scispace - formally typeset
Search or ask a question

Showing papers by "T. G. Sitharam published in 2011"


Journal ArticleDOI
TL;DR: 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 and highlights the capability of the SVM over the ANN models.
Abstract: . 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 [(N1)60] and cyclic stress ratio (CSR). Further, an attempt has been made to simplify the models, requiring only the two parameters [(N1)60 and peck ground acceleration (amax/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.

155 citations


Journal ArticleDOI
TL;DR: In this article, the effect of non-plastic fines (silt) on the undrained monotonic response of saturated and isotropically consolidated sand specimens prepared to various measures of their density was studied in detail through various approaches.
Abstract: The effect of non-plastic fines (silt) on the undrained monotonic response of saturated and isotropically consolidated sand specimens prepared to various measures of their density was studied in detail through various approaches namely gross void ratio approach, relative density approach, sand skeleton void ratio approach, and interfine void ratio approach. Specimens of 50 mm in diameter and 100 mm in height were tested at a rate of loading of 0.6 mm/min for this purpose. The limiting silt content and the relative density of a specimen were found to influence the undrained monotonic response of sand–silt mixtures to a great extent. Undrained monotonic response was observed to be independent of silt content at very high relative densities; however the presence of fines significantly influenced this response of loose to medium dense specimens. Individual and combined analyses of undrained monotonic peak strengths which are closely related to the liquefaction related problems have been done in this paper to ...

39 citations


Journal ArticleDOI
TL;DR: In this paper, the potential of support vector machine (SVM) based classification approach has been used to assess the liquefaction potential from actual shear wave velocity data, which is an approximate implementation of a structural risk minimization (SRM) induction principle is done, which aims at minimizing a bound on the generalization error of a model rather than minimizing only the mean square error over the data set.

32 citations


Journal ArticleDOI
TL;DR: In this article, a series of stress controlled cyclic and strain controlled monotonic triaxial tests was carried out on sand-silt mixture specimens of 50 mm diameter and 100 mm height with varying silt content.
Abstract: In an attempt to correlate the monotonic peak strength and the cyclic strength of sand-silt mixtures over a wide range of parameters and to clarify some of the existing confusing conclusions in the literature regarding the undrained strength response of sand-silt mixtures, a series of stress controlled cyclic and strain controlled monotonic triaxial tests was carried out on sand-silt mixture specimens of 50 mm diameter and 100 mm height with varying silt content. In these experiments, various measures of sample density was adopted through different approaches such as constant gross void ratio approach, constant relative density approach, constant sand skeleton void ratio approach, and constant interfine void ratio approach. Also the effect of relative density and confining pressure on these strengths was studied. It is observed that the limiting fines content and the relative density of a specimen play the key role in deciding the cyclic and monotonic resistance of sand-silt mixtures when studied through any approach. For any silt content with relative density more than 70%, cyclic and monotonic resistances are observed to be independent of silt content. When the undrained cyclic strengths of these specimens are plotted against their respective undrained monotonic peak strengths, it is observed that there exists a definite exponential relationship between the two with an excellent correlation coefficient. An expression is proposed in this regard to help engineers assess the cyclic strength of sand-silt mixtures from monotonic test results.

19 citations


Journal ArticleDOI
TL;DR: In this article, an attempt has been made to evaluate the spatial variation of peak horizontal acceleration (PHA) and spectral acceleration (SA) values at rock level for south India based on the probabilistic seismic hazard analysis (PSHA).
Abstract: In this work, an attempt has been made to evaluate the spatial variation of peak horizontal acceleration (PHA) and spectral acceleration (SA) values at rock level for south India based on the probabilistic seismic hazard analysis (PSHA). These values were estimated by considering the uncertainties involved in magnitude, hypocentral distance and attenuation of seismic waves. Different models were used for the hazard evaluation, and they were combined together using a logic tree approach. For evaluating the seismic hazard, the study area was divided into small grids of size 0.1° × 0.1°, and the hazard parameters were calculated at the centre of each of these grid cells by considering all the seismic sources within a radius of 300 km. Rock level PHA values and SA at 1 s corresponding to 10% probability of exceedance in 50 years were evaluated for all the grid points. Maps showing the spatial variation of rock level PHA values and SA at 1 s for the entire south India are presented in this paper. To compare the seismic hazard for some of the important cities, the seismic hazard curves and the uniform hazard response spectrum (UHRS) at rock level with 10% probability of exceedance in 50 years are also presented in this work.

