Department of Civil Engineering

Permanent URI for this collectionhttp://localhost:4000/handle/123456789/1927

Browse

Filter results by year or month
Now showing 1 - 20 of 1393
  • No Thumbnail Available
    Item
    A study of fatigue crack growth IS-1020 steel under constant-amplitude loading
    (Elsiever, 1992) Singh, Shamsher Bahadur
    Crack propagation experiments were performed on IS-1020 steel for various load ranges and stress ratios. At constant maximum load, the life of the specimen increased as the load ratio increased. The crack growth data were analysed in terms of as a function of stress ratio R. The data covered R values of 0, 0·1, 0·2, and 0·3, and a good relation was obtained for . A crack growth rate equation was also developed.
  • No Thumbnail Available
    Item
    Investigation of fatigue crack growth after a single cycle peak overload in IS 1020 steel
    (Elsiever, 1992) Singh, Shamsher Bahadur
    Crack propagation experiments were conducted on IS 1020 steel for various overload ratios (1·2, 1·3 and 1·4). On the basis of these experiments one power law is developed to predict the crack propagation delay period. The delay period after application of a single overload was found to increase as the magnitude of the overload increased. Crack growth also decreased after the application of an overload cycle but after a certain number of cycles it tended to return to the crack propagation rate for constant amplitude loading (CAL).
  • No Thumbnail Available
    Item
    Experimental observations of fatigue crack growth in IS-1020 steel under constant amplitude loading
    (Elsiever, 1993) Singh, Shamsher Bahadur
    Crack propagation experiments were performed on a 0·2% C 0·9% Mn (IS-1020) steel for various load ranges and stress ratios. At constant load range, the life of the specimen decreased as the load ratio increased. The crack growth data were analysed in terms of as a function of stress ratio R. The data covered R values for 0, 0·15, 0·30, 0·35 and 0·40 and a good relation was obtained for . A crack growth rate equation was also developed.
  • No Thumbnail Available
    Item
    Sensitivity‐Based Weighted‐Average in Structural Damage Assessment
    (ASCE, 1994) Barai, Sudhir Kumar
    To quantify the global damageability of a structure, the weighted‐average function defined in terms of damage condition and the importance or weightages of structural elements has generally been used in literature. Judgment of structural importance of a structural element is a difficult task. The opinion provided by experts could have different perspective or bias, which may reflect in the overall assessment process. Structural analysis being precise these days, structural importance of a member could be realistically estimated using standard structural procedures. In this paper it is proposed that the weightage computation could be linked to damage sensitivity of the element response. It is argued that the weighted‐average computation based on damage sensitivity is a more realistic index for integrity assessment. In this perspective, the concept of damage sensitivity and its computational aspects based on the finite‐element method are also presented. Finally, examples of weighted‐average computation comprising fuzzy sets and importance factor obtained from normalized damage sensitivity are illustrated.
  • No Thumbnail Available
    Item
    Neural Networks for Damage Detection in Steel Railway Bridges
    (IABSE, 1995) Barai, Sudhir Kumar
    The paper presents Artificial Neural Networks developed for typical steel railway bridges for the purpose of damage detection. Multilayer perceptrons have been used for generating the architecture for the bridges of different configurations. The back propagation algorithm has been adopted for training the network with simulated damage states. The training pairs have been generated using a standard finite element program. The weights of the trained networks have been stored and can be used as a knowledge source independently. It is demonstrated that the trained networks have practical relevance.
  • No Thumbnail Available
    Item
    Performance of the generalized delta rule in structural damage detection
    (Elsiever, 1995-04) Barai, Sudhir Kumar
    The paper examines the suitability of the generalized data rule in training artificial neural networks (ANN) for damage identification in structures. Several multilayer perceptron architectures are investigated for a typical bridge truss structure with simulated damage states generated randomly. The training samples have been generated in terms of measurable structural parameters (displacements and strains) at suitable selected locations in the structure. Issues related to the performance of the network with reference to hidden layers and hidden neurons are examined. Some heuristics are proposed for the design of neural networks for damage identification in structures. These are further supported by an investigation conducted on five other bridge truss configurations.
