slough into a drilled shaft during the drilling process. The PISCIS fits down a Cone Penetration Test (CPT) hole and uses air pressure to agitate a sample off of the hole wall that is then collected and weighed. Large-scale lab testing was conducted using sand under a variety of simulated overburden pressures and fines contents. The tests were conducted with a dual purpose in mind. First, the tests confirmed the functionality of the PISCIS prototype and its ability to collect samples in a consistent and repeatable manner. Second, the tests resulted in a calibration curve that shows a very strong (nearly exponential) relationship between collected sample weight and the fines content of the test sand; higher fines contents resulted in lower collection weights. The PISCIS was designed to supplement information found in a geotechnical report with information that would specifically inform drilling contractors about potential caving/sloughing hazards found in the stratigraphy.

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px 'Times New Roman'} A methodology is presented for evaluating the potential surface fault displacement on reverse faults in a probabilistic manner. This methodology follows the procedures put forth for Probabilistic Fault Displacement Hazard Analysis (PFDHA). Empirical probability distributions that are central to performing a PFDHA are derived from field investigations of reverse faulting events. Statistical analyses are used to test previously assumed properties of scale invariance with respect to magnitude for normalized displacement. It is found that normalized displacement is statistically invariant with respect to magnitude and focal mechanism, allowing for the combination of a large number of events into a single dataset for regression purposes. An empirical relationship is developed using this single dataset to be used as a fault displacement prediction equation. A PFDHA is conducted on the Los Osos fault zone in central California and a hazard curve for fault displacement is produced. A full sensitivity analysis is done using this fault as a reference, to test for the sources of variability in the PFDHA methodology. The influence of the major primary variables is quantified to provide a future direction for PFDHA.

Lifelines such as highways, pipelines, telecommunication lines, and powerlines provide communities with vital services, and their functionality is dependent upon the foundation soil that supports them. However, when designing the infrastructure, it can be difficult to know where to test the soil in order to give spatially representative sampling, particularly for long, lifeline structures. Finding this distance requires knowledge of the spatial correlation and/or the spatial variability of the soil parameter (stiffness, cohesion, etc.). But this correlation distance is not typically found in practice because it requires large amounts of data and the costs of retrieving that data can be high. Lack of representative sampling can lead to an overly conservative design and too much sampling can create an overly expensive sampling program. In this study, multiple tests using the geophysical method of spatial autocorrelation (SPAC) were conducted to find the soil stiffness along a 310 meter long profile. SPAC records passive surface waves which sample the underlying soil, and these surface waves can be used to create a shear wave velocity profile of the site. The spatial continuity of the stiffness (the soil velocity values) was then found using geostatistics. The geostastical tool primarily used in this study was the (semi-)variogram, but the covariance function and the correlogram are also shown. By using these tools, the spatial correlation/variability can give an estimate of the how far apart to test the foundation soil so that the data is spatially representative. In other words, finding the distance that the soil parameter is minimally correlated with itself. This study found the distance (the range of the semi-variogram) to be 70 meters for 5 meters depth, 100 meters for 10 to 15 meters depth, and 90 meters for 30 meters depth.

This dissertation describes enhancements made to the finite element program, SAGE, and research on the performance of deep excavations. SAGE was developed at Virginia Tech for analysis of soil-structure interaction problems (Morrison, 1995). The purpose of the work described in this text with SAGE was to increase the capabilities of the program for soil-structure analysis. The purpose of the research on deep excavations was to develop a deeper understanding of the behavior of excavation support systems. The significant changes made to SAGE during this study include implementation of Biot Consolidation, implementation of axisymmetric analysis, and creation of a steady state seepage module. These changes as well as several others are described. A new manual for the program is also included. A review of published studies of deep excavation performance and recent case histories is presented. Factors affecting the performance of excavation support systems are examined, and performance data from recent published case histories is compared to data from Goldberg et al.'s 1976 report to the Federal Highway Administration. The design, construction, and performance of the deep excavation for the Dam Number 2 Hydroelectric Project is described. Finite element analyses of the excavation that were performed with SAGE are presented and discussed. / Ph. D.

The phenomenon of vibration-induced shear resistance reduction (ViSRR) in granular soil is characterized by the loss of shear resistance without significant excess pore pressure generation. It has diverse potential applications in various industries such as mining, pharmaceuticals, and civil engineering, including the installation of vibratory-driven piles. Despite limited research on this topic, both experimentally and theoretically, the mechanism associated with ViSRR remains challenging to explain. There is currently no established constitutive model to properly describe it. This dissertation investigates the fundamental features of ViSRR and develops a model to describe the process that leads ViSRR. To achieve these objectives, three main areas of investigation were undertaken. First, a series of laboratory tests were conducted using a modified triaxial apparatus that allowed for vibrations superimposed on the monotonic shearing of granular soil samples. Second, by correlating macroscopic plastic strains with the transition, creation, and destruction of mesoscopic shear-transformation-zones (STZs), which can be considered as weak particle loops in granular assemblies, the conventional thermodynamic-based STZ model was extended to soil mechanics. Third, the concept of "vibration-induced shear resistance relaxation" was proposed, which refers to the loss of shear resistance in granular material subjected to restricted deformations in response to plastic strains induced by vibrations. In other words, ViSRR occurs when the total deformation rate of the granular material is constrained and does not keep up with the rate of plastic deformation induced by vibrations. By conducting laboratory tests, developing the extended STZ model, and proposing the concept of "vibration-induced shear resistance relaxation", this dissertation contributes to a better understanding of ViSRR in granular soil and provides insights into the mechanisms governing this phenomenon. The results of this research can be used to improve the design and construction of geotechnical structures. / Thesis / Doctor of Philosophy (PhD)