Civil Engineering Infrastructures Journal
Volume:45 Issue: 1, 2011

The purpose of the present study is to evaluate mechanical {stress-strain and volumetric} behavior of granular materials considering their internal physical properties in particle scale. To this end, two-dimensional numerical modeling of drained compressive tri-axial te¬st, using Discrete Element Method (DEM) was utilized. DEM-2D computer code was adopted for modeling these tests based on Discrete Element Method formulation. Drained biaxial numerical tests on specimens with different void ratios and confining pressures were carried out and the results are compared with the associated laboratory tests results reported in the literature.

Webs of rolled and built-up beam and plate girders could be subjected to local in-plane compressive patch loads. Examples are wheel loads, loads from purlins and roller loads during construction. The behavior of plate girders under patch loading, represents complex stability and elasto-plastic problems. It is influenced by many different factors and even the increasing number of experimental results and laborious theoretical work could not give complete insight into the problem. Some empirical and semi-emprical formulas have been established, but significant errors still present in the current design formulas. As an alternative solution, numerical methods, such as finite element have been used to model the problem but they still present significant difference when compared to experimental results. This study considers the use of artificial neural networks (NN) to predict the ultimate resistance of plate girders subjected to patch loading. In this work, multilayer perceptrons with backpropagation are used. This choice is mainly due to their adaptive structure and to the efficient learning algorithms nowadays available. The training and testing patterns of the proposed neural network system are based on well established experimental results taken from literature. Among these 200 tests, 50 tests are used as the test data set and remaining ones used as the training set for NN training. Each training data sample is composed of the eight geometrical and material parameters and by the experimental ultimate load. The performance of an NN model mainly depends on the network architecture and parameter settings. One of the most difficult tasks in NN studies is to find this optimal network architecture, which is based on the determination of numbers of optimal layers and neurons in the hidden layers by a trial and error approach. In this study the MATLAB toolbox is used for NN applications. To produce reasonable results, the available experimental data divided in two classes according to their ultimate load capacity. Backpropagation algorithm was used to train the first and second class neural networks. It is observed that all networks presented error values below 11% and no higher than 4.35% for majority of them. The predictions of the proposed NN model are compared with experimental values. It is obvious that the proposed NN model learned well to map the relationship between the ultimate path resistance and its geometric and mechanical properties. One of the advantages of the proposed NN model is its wide range of input parameters, which enables the NN model to be used in practical applications. The division of neural training data into two load classes provided better training and better suitability to distinct characteristics of the problem. The well trained NN model could be also used to conduct parametric studies. The results of the proposed NN are compared with current design formulae and are found to be considerably more accurate. The proposed NN presents a maximum error lower than 11% while existing design formulas errors are over 20%.

In this study, a new unconditionally stable time integration is proposed for solving the differential equation of motion. Due to increase in order of variations of acceleration, proposed method has higher accuracy than classical methods. Two variable parameters are used to increase the stability and accuracy of the method. The proposed method is second order accurate and also it includes methods with numerical dissipation which can be used to filter the high undesirable frequencies. Moreover in the proposed method, equation of motion is exactly satisfied at the beginning and at the end of the time step.

Spar platforms are large structures which are used in deep waters. Because of their large dimension, Morrison’s Equation is not suitable to estimate hydrodynamic forces on this kind of platforms. Therefore in this study wave forces and hydrodynamic coefficients of a spar platform have been estimated in six degree of freedom using diffraction theory. The incident wave is assumed to be harmonic and linear. Diffraction and radiation velocity potential were solved with finite difference method. The results showed that the developed model is able to estimate the wave forces and hydrodynamic coefficients on large structures considering diffraction theory with acceptable accuracy.

In this study, the effect of state parameter in elasto-plastic constitutive models on improvement of predicting sand behavior was studied, using DIANA-SWANDYNE II program. The centrifuge test results, which were obtained in VELACS workshop, were used in this study. The mechanical behaviors of centrifuge models under earthquake loading was predicted by using Pastor & Zienkiewicz Mark III constitutive model and the modified model and were compared with experimental results. The results show that the use of current state parameter can have a significant effect on improvement of predicting deformation, water pore pressure and acceleration caused by earthquake loading in soil structures.

With development of new seismic design concepts like performance based design, rehabilitation of existing reinforced concrete structures is of great importance nowadays, because they do not satisfy recent design criteria considering optimum performance level with providing both ductility and strength capacities in earthquakes with different hazard levels. Since adding Reinforced Concrete (RC) infills to RC frames is an effective rehabilitation procedure and openings in RC infills are inevitable, behavior study of RC frames strengthened by RC infills with openings seems necessary. In this paper, nonlinear behavior study of nine five-story concrete frames strengthened by RC infills with openings was done by performing nonlinear static analysis (Push Over), for which appropriate nonlinear modeling of system in PERFORM-3D software was verified by comparing results of a single story RC partially infilled frames under cyclic loading with experimental results of a similar existing study.Infill span, position and lateral load pattern affects strength, stiffness and ductility of a typical existing RC frame as behavior parameters. Adding RC infills obviously increases strength and stiffness of a RC frame, although it results in ductility decrease which can be prevented by proper positioning of infills. As a result, infill span is the most effective parameter on system strength and triangular lateral load pattern induce more ductility and strength capacities in system. Therefore; adding RC infills in proper positions is a way to enhance strength and stiffness of RC frames significantly with provision of appropriate ductility. As an example, a frame which is strengthened by infills with opening at the midspan shows better strength, stiffness and even ductility compared to those with openings near the columns.