18 citations


Journal ArticleDOI
TL;DR: In this paper, a simple and ordinary kriging model is used to compute variance of estimated reduced level of the rock in the subsurface of Bangalore and a new type of cross-validation analysis proves the robustness of the developed models.

14 citations


Journal ArticleDOI
TL;DR: In this article, the authors evaluated the seismic hazard of the South Indian Peninsular region based on the deterministic seismic hazard analysis (DSHA) methodology in order to consider the epistemic uncertainties in a better manner, a logic tree approach was adopted in the evaluation of seismic hazard.
Abstract: The most important seismic hazard parameters required to demarcate seismic zones are the peak horizontal acceleration (PHA) and spectral acceleration (SA). The two approaches for evaluation of seismic hazard are the probabilistic seismic hazard analysis and the deterministic seismic hazard analysis (DSHA). The present study evaluates the seismic hazard of the South Indian Peninsular region based on the DSHA methodology. In order to consider the epistemic uncertainties in a better manner, a logic tree approach was adopted in the evaluation of seismic hazard. Two types of seismic sources and three different attenuation relations were used in the analysis. The spatial variation of PHA (mean and 84th percentile values) and SA values for 1 Hz and 10 Hz at bedrock level (84th percentile values) for the entire study area were evaluated and the results are presented here. The surface level peak ground acceleration (PGA) values will be different from that of the bedrock level values due to the local site condition...

7 citations


01 Jan 2011
TL;DR: An updated probabilistic seismic hazard map of Kerala and its vicinity covering 5°-16°N and 72°-81°E was prepared by compiling the data from different national and international agencies.
Abstract: In view of the major advancement made in understanding the seismicity and seismotectonics of the region, an updated probabilistic seismic hazard map of Kerala and its vicinity covering 5°-16°N and 72°-81°E was prepared. The earthquake catalogue was prepared by compiling the data from different national and international agencies. The sesismotectonic map of the study area was prepared by considering the faults, lineaments and the shear zones which are associated with earthquakes of magnitude 3 and above. The seismicity parameters were estimated for the region and and the hazard parameters (Rock level Peak Horizontal Acceleration (PHA) and spectral accelerations for periods 0.1s and 1s) were calculated considering all the seismic sources within a radius of 300 km. Epistemic uncertainty in hazard has been tackled within a logic-tree framework considering three types of source models and three attenuation models. The contour maps showing spatial variation of spectral acceleration values both at bed rock and surface level are presented in the paper.

5 citations


Proceedings ArticleDOI
11 Mar 2011
TL;DR: In this paper, the authors explored the potential of discrete element simulation (DEM) to model the behavior of reinforced granular soils from grain scale level using a chain of circular particles by assigning contact bond between particles horizontally at all predetermined contact locations.
Abstract: This paper explores the potential of Discrete Element Simulation (DEM) to model the behavior of reinforced granular soils. Biaxial element test were carried out on these assembles with and with out horizontal reinforcement layers to understand the behavior of reinforced granular soils from grain scale level. Numerical simulations were carried out using PFC 2D with assemblies of 1800 uni sized circular particles of size 0.1 mm. In the current investigation, reinforcing element was modeled by a chain of circular particles by assigning contact bond between particles horizontally at all predetermined contact locations. Influence of parameters such as number of reinforcement layer, thickness and packing geometry on the behavior of reinforced granular materials were studied and reported.

5 citations



01 Jan 2011
TL;DR: In this paper, a performance-based approach was proposed to evaluate the liquefaction potential of the Bhuj earthquake based on the corrected surface level peak ground acceleration (PGA) values.
Abstract: The Bhuj earthquake of 2001 is considered as the deadliest stable continental shield region earthquake. One of the main reasons for heavy damage during the Bhuj earthquake was the effects of site amplification and liquefaction. This paper tries to evaluate one of these geotechnical aspects - liquefaction, based on a performance-based approach. The seismic hazard assessment of Gujarat (at bed rock level) was evaluated using Probabilistic Seismic hazard Assessment (PSHA) method. Based on the rock level peak horizontal acceleration (PHA) values obtained, the surface level peak ground acceleration (PGA) values were evaluated by amplification coefficients suggested by Raghu Kanth and Iyengar for Peninsular India. Using the PGA values, the liquefaction potential was evaluated based on a performance-based approach. The liquefaction potential was evaluated based on the corrected SPT value required to prevent liquefaction. The spatial variation of SPT values required to prevent liquefaction for return periods of 475 and 2500 years are presented in this paper. The results obtained in this study can be used for macro level planning and liquefaction hazard mitigation of Gujarat.

