Application of Bayesian approach in geotechnical engineering

En esta Tesis Doctoral se emplean y desarrollan Métodos Bayesianos para su aplicación en análisis geotécnicos habituales, con un énfasis particular en (i) la valoración y selección de modelos geotécnicos basados en correlaciones empíricas; en (ii) el desarrollo de predicciones acerca de los resultados esperados en modelos geotécnicos complejos. Se llevan a cabo diferentes aplicaciones a problemas geotécnicos, como es el caso de: (1) En el caso de rocas intactas, se presenta un método Bayesiano para la evaluación de modelos que permiten estimar el módulo de Young a partir de la resistencia a compresión simple (UCS). La metodología desarrollada suministra estimaciones de las incertidumbres de los parámetros y predicciones y es capaz de diferenciar entre las diferentes fuentes de error. Se desarrollan modelos "específicos de roca" para los tipos de roca más comunes y se muestra cómo se pueden "actualizar" esos modelos "iniciales" para incorporar, cuando se encuentra disponible, la nueva información específica del proyecto, reduciendo las incertidumbres del modelo y mejorando sus capacidades predictivas. (2) Para macizos rocosos, se presenta una metodología, fundamentada en un criterio de selección de modelos, que permite determinar el modelo más apropiado, entre un conjunto de candidatos, para estimar el módulo de deformación de un macizo rocoso a partir de un conjunto de datos observados. Una vez que se ha seleccionado el modelo más apropiado, se emplea un método Bayesiano para obtener distribuciones predictivas de los módulos de deformación de macizos rocosos y para actualizarlos con la nueva información específica del proyecto. Este método Bayesiano de actualización puede reducir significativamente la incertidumbre asociada a la predicción, y por lo tanto, afectar las estimaciones que se hagan de la probabilidad de fallo, lo cual es de un interés significativo para los diseños de mecánica de rocas basados en fiabilidad. (3) En las primeras etapas de los diseños de mecánica de rocas, la información acerca de los parámetros geomecánicos y geométricos, las tensiones in-situ o los parámetros de sostenimiento, es, a menudo, escasa o incompleta. Esto plantea dificultades para aplicar las correlaciones empíricas tradicionales que no pueden trabajar con información incompleta para realizar predicciones. Por lo tanto, se propone la utilización de una Red Bayesiana para trabajar con información incompleta y, en particular, se desarrolla un clasificador Naïve Bayes para predecir la probabilidad de ocurrencia de grandes deformaciones (squeezing) en un túnel a partir de cinco parámetros de entrada habitualmente disponibles, al menos parcialmente, en la etapa de diseño. This dissertation employs and develops Bayesian methods to be used in typical geotechnical analyses, with a particular emphasis on (i) the assessment and selection of geotechnical models based on empirical correlations; on (ii) the development of probabilistic predictions of outcomes expected for complex geotechnical models. Examples of application to geotechnical problems are developed, as follows: (1) For intact rocks, we present a Bayesian framework for model assessment to estimate the Young’s moduli based on their UCS. Our approach provides uncertainty estimates of parameters and predictions, and can differentiate among the sources of error. We develop ‘rock-specific’ models for common rock types, and illustrate that such ‘initial’ models can be ‘updated’ to incorporate new project-specific information as it becomes available, reducing model uncertainties and improving their predictive capabilities. (2) For rock masses, we present an approach, based on model selection criteria to select the most appropriate model, among a set of candidate models, to estimate the deformation modulus of a rock mass, given a set of observed data. Once the most appropriate model is selected, a Bayesian framework is employed to develop predictive distributions of the deformation moduli of rock masses, and to update them with new project-specific data. Such Bayesian updating approach can significantly reduce the associated predictive uncertainty, and therefore, affect our computed estimates of probability of failure, which is of significant interest to reliability-based rock engineering design. (3) In the preliminary design stage of rock engineering, the information about geomechanical and geometrical parameters, in situ stress or support parameters is often scarce or incomplete. This poses difficulties in applying traditional empirical correlations that cannot deal with incomplete data to make predictions. Therefore, we propose the use of Bayesian Networks to deal with incomplete data and, in particular, a Naïve Bayes classifier is developed to predict the probability of occurrence of tunnel squeezing based on five input parameters that are commonly available, at least partially, at design stages.