  • No Thumbnail Available
    Item
    Progressive failure of symmetrically laminated plates under uni-axial compression, Structural Engineering and Mechanics
    (Korea Institute of Science and Technology Information, 1997) Singh, Shamsher Bahadur
    The objective of this work is to predict the failure loads, associated maximum transverse displacements, locations and the modes of failure, including the onset of delamination, of thin, flat, square symmetric laminates under the action of uni-axial compression. Two progressive failure analyses, one using Hashin criterion and the other using Tensor polynomial criteria, are used in conjunction with the finite element method. First order shear deformation theory and geometric nonlinearity in the von Karman sense have been employed. Five different types of lay-up sequence are considered for laminates with all edges simply supported. In addition, two boundary conditions, one with all edges fixed and other with mixed boundary conditions for (+45/−45/0/90)2s quasi-isotropic laminate have also been considered to study the effect of boundary restraints on the failure loads and the corresponding modes of failure. A comparison of linear and nonlinear results is also made for (±45/0/90)2s quasi-isotropic laminate. It is observed that the maximum difference between the failure loads predicted by various criteria depend strongly on the laminate lay-ups and the flexural boundary restraints. Laminates with clamped edges are found to be more susceptible to failure due to the transverse shear and delamination, while those with the simply supported edges undergo total collapse at a load slightly higher than the fiber failure load.
  • No Thumbnail Available
    Item
    Progressive failure of symmetric laminates under in-plane shear: Il-Negative shear
    (Korea Institute of Science and Technology Information, 1998) Singh, Shamsher Bahadur
    The objective of the present work is to estimate the strength and failure characteristics of symmetric thin square laminates under negative shear load. Two progressive failure analyses, one using the Hashin criterion and the other using a Tensor polynomial criterion, are used in conjunction with the finite element method. First-order shear-deformation theory along with geometric nonlinearity in the von Karman sense has been incorporated in the finite element modeling. Failure loads, associated maximum transverse displacements, locations and modes of failure including the onset of delamination are discussed in detail; these are found to be quite different from those for the positive sheer load reported in Part I of this study (Singh et al. 1998).
  • No Thumbnail Available
    Item
    Progressive failure of symmetric laminates under in-plane shear, Part I-Positive Shear, Structural Engineering and Mechanics
    (Korea Institute of Science and Technology Information, 1998) Singh, Shamsher Bahadur
    The objective of this present work is to estimate the failure loads, associated maximum transverse displacements, locations and the modes of failure, including the onset of delamination, of thin, square symmetric laminates under the action in-plane positive (+ve) shear load. Two progressive failure analyses, one using the Hashin criterion and the other using a Tensor polynomial criterion, are used in conjunction with finite element method. First order shear deformation theory along with geometric non-linearity in the von Karman sense have been employed. Variation of failure loads and failure characteristics with five type of lay-ups and three types of boundary conditions has been investigated in detail. It is observed that the maximum difference between failure loads predieted by various criteria depends strongly on the laminate lay-up and the flexural boundary restraint. Laminates with clamped edges are found to be more susceptible to failure due to transverse shear (ensuing from the out of plane bending) and delamination, while those with simply supported edges undergo total collapse at a load slightly higher than the fiber failure load. The investigation on negative (-ve) in-plane shear load is in progress and will be communicated as part-II of the present work.
  • No Thumbnail Available
    Item
    Ensemble modeling or selecting the best model: Many could be better than one
    (1999) Barai, Sudhir Kumar
    In the course of data modeling, many models could be created. Much work has been done on formulating guidelines for model selection. However, by and large, these guidelines are conservative or too speci c. Instead of using general guidelines, models could be selected for a particular task based on statistical tests. When selecting one model, others are discarded. Instead of losing potential sources of information, models could be combined to yield better performance. We review the basics of model selection and combination and discuss their di erences. Two examples of opportunistic and principled combinations are presented. The rst demonstrates that mediocre quality models could be combined to yield signi cantly better performance. The latter is the main contribution of the paper; it describes and illustrates a novel heuristic approach called the SG (k-NN) ensemble for the generation of good quality and diverse models that can even improve excellent quality models.
  • No Thumbnail Available
    Item
  • No Thumbnail Available
    Item
    Postbuckling response and strength of laminates under combined in-plane loads
    (Elsiever, 1999-04) Singh, Shamsher Bahadur
    The objective of this work is to study the postbuckling behaviour and progressive failure response of thin, symmetric laminates under uniaxial compression and uniaxial compression combined with in-plane shear loads (positive and negative). First-order shear deformation theory and geometric non-linearity in the von Karman sense are used with a finite-element procedure. The 3-D Tsai-Hill criterion is used to predict failure of a lamina and the maximum stress criterion is used to predict onset of delamination at the interface of two adjacent layers. The effect of plate aspect ratio and ply lay-ups on the load deflection response is presented. Load interaction diagrams for (±45/0/90)2s, (±45)4s and (0/90)4s laminates are obtained in terms of the buckling, the first-ply and the ultimate failure loads. In addition, progressive failure response of the (±45/0/90)2s laminate is also presented to show the buckling loads, failure loads, maximum transverse displacements associated with the failure loads and failure modes and locations at various load ratios.