Journal ArticleDOI
TL;DR: In this paper, the results of the dynamic properties such as shear modulus and damping ratio of sandy soils, especially at large shear strain levels (> 0.2%), were presented.
Abstract: This paper presents the results of the dynamic properties such as shear modulus and damping ratio of sandy soils, especially at large shear strain levels (>0.2%). A series of strain controlled cyclic triaxial tests were carried out on sand samples collected from the earthquake affected areas of Gujarat, India. Laboratory investigations were conducted on natural sand (Base sand) and clean sand samples. The shear modulus and damping ratios have been estimated based on the first cycle information. The effect of different parameters such as number of loading cycles, relative density, confining pressures, and non-plastic fines on the dynamic properties of the soils has been studied. It was observed that shear modulus decreases with an increase in the percentage of non-plastic fines. However, a slight increase in damping ratio was observed with increase in non-plastic fines.


01 Jan 2011
TL;DR: In this article, the authors deal with the variation of stresses and deformations around a twin tunnel using discrete element approach, and the numerical simulations adopting discrete element method predicts the mechanical behaviour of soil/rock masses through grain scale modeling.
Abstract: In recent times due to rapid urbanization and sustainable building practices, there is an exponential increase in the use of underground space for urban development. This has in turn created the necessity of understanding the behaviour of underground structures like tunnels under various conditions. Twin tunnels are two closely constructed tunnels. An underground structure is basically acted upon by the pressure due to overburden and the medium surrounding the underground space. This ground which supports the underground structure acts as an additional constraint and hence the underground structures are not as vulnerable as compared to their surface counterparts during an earthquake event. One of the major challenges in underground space construction is the control of ground stability and deformations. Numerical modeling has become one of the most powerful tools in understanding the response of underground spaces such as tunnels and caverns. The evaluation of the mechanical behaviour of the ground is very significant in the case of urban underground spaces. The excavation of the underground structure disturbs the soil and rock masses. This results in ground settlement and subsurface movements leading to significant damages to the existing structures. This paper deals with the variation of stresses and deformations around a twin tunnel using discrete element approach. The numerical simulations adopting discrete element method predicts the mechanical behaviour of soil/rock masses through grain scale modeling. Here, a three dimensional ground surface is modeled in a weak rock like sandstone and two circular tunnels which are closely spaced are excavated inside the modeled ground surface. The response of the system after excavating the tunnels along with the effect of spacing between the tunnels is studied and the results are presented

01 Jan 2011
TL;DR: The fluid coupled discrete element method is an effective technique that helps to model the interactions of fluid flow with particles and has wide range of applications in geotechnical engineering.
Abstract: Flow coupled Discrete Element Method is an effective technique that helps to model the interactions of fluid flow with particles. Coupled models allow particle level studies and have wide range of applications in geotechnical engineering. This paper gives an overview of the fluid coupled methods, explains briefly the forces that act on the soil particles as a result of fluid flow through the system and the formulations that are used. The method of coupling between solid phase and fluid phase is explained. The paper illustrates the applications of fluid coupled models in geotechnical engineering.

Journal ArticleDOI
TL;DR: In this article, the undrained cyclic response of sand-silt mixture specimens prepared to constant sand skeleton void ratios and with silt contents less than the limiting silt content was studied under three headings namely the cyclic resistance, pore pressure generation, and the first cycle shear modulus at a particular cyclic stress ratio.
Abstract: The undrained cyclic response of sand-silt mixture specimens prepared to constant sand skeleton void ratios and with silt contents less than the limiting silt content was studied under three headings namely the cyclic resistance, pore pressure generation, and the first cycle shear modulus at a particular cyclic stress ratio. A similar behaviour was observed in all these strength parameters at both the constant sand skeleton void ratios chosen. Both the cyclic resistance and the first cycle shear modulus at a particular cyclic stress ratio were seen to initially remain constant till a particular silt content corresponding to a relative density of around 70 % but a sudden increase in both the strength parameters were observed beyond that. A just reverse behaviour was observed in the rate of generation of excess pore water pressure. The rate of generation of excess pore water pressure with respect to cycles of loading was found to be relatively same till this silt content but a sudden fall in the rat...