Machine learning modelling for predicting soil liquefaction susceptibility

This study describes two machine learning techniques applied to predict liquefaction susceptibility of soil based on the standard penetration test (SPT) data from the 1999 Chi-Chi, Taiwan earthquake. The first machine learning technique which uses Artificial Neural Network (ANN) based on multi-layer perceptions (MLP) that are trained with Levenberg-Marquardt backpropagation algorithm. The second machine learning technique uses the Support Vector machine (SVM) that is firmly based on the theory of statistical learning theory, uses classification technique. ANN and SVM have been developed to predict liquefaction susceptibility using corrected SPT (N-1)(60)] and cyclic stress ratio (CSR). Further, an attempt has been made to simplify the models, requiring only the two parameters (N-1)(60) and peck ground acceleration (a(max)/g)], for the prediction of liquefaction susceptibility. The developed ANN and SVM models have also been applied to different case histories available globally. The paper also highlights the capability of the SVM over the ANN models.

Collapse Behavior of an Artificially Cemented Clayey Silt

The present study examines the role of interparticle cementation in the collapse behavior of two partly saturated (S-r = 4 to 12%) and very highly porous (initial void ratio = 1.5 to 2) laboratory-desiccated clayey silt specimens containing varying amounts (5 and 15% by dry weight of the respective specimens) of the cementitious iron oxides hematite and goethite, which are generally encountered in tropical residual soils. Kaolinite is the representative clay mineral of the soil matrix used for this research. Interparticle cementation by the crystalline iron oxides was generated in the laboratory by repeated (six times) wetting and drying of the iron-hydroxide-admixed clayey silt specimens under ambient conditions of temperature and humidity. Results showed that, for a given laboratory-desiccated clayey silt specimen (i.e., a specimen containing 5 or 15% of iron oxide on a dry weight basis), the amount of collapse (represented by Delta epsilon, the change in vertical strain upon wetting under constant pressure) increases with an increase in the experimental loading under which the specimen is inundated. The laboratory results also show that the desiccated specimen with a higher iron oxide content (containing 15% iron oxide by dry weight of the desiccated specimen) in spite of a lower dry unit weight (gamma(d) = 8.8 kN/m(3)) undergoes a lesser amount of collapse on soaking under a constant external stress (50 or 100 kPa) than the desiccated specimen with a lower iron oxide content (i.e., containing 5% iron oxide by dry weight of the desiccated specimen, gamma(d) = 10.4 KN/m(3)). Based on the X-ray diffraction results and the stress-strain relationships obtained from isotropically consolidated undrained triaxial tests, it is suggested that the laboratory-desiccated specimens are characterized by a metastable bonding provided by capillary suction and the crystalline iron oxides. On soaking under load owing to the loss of the metastable bonding, collapse of the laboratory-desiccated specimens occurs. Also, in the case of the laboratory-desiccated specimen with a higher iron oxide content, the presence of a stronger interparticle cementation (due to a greater abundance of crystalline iron oxides) and a higher initial moisture content are considered responsible for the specimen exhibiting a lower amount of collapse in comparison to that exhibited by the desiccated specimen with a lesser iron oxide content.

Potential-distance relationships of clay-water systems considering the Stern theory

This paper deals with the use of Stem theory as applied to a clay-water electrolyte system, which is more realistic to understand the force system at micro level man the Gouy-Chapman theory. The influence of the Stern layer on potential-distance relationship has been presented quantitatively for certain specified clay-water systems and the results are compared with the Gouy-Chapman model. A detailed parametric study concerning the number of adsorption spots on the clay platelet, the thickness of the Stern layer, specific adsorption potential and the value of dielectric constant of the pore fluid in the Stern layer, was carried out. This study investigates that the potential obtained at any distance using the Stern theory is higher than that obtained by the Gouy-Chapman theory. The hydrated size of the ion is found to have a significant influence on the potential-distance relationship for a given clay, pore fluid characteristics and valence of the exchangeable ion.