  • No Thumbnail Available
    Item
    Ensemble modelling or selecting the best model: Many could be better than one
    (CUP, 1999-11-01) Barai, Sudhir Kumar
    In the course of data modelling, many models could be created. Much work has been done on formulating guidelines for model selection. However, by and large, these guidelines are conservative or too specific. Instead of using general guidelines, models could be selected for a particular task based on statistical tests. When selecting one model, others are discarded. Instead of losing potential sources of information, models could be combined to yield better performance. We review the basics of model selection and combination and discuss their differences. Two examples of opportunistic and principled combinations are presented. The first demonstrates that mediocre quality models could be combined to yield significantly better performance. The latter is the main contribution of the paper; it describes and illustrates a novel heuristic approach called the SG(k-NN) ensemble for the generation of good-quality and diverse models that can even improve excellent quality models.
  • No Thumbnail Available
    Item
    Soil Bearing Capacity Determination using Ultrasonics
    (Journal of Pure and Applied Ultrasonics, 2000) Kumar, Kamalesh
  • No Thumbnail Available
    Item
    Extended Use of Linear Graph Theory for Analysis of Pipe Networks
    (ASCE, 2000-01) Gupta, Rajiv
    Linear graph theory used for pipe network analysis is to make the method systematic. A numerical method that uses linear graph theory is presented for the steady-state analysis of flow and pressure in a pipe network including its hydraulic components (pumps, valves, junctions, etc.). The proposed method differs from other linear graph methods in terms of the linear graph and the selection of its tree. The solution algorithm uses a function that depends on a power law to update the pipe flows in successive iterations. The exponents of this function are chosen to obtain a fast convergence rate even for large errors in the assumption of initial pipe flows. The convergence rate of the proposed method is validated using an error function and is compared to those of other methods. Some typical networks are analyzed to check the reliability of the proposed method. The results demonstrate the superior conditioning of the proposed method.
  • No Thumbnail Available
    Item
    Integration of damage assessment paradigms of steel bridges on a blackboard architecture
    (Elsiever, 2000-10) Barai, Sudhir Kumar
    This paper presents damage assessment paradigms integrated on blackboard platforms with special reference to steel bridge structures. The implementation of knowledge sources developed using paradigms of domain-specific algorithms, Artificial Intelligence (AI) techniques and Artificial Neural Networks (ANN) models are discussed. The components of the blackboard and the control aspects are explained in particular. The dynamics of the blackboard is discussed with simulation steps and identification of global damageability of bridge structure is demonstrated
  • No Thumbnail Available
    Item
    Pore fluid effect on shear strength of Pilani soil
    (Journal of the Institution of Engineers, 2001) Kumar, Kamalesh
    Soil strength and its accurate determination are always important issues to a soil engineer. Pore fluid and its characteristics are found to have significant effect in modifying shear strength parameters. Present study focuses on effect of pore fluid quantity and its electrolyte concentration on shear strength parameters of soil samples collected locally. Soil samples of specific particle size were subjected to direct shear testing when pore water content and its electrolyte concentration was changed. Variation of cohesion and angle of internal friction with above parameters was thus studied. Cohesion and angle of internal friction can be determined conveniently and accurately and provide valuable information about shear strength status of soil samples when pore water content and electrolyte concentration of pore water of the soil system changes. Thus optimum level of pore water quantity and electrolyte concentration in it has been determined to obtain optimum shear strength of soil mix tested. 
  • No Thumbnail Available
    Item
  • No Thumbnail Available
    Item
  • No Thumbnail Available
    Item
    Neuro-ensemble for air quality prediction
    (Loughborough University, 2002) Barai, Sudhir Kumar
    The present study investig; ttes the advantage of ensemble of neural networks (Haykin, 2000) for forecasting the air pollution. The aim is to find accurate air quality predictors, which can work with low number of data sets and should be robust enough to handle data with noise and errors