Influence de la taille et de la gradation des particules sur la résistance au cisaillement et le comportement dilatant des matériaux granulaires

Résumé : Cette étude examine l'impact de la taille et de la gradation de particules sur les corrélations théoriques et empiriques existantes les plus connues entre la résistance au cisaillement et le comportement dilatatant des matériaux granulaires en condition de déformation plane et en compression triaxiale drainée. À cette fin, 276 tests de cisaillements symétriques directs et 35 tests de compressions triaxiales drainées ont été menés sur des échantillons composés de billes de basalte (particules rondes), et de sables constitués de particules angulaires (sable de Péribonka et sable d'Eastmain) sur une échelle de 63 µm à 2000 µm afin d'évaluer leur résistance au cisaillement et leur comportement de dilatance sur une vaste échelle de pressions normales et de densités relatives initiales. Premièrement, la fiabilité et l'applicabilité des limites de mesure à l’aide de tests physiques de cisaillements symétriques directs dans l'interprétation de la résistance au cisaillement frictionnel en déformation plane des matériaux granulaires ont été discutées et confirmées par l'usage du code informatique DEM, SiGran. L'accent a été particulièrement mis sur la validation du modèle DEM au moyen de comparaison des résultats des simulations DEM avec leurs équivalents physiques à une échelle macro. Les résultats virtuels DSA sont abordés du point de vue de la coaxialité entre les principales tensions et les principales directions des paliers de pression ainsi que de la déviation de la direction d'extension nulle à partir de la direction horizontale. Les résultats numériques fournissent également des données quantitatives sur les différentes formes d'énergie consommées durant le cisaillement confirmées par d'autres résultats physiques et numériques publiés. Sur la base des postulats précédents, un examen minutieux des résultats des essais de cisaillements directs et de données issues de la littérature a été accompli afin d'évaluer la fiabilité des formules empiriques bien connues de Bolton et Collins et al. avec leurs constantes couramment employées en condition de déformation plane. L'étude montre qu'une application des relations empiriques de force-dilatation de cisaillement avec les constantes proposées par Bolton (1986) et Collins et al. (1992) aux sables ayant une distribution de taille de particules différente peut conduire à surestimer leurs valeurs en terme de force de cisaillement. Dans cette étude, les coefficients des équations de Bolton et Collins et al. ont donc été ajustée afin de prendre en compte les caractéristiques des particules, en particulier le diamètre médian, D50. De manière analogue, les effets microstructuraux imposés par la géométrie interne des particules (par exemple la taille, la forme et la gradation des particules) sur la relation tension-dilatation très connue, celle de Rowe (1962), et son ajustement empirique en condition triaxiale drainée ont été examinés dans cette étude. Une comparaison des prédictions des formules proposées avec les données de force de cisaillement issues de la littérature fournit de nombreuses preuves en faveur des contraintes mises en place au sein des relations existantes de force-dilatation de cisaillement en condition de déformation plane et triaxiale. Ces comparaisons prouvent également que la prise en compte de la taille des grains conduit à des résultats plus tangibles que lorsque la taille de la particule n'est pas considérée. Les formules de force-dilatation ajustées peuvent se révéler avantageuses pour évaluer indépendamment la cohérence des forces de cisaillement déterminées expérimentalement et pour introduire des lois d’écoulement plus précises dans les analyses géotechniques analytiques et numériques.

Geotechnical and structural behavior of the railroad track in large-scale apparatus

This paper analyses the static and dynamic behavior of the railroad track model in laboratory. Measurements of stresses and strains on a large-scale railroad track apparatus were studied. The model includes: compacted soil, representing the final layers of platform, ballast layer, and ties (steel, wooden, and pre-stressed concrete). The soil and soil ballast interface were instrumented with pneumatic stress gauge. Settlement measurement device were positioned at the same levels as the load cells. Loads were applied by hydraulic actuators, statically and dynamically. After the prescribed number of load cycles, in pre-determined intervals, stresses and strains were measured. Observations indicate that stress and strain distributions, transmitted by wooden or steel ties, behave similarly. A more favorable behavior was observed with pre-stressed concrete mono block ties. Non-linear response was observed after a threshold numbers of cycles were surpassed, showing that the strain modulus increases with the numbers of cycles. © 2009 IOS Press.