Permeability and Compressibility Behavior of Bentonite-Sand/Soil Mixes

As a seepage barrier slurry trench material should have a relatively low coefficient of permeability, in the range of 10(-7) cm/s, and at the same time should be compatible with surrounding material with regard to compressibility. Although bentonite-sand/soil mixes are used widely, there is no specific engineering approach to proportion these mixes that satisfies the above practical requirements. In this paper, a generalized approach is presented for predicting the permeability and compressibility characteristics of mixes with minimum input parameters. This approach will be helpful in proportioning mixes and predicting corresponding changes in engineering behavior. It is possible to proportion a mix to arrive at the required compressibility without affecting the permeability. This is explained using an illustrative example.

Characteristic water contents of a fine-grained soil-water system

This paper critically appraises the limitations of the liquid-limit water content of clayey soils determined conventionally either by percussion cup or by the cone penetration method. It is shown that the conventional liquid limit and plastic limit are arbitrary, strength-based water contents and that they cannot represent the plasticity limits, and that the state of the soil-water system at the conventional liquid limit does not correspond to a stress-free reference state. The present investigation identifies three characteristic limiting water contents for a soil-water system which have well-defined, unique mechanisms controlling them, namely the free swell limit, settling limit and shrinkage limit. Simple procedures for the determination of the free swell limit and settling limit of natural soils are presented. The settling limit is shown to be the 'real liquid limit' of any clayey soil. With a number of experimental illustrations, it is clearly shown that the settling limit represents the maximum water-holding capacity of clayey soils and that it corresponds to the stress-free reference state.

Relative performances of textural models in estimating soil moisture characteristic

The soil moisture characteristic (SMC) forms an important input to mathematical models of water and solute transport in the unsaturated-soil zone. Owing to their simplicity and ease of use, texture-based regression models are commonly used to estimate the SMC from basic soil properties. In this study, the performances of six such regression models were evaluated on three soils. Moisture characteristics generated by the regression models were statistically compared with the characteristics developed independently from laboratory and in-situ retention data of the soil profiles. Results of the statistical performance evaluation, while providing useful information on the errors involved in estimating the SMC, also highlighted the importance of the nature of the data set underlying the regression models. Among the models evaluated, the one possessing an underlying data set of in-situ measurements was found to be the best estimator of the in-situ SMC for all the soils. Considerable errors arose when a textural model based on laboratory data was used to estimate the field retention characteristics of unsaturated soils.

Comparison of electrical resistivity by geophysical method and neutron probe logging for soil moisture monitoring in a forested watershed

Geophysical methods are becoming more popular nowadays in the field of hydrology due to their time and space efficiency. So an attempt has been made here to relate electrical resistivity with soil moisture content in the field. The experiments were carried out in an experimental watershed `Mulehole' in southern India, which is a forested watershed with approximately 80% red soil. Five auger holes were drilled to perform the soil moisture and electrical resistivity measurements in a toposequence having red and black soils, with sandy weathered soil at the bottom. Soil moisture was measured using neutron probe and electrical resistivity was measured using electrical logging tool. The results indicate that electrical resistivity measurements can be used to measure soil moisture content for red soils only.

Triaxial A-value versus swell or collapse for compacted soil - Discussion

A discussion of a technical note with the aforementioned title by Day and Marsh, published in this journal (Volume 121, Number 7, July 1995), is presented. Discussers Robinson and Allam assert that the authors' application of the pore-pressure parameter A to predict and quantify swell or collapse of compacted soils is hard to use because the authors visualize the collapse-swell phenomenon to occur in compacted soils broadly classified as sands and clays. The literature demonstrates that mineralogy has an important role in the volume change behavior of fine-grained soils. Robinson and Allam state that the A-value measurements may not completely predict the type of volume change anticipated in compacted soils on soaking without soil clay mineralogy details. Discussion is followed by closure from the authors.

California bearing ratio behavior of cement-stabilized fly ash-soil mixes

Thermal power stations using pulverized coal as fuel generate large quantities of fly ash as a byproduct, which has created environmental and disposal problems. Using fly ash for gainful applications will solve these problems. Among the various possible uses for fly ash, the most massive and effective utilization is in geotechnical engineering applications like backfill material, construction of embankments, as a subbase material, etc. A proper understanding of fly ash-soil mixes is likely to provide viable solutions for its large-scale utilization. Earlier studies initiated in the laboratory have resulted in a good understanding of the California Bearing Ratio (CBR) behavior of fly ash-soil mixes. Subsequently, in order to increase the CBR value, cement has been tried as an additive to fly ash-soil mixes. This paper reports the results.

