• Structural Engineering and Mechanics
• /
• v.72 no.6
• /
• pp.763-774
• /
• 2019

The main aim of this paper is enhancing design of traditional laminated composite plates subjected to static loads. In this regard, this paper suggests embedding a lightweight porous layer in the middle of laminated composite as the core layer of the resulted sandwich plate. The static responses of the suggested structures with uniform, symmetric and non-symmetric porosity distributions are compared to optimize their design. Using the first order shear deformation theories, the static governing equations of the suggested laminated composite plates with a porous layer (LCPPL) rested on two-parameter foundation are obtained. A finite element method is also utilized to solve the governing equations of LCPPLs. Effects of laminated composite and porosity characteristics as well as geometry dimension, edges' boundary conditions and foundation coefficients on the static deflection and stress distribution of the suggested composite plates have been investigated. The results reveal that the use of core between the layers of laminated composites leads to a sharp reduction in the static deflections of LCPPLs. Furthermore, in compare with perfect cores, the use of porous core between the layers of laminated composite plates can offer a considerable reduction in structural weight without a significant difference in their static responses.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.9 no.4
• /
• pp.452-462
• /
• 1985

The axisymmetric forward extrusion is analyzed by using the rigid-plastic finite element formulation. The distribution of stresses and strains as well as the deformation pattern in solid extrusion is very important for the improvement of product quality. The initial velocity field is determined by assuming the material as a Newtonian fluid through an arbitrarily shaped axisymmetric die. The workhardening effect and the friction of the die-material interface are considered in the formulation. Some reduction of area and die shapes(conical and biquadratic-curved) are chosen for computation. Experiments are carried out for steel alloy(SCM4) specimens using conical and curved dies. It is found that experimental observation is in good agreement with FEM results. The strain distribution is curved(biquadratic) dies is shown to be more uniform than in conical dies at the same reduction of area.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.3
• /
• pp.646-655
• /
• 1990

A practically useful system for elastic-plastic fracture mechanics analysis has been developed. The developed system is comprised of the deformation plasticity failure assessment diagram(DPFAD) approach and the J-integral/Tearing modulus(J/T) approach. The system contains analysis routines for five types of fracture specimens : compact tension, center cracked tension, single edge craked plate in uniform tension, single edge cracked plate in three point bending and double edge cracked plate in tension. A double interpolation scheme was adopted to interpolate J values from the EPRI developed EPFM handbook and the Newton-Raphson method was used to obtain proper loadings for displacement control conditions. A graphic output system was utilized to present numerical results. Several case studies were performed to evaluate the accuracy and the usefulness of the code. It was found that the J/T approach and the DPFAD approach yielded similar results. However, the DPFAD approach is more convenient for qick assessment of integrity of cracked structures while the J/T approach is more useful in evaluating the full history of the fracture process.

• Smart Structures and Systems
• /
• v.25 no.6
• /
• pp.693-705
• /
• 2020

In the current investigation, large amplitude free vibration behavior of shallow curved pipes (tubes) made of functionally graded materials is investigated. Properties of the tube are distributed across the radius of the tube and are obtained by means of a power law function. It is also assumed that all thermo-mechanical properties are temperature dependent. The governing equations of the tube are obtained using a higher order shear deformation tube theory, where the traction free boundary conditions are satisfied on the top and bottom surfaces of the tube. The von Kármán type of geometrical non-linearity is included into the formulation to consider the large displacements and small strains. Uniform temperature elevation of the tube is also included into the formulation. For the case of tubes which are simply supported in flexure and axially immovable, the governing equations are solved using the two-step perturbation technique. Closed form expressions are provided to obtain the small and large amplitude fundamental natural frequencies of the FGM shallow curved tubes in thermal environment. Numerical results are given to explore the effects of thermal environment, radius ratio, and length to thickness ratio of the tube on the fundamental linear and non-linear frequencies.

• Nuclear Engineering and Technology
• /
• v.53 no.2
• /
• pp.647-656
• /
• 2021

Knowing a material's true stress-strain curve is essential for performing a nonlinear finite element analysis to solve an elastoplastic problem. This study presents a simple methodology to determine the true stress-strain curve of type 304 and 316 austenitic stainless steels in the full range of strain from a typical tensile test. Before necking, the true stress and strain values are directly converted from engineering stress and strain data, respectively. After necking, a true stress-strain equation is determined by iteratively conducting finite element analysis using three pieces of information at the necking and the fracture points. The Hockett-Sherby equation is proposed as an optimal stress-strain model in a non-uniform deformation region. The application to the stainless steel under different temperatures and loading conditions verifies that the strain hardening behavior of the material is adequately described by the determined equation, and the estimated engineering stress-strain curves are in good agreement with those of experiments. The presented method is intrinsically simple to use and reduces iterations because it does not require much experimental effort and adopts the approach of determining the stress-strain equation instead of correcting the individual stress at each strain point.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.1740-1743
• /
• 2007

In this study, two types of fatigue tests were conducted. First, cyclic bending tests were performed using the micro-bending tester. A four-point bending test method was adopted, because it induces uniform stress fields within a loading span. Second, thermal fatigue tests were conducted using a pseudo power cycling machine which was newly developed for a realistic testing condition. The pseudo-power cycling method makes up for the weak points in a power cycling and a chamber cycling method. Two compositions of solder are tested in all test condition, one is lead-free solder (95.5Sn4.0Ag0.5Cu) and the other is eutectic lead-contained solder (63Sn37Pb). In the cyclic bending test, the solder that exhibits a good reliability can be reversed depending on the load conditions. The lead-contained solders have a longer fatigue life in the region where the applied load is high. On the contrary, the lead-free solder sustained more cyclic loads in the small load region. A similar trend was detected at the thermal cycling test. A three-dimensional finite element analysis model was constructed. A finite element analysis using ABAQUS was performed to extract the applied stress and strain in the solder joints. A constitutive model which includes both creep and plasticity was employed. Thermal fatigue was occurred due to the creep. And plastic deformation is main damage for bending failure. From the inelastic energy dissipation per cycle versus fatigue life curve, it can be found that the bending fatigue life is longer than the thermal fatigue life.

• Structural Engineering and Mechanics
• /
• v.59 no.3
• /
• pp.503-526
• /
• 2016

In-plane and out-of-plane free vibration analysis of Timoshenko curved beams is addressed based on the isogeometric method, and an effective scheme to avoid numerical locking in both of the two patterns is proposed in this paper. The isogeometric computational model takes into account the effects of shear deformation, rotary inertia and axis extensibility of curved beams, and is applicable for uniform circular beams, and more complicated variable curvature and cross-section beams as illustrated by numerical examples. Meanwhile, it is shown that, the $C^{p-1}$-continuous NURBS elements remarkably have higher accuracy than the finite elements with the same number of degrees of freedom. Nevertheless, for in-plane or out-of-plane vibration analysis of Timoshenko curved beams, the NURBS-based isogeometric method also exhibits locking effect