Geotechnical properties of shelf sdiments from the Prydz Bay

This paper presents a geotechnical characterization of the glacigenic sediments in Prydz Bay, East Antarctica, based on the shipboard physical properties data obtained during Leg 119, combined with results of land-based analyses of 24 whole-round core samples. Main emphasis is placed on the land-based studies, which included oedometer consolidation tests, triaxial and simple shear tests for undrained shear strength, permeability tests in oedometer and triaxial cell, Atterberg limits, and grain-size analyses. The bulk of the tested sediments comprise overconsolidated diamictites of a relatively uniform lithology. The overconsolidation results from a combination of glacial loading and sediment overburden subsequently removed by extensive glacial erosion of the shelf. This leads to downhole profiles of physical properties that have been observed not to change as a function of the thickness of present overburden. A number of fluctuations in the parameters shows a relatively systematic trend and most likely results from changes in the proximity to the ice sheet grounding line in response to variations in the glacial regime. Very low permeabilities mainly result from high preconsolidation stresses (Pc'). Pc' values up to 10,000 kPa were estimated from the oedometer tests, and empirical estimates based on undrained shear strengths (up to 2500 kPa) indicate that the oedometer results are conservative. The diamictites generally classify as inactive, of low to medium plasticity, and they consolidate with little deformation, even when subjected to great stresses. This is the first report of geotechnical data from deep boreholes on the Antarctic continental shelf, but material of similar character can also be expected in other areas around the Antarctic.

Slope stability analyses based on sediment cores of the Ligurian Margin, Southern France

Submarine slope failures of various types and sizes are common along the tectonic and seismically active Ligurian margin, northwestern Mediterranean Sea, primarily because of seismicity up to ~M6, rapid sediment deposition in the Var fluvial system, and steepness of the continental slope (average 11°). We present geophysical, sedimentological and geotechnical results of two distinct slides in water depth >1,500 m: one located on the flank of the Upper Var Valley called Western Slide (WS), another located at the base of continental slope called Eastern Slide (ES). WS is a superficial slide characterized by a slope angle of ~4.6° and shallow scar (~30 m) whereas ES is a deep-seated slide with a lower slope angle (~3°) and deep scar (~100 m). Both areas mainly comprise clayey silt with intermediate plasticity, low water content (30-75 %) and underconsolidation to strong overconsolidation. Upslope undeformed sediments have low undrained shear strength (0-20 kPa) increasing gradually with depth, whereas an abrupt increase in strength up to 200 kPa occurs at a depth of ~3.6 m in the headwall of WS and ~1.0 m in the headwall of ES. These boundaries are interpreted as earlier failure planes that have been covered by hemipelagite or talus from upslope after landslide emplacement. Infinite slope stability analyses indicate both sites are stable under static conditions; however, slope failure may occur in undrained earthquake condition. Peak earthquake acceleration from 0.09 g on WS and 0.12 g on ES, i.e. M5-5.3 earthquakes on the spot, would be required to induce slope instability. Different failure styles include rapid sedimentation on steep canyon flanks with undercutting causing superficial slides in the west and an earthquake on the adjacent Marcel fault to trigger a deep-seated slide in the east.

Geotechnical properties of Lau Basin sediments

Distribution of pore space and degree of cementation appear to be the main factors controlling the permeability of sediments retrieved from the Lau Basin. The undisturbed microfabrics of two lithologies, nannofossil ooze and vitric sandy silt, commonly found at Holes 834A, 835A, 838A, and 839Aof Leg 135 were examined by scanning electron microscopy equipped with energy dispersive X-ray spectral analysis and image analysis systems. The results of these analyses were compared with laboratory determinations of porosity, grain-size distribution, and permeability on discrete samples from the same sediment depths. The permeability of the vitric sandy silt is 3-5 orders of magnitude higher than the nannofossil ooze samples. The porosity of nannofossil ooze ranges from 6% to 12% greater than the porosity of vitric sandy silt, which partially reflects the finer texture of nannofossil ooze. Although the correlation of higher porosity with lower permeability is not surprising, factors other than simply grain-size distribution must be invoked to explain the large differences in permeability found in these samples.