Collapse behaviour of a residual soil

Residually derived red soils occur in Bangalore District of Karnataka State, India. The porous and unsaturated nature of the red soils makes them susceptible to collapse on wetting under load. The present study analyses the collapse behaviour of an unsaturated bonded (undisturbed) red soil from Bangalore referenced to tests on samples in an unbonded (remoulded) state. A filter paper method was used to determine the matric suction of the bonded and unbonded specimens, and mercury intrusion porosimetry (MIP) was used to determine their soil structure. Analysis of the experimental results shows that bonding plays an important role in the collapse behaviour of the unsaturated residual soil. The results of the study also provide insight into the volume change behaviour of unsaturated bonded soils on wetting within and beyond the yield locus.

Reliability Based Design Optimization for Internal Stability of Reinforced Soil Structures Using Pseudo‐Static Method

Seismic design of reinforced soil structures involves many uncertainties that arise from the backfill soil properties and tensile strength of the reinforcement which is not addressed in current design guidelines. This paper highlights the significance of variability in the internal stability assessment of reinforced soil structures. Reliability analysis is applied to estimate probability of failure and pseudo‐static approach has been used for the calculation of the tensile strength and length of the reinforcement needed to maintain the internal stability against tension and pullout failures. Logarithmic spiral failure surface has been considered in conjunction with the limit equilibrium method. Two modes of failure namely, tension failure and pullout failure have been considered. The influence of variations of the backfill soil friction angle, the tensile strength of reinforcement, horizontal seismic acceleration on the reliability index against tension failure and pullout failure of reinforced earth structure have been discussed.

Effect of aging on swelling and swell-shrink behavior of a compacted expansive soil

Effect of aging on swelling and swell-shrink behavior of a compacted expansive soil is investigated in this paper. An expansive soil having a liquid limit of 100% is used for this purpose. Compacted specimens were prepared and aged for a predetermined number of days (7, 15, 30, and 90 days) to study their swelling and swell-shrink behavior. It has been shown that aging improves the resistance to compression of compacted specimens. The swelling potentials of specimens also decreased with aging. The dominant factors that influence the aging effects are the water content and degree of saturation at the beginning of the aging process. The changed behavior of aged specimens is attributed to particle rearrangements and formation of bonds, which affect the surface area absorbing water during swelling. The cyclic swell-shrink tests on aged specimens indicated that the differences in vertical displacement during the first swelling were eliminated in the subsequent cycles when specimens were shrunk more, but the aging effect was found to persist with cycles for specimens subjected to lower shrinkage magnitudes.

Role of soil structure and matric suction in collapse of a compacted clay soil

This paper examines the role of microstructure and matric suction in the collapse behavior of a compacted clay soil from Bangalore District in Karnataka State, India. The microstructure of the compacted specimens was examined by mercury intrusion porosimetry (MIP), and the ASTM Filter Paper Method was used to determine their matric suction. The microstructure and matric suction of the compacted specimens were changed by varying their compaction water content, dry density, and clay content (< 2 mum fraction). Experimental results showed that relative abundance of coarse (60 to 6 mum) pores was mainly affected by increasing the dry density of the specimens from 1.49 to 1.77 g/cm(3). The relative abundance of coarse and fine (0.01 to 0.002 mum) pores was affected by increasing the compaction water content from 10.6 to 26.4%. Variations in dry density, compaction water content, and clay contents notably affected the matric suction of the compacted specimens. The collapse behavior of the compacted specimens is explained from analysis of the MIP and matric suction results.

Reliability-based load and resistance factors for soil-nail walls

Existing soil nailing design methodologies are essentially based on limit equilibrium principles that together with a lumped factor of safety or a set of partial factors on the material parameters and loads account for uncertainties in design input parameter values. Recent trends in the development of design procedures for earth retaining structures are towards load and resistance factor design (LRFD). In the present study, a methodology for the use of LRFD in the context of soil-nail walls is proposed and a procedure to determine reliability-based load and resistance factors is illustrated for important strength limit states with reference to a 10 m high soil-nail wall. The need for separate partial factors for each limit state is highlighted, and the proposed factors are compared with those existing in the literature.