Concrete buildings that have been built before 1960 are mainly reinforced by plain bars. Because of defective bond between plain bars and concrete, the response of such buildings to earthquakes is uncertain. Despite several experimental reports of cyclic behavior of such beams, their mechanical and physical modeling is vague, and little report on this issue can be found in the well-known literature. It is known that the displacement of concrete beams reinforced by plain bars consists of three components: slip, bending, and shear. In this paper, an attempt has been made to develop a model for flexural behavior of beams with plastic hinges, and to expand a formulation for slip of plain bars in concrete. The theoretical model has been verified by tests of three beams. The results show that the effective stiffness of plain-bar- beams is equal to 22% of that of gross section. In addition, displacement ductility of the beams is about 4.5. These properties vary with ratio of longitudinal reinforcement, ratio of sectional dimensions, and arrangement of reinforcements, where all these parameters have been assessed in the study.

The hydraulic jump on negative steps forms in stilling basins. In this paper the hydraulic and hydrodynamic behavior of these flows were studied. The experiments were performed in a channel with length:6m depth:0.5m and width: 0.47m. The step heights were 2/5, 4 and 7cm. The varied parameters in hydraulic study were 1. Froud number 2. The ratio of secondary depth to initial depth 3. The ratio of the step height to initial depth. By changing the initial velocity of hydraulic jump, step height and tailwater depth different kinds of hydraulic jumps were formed. For estimating the hydraulic jump shapes different curves were derived. Study about the pressure correction coefficient were done. In hydrodynamic study PIV apparatus was used. This apparatus can measure the velocities in confined surface. In hydrodynamic study the mean velocities and turbulent parameters were investigated. Because of limitations of PIV, the hydrodynamic studies were done only for two different kind of wavy hydraulic jumps and an A jump.

In this article we use wavelet transform to solve scalar elastic wave equation with real boundary condition in earthquake engineering problems (i.e., soil-structure interaction). Multiresolution property, localization and inherent adaptive features of wavelet transform make this approach appropriate for solving PDEs; moreover, fast algorithms are developed. Working in the framework of wavelet, here projection schemes are used, i.e., operators (such as derivatives dmf/dxm) are directly projected in the wavelet spaces. There, the projected operators are sparse and banded matrixes; moreover they are represented in different levels of resolution. The width of the bands could be reduced by thresholding the coefficients of the matrixes; i.e., at first a predefined threshold is assumed, then the coefficients, whose values are smaller than the threshold value, are replaced by zero; thereby, an inherent adaption is attained, increasing the computational speed. The wavelet families used here is Daubechies wavelet, an orthogonal family with compact support property. This kind of wavelet leads to the most regular approximation with the smallest support of wavelets. Temporal time integration is done by explicit semi-group method. In this method, by using explicit approach, the obtained results are comparable with common implicit schemes. In this work propagation of SH waves in a infinite region, infinite media containing a hole (such as tunnels), and interaction of a shear wall and a semi-infinite domain are investigated. To simulations the free surfaces, here, the material with nearly zero mechanical properties are used, i.e., an artificial domain (like air) is modeled. Hence, it is not necessary to simulate the free boundaries. Simulation of such boundaries is a challenge problem in wavelet-based projection schemes; in common wavelet theories (first generation wavelet families) it is needed to modify the wavelets in the vicinity of the boundaries to simulate Dirichlet and Neumann boundary conditions. The infinite boundaries are simulated via absorbing boundaries. They are introduced explicitly by modification the wave equation, adding artificial viscosity in the computational domain. The viscosity function is zero in domain and increases slightly while approaches to the infinite boundaries. The results indicate that for a grid with constant points, the proposed method is more stable than the common finite difference scheme. Moreover, the results show that interaction of structure and soil is an important factor, for example, for shear walls with higher density, smaller responses are captured during propagation of waves in the earth. Also when length of propagating waves is comparable with structure size, high localized responses in the structure will be occurred; therefore common beam element formulation could not be used.

Spatial data collection is a time/money consuming task in activities related to geospatial sciences. Spatial data integration increases the efficiency of spatial data through cost reduction. Therefore, developing matching algorithms is of vital importance in spatial data integration, data updating and data accuracy enhancement. The main objective of this research is to design and develop a semi-automatic vector matching system. The developed system simulates human brain process in its matching procedure. Priliminary results are promissing. Further reaseaches are required to examine the system efficiency.