A linearization-based system reliability method with application to geotechnical analysis

Los análisis de fiabilidad representan una herramienta adecuada para contemplar las incertidumbres inherentes que existen en los parámetros geotécnicos. En esta Tesis Doctoral se desarrolla una metodología basada en una linealización sencilla, que emplea aproximaciones de primer o segundo orden, para evaluar eficientemente la fiabilidad del sistema en los problemas geotécnicos. En primer lugar, se emplean diferentes métodos para analizar la fiabilidad de dos aspectos propios del diseño de los túneles: la estabilidad del frente y el comportamiento del sostenimiento. Se aplican varias metodologías de fiabilidad — el Método de Fiabilidad de Primer Orden (FORM), el Método de Fiabilidad de Segundo Orden (SORM) y el Muestreo por Importancia (IS). Los resultados muestran que los tipos de distribución y las estructuras de correlación consideradas para todas las variables aleatorias tienen una influencia significativa en los resultados de fiabilidad, lo cual remarca la importancia de una adecuada caracterización de las incertidumbres geotécnicas en las aplicaciones prácticas. Los resultados también muestran que tanto el FORM como el SORM pueden emplearse para estimar la fiabilidad del sostenimiento de un túnel y que el SORM puede mejorar el FORM con un esfuerzo computacional adicional aceptable. Posteriormente, se desarrolla una metodología de linealización para evaluar la fiabilidad del sistema en los problemas geotécnicos. Esta metodología solamente necesita la información proporcionada por el FORM: el vector de índices de fiabilidad de las funciones de estado límite (LSFs) que componen el sistema y su matriz de correlación. Se analizan dos problemas geotécnicos comunes —la estabilidad de un talud en un suelo estratificado y un túnel circular excavado en roca— para demostrar la sencillez, precisión y eficiencia del procedimiento propuesto. Asimismo, se reflejan las ventajas de la metodología de linealización con respecto a las herramientas computacionales alternativas. Igualmente se muestra que, en el caso de que resulte necesario, se puede emplear el SORM —que aproxima la verdadera LSF mejor que el FORM— para calcular estimaciones más precisas de la fiabilidad del sistema. Finalmente, se presenta una nueva metodología que emplea Algoritmos Genéticos para identificar, de manera precisa, las superficies de deslizamiento representativas (RSSs) de taludes en suelos estratificados, las cuales se emplean posteriormente para estimar la fiabilidad del sistema, empleando la metodología de linealización propuesta. Se adoptan tres taludes en suelos estratificados característicos para demostrar la eficiencia, precisión y robustez del procedimiento propuesto y se discuten las ventajas del mismo con respecto a otros métodos alternativos. Los resultados muestran que la metodología propuesta da estimaciones de fiabilidad que mejoran los resultados previamente publicados, enfatizando la importancia de hallar buenas RSSs —y, especialmente, adecuadas (desde un punto de vista probabilístico) superficies de deslizamiento críticas que podrían ser no-circulares— para obtener estimaciones acertadas de la fiabilidad de taludes en suelos. Reliability analyses provide an adequate tool to consider the inherent uncertainties that exist in geotechnical parameters. This dissertation develops a simple linearization-based approach, that uses first or second order approximations, to efficiently evaluate the system reliability of geotechnical problems. First, reliability methods are employed to analyze the reliability of two tunnel design aspects: face stability and performance of support systems. Several reliability approaches —the first order reliability method (FORM), the second order reliability method (SORM), the response surface method (RSM) and importance sampling (IS)— are employed, with results showing that the assumed distribution types and correlation structures for all random variables have a significant effect on the reliability results. This emphasizes the importance of an adequate characterization of geotechnical uncertainties for practical applications. Results also show that both FORM and SORM can be used to estimate the reliability of tunnel-support systems; and that SORM can outperform FORM with an acceptable additional computational effort. A linearization approach is then developed to evaluate the system reliability of series geotechnical problems. The approach only needs information provided by FORM: the vector of reliability indices of the limit state functions (LSFs) composing the system, and their correlation matrix. Two common geotechnical problems —the stability of a slope in layered soil and a circular tunnel in rock— are employed to demonstrate the simplicity, accuracy and efficiency of the suggested procedure. Advantages of the linearization approach with respect to alternative computational tools are discussed. It is also found that, if necessary, SORM —that approximates the true LSF better than FORM— can be employed to compute better estimations of the system’s reliability. Finally, a new approach using Genetic Algorithms (GAs) is presented to identify the fully specified representative slip surfaces (RSSs) of layered soil slopes, and such RSSs are then employed to estimate the system reliability of slopes, using our proposed linearization approach. Three typical benchmark-slopes with layered soils are adopted to demonstrate the efficiency, accuracy and robustness of the suggested procedure, and advantages of the proposed method with respect to alternative methods are discussed. Results show that the proposed approach provides reliability estimates that improve previously published results, emphasizing the importance of finding good RSSs —and, especially, good (probabilistic) critical slip surfaces that might be non-circular— to obtain good estimations of the reliability of soil slope systems.

Public perception of carbon capture and storage in relation to hydraulic fracturing: content analyses, survey results, and policy proposals

Carbon capture and storage (CCS) can contribute significantly to addressing the global greenhouse gas (GHG) emissions problem. Despite widespread political support, CCS remains unknown to the general public. Public perception researchers have found that, when asked, the public is relatively unfamiliar with CCS yet many individuals voice specific safety concerns regarding the technology. We believe this leads many stakeholders conflate CCS with the better-known and more visible technology hydraulic fracturing (fracking). We support this with content analysis of media coverage, web analytics, and public lobbying records. Furthermore, we present results from a survey of United States residents. This first-of-its-kind survey assessed participants’ knowledge, opinions and support of CCS and fracking technologies. The survey showed that participants had more knowledge of fracking than CCS, and that knowledge of fracking made participants less willing to support CCS projects. Additionally, it showed that participants viewed the two technologies as having similar risks and similar risk intensities. In the CCS stakeholder literature, judgment and decision-making (JDM) frameworks are noticeably absent, and public perception is not discussed using any cognitive biases as a way of understanding or explaining irrational decisions, yet these survey results show evidence of both anchoring bias and the ambiguity effect. Public acceptance of CCS is essential for a national low-carbon future plan. In conclusion, we propose changes in communications and incentives as programs to increase support of CCS.

Evaluation of One-Dimensional Seismic Site Response Analyses at Small to Large Strain Levels

Thesis (Master's)--University of Washington, 2016-08

GEOTECHNICAL AND ENVIRONMENTAL IMPACTS OF STEEL SLAG IN EMBANKMENTS

Steel slag is a byproduct of iron and steel production by the metallurgical industries. Annually, 21 million tons of steel slag is produced in the United States. Most of the slag is landfilled, which represents a significant economic loss and a waste of valuable land space. Steel slag has great potential for the construction of highway embankments; however, its use has been limited due to its high swelling potential and alkalinity. The swelling potential of steel slags may lead to deterioration of the structural stability of highways, and high alkalinity poses an environmental challenge as it affects the leaching behavior of trace metals. This study seeks a methodology that promotes the use of steel slag in highway embankments by minimizing these two main disadvantages. Accelerated swelling tests were conducted to evaluate the swelling behavior of pure steel slag and water treatment residual (WTR) treated steel slag, where WTR is an alum-rich by-product of drinking water treatment plants. Sequential batch tests and column leach tests, as well as two different numerical analyses, UMDSurf and WiscLEACH, were carried out to check the environmental suitability of the methods. Tests were conducted to study the effect of a common borrow fill material that encapsulated the slag in the embankment and the effects of two subgrade soils on the chemical properties of slag leachate. The results indicated that an increase in WTR content in the steel slag-WTR mixtures yields a decrease in pH and most of the leached metal concentrations, except aluminum. The change in the levels of pH, after passing through encapsulation and subgrade, depends on the natural pHs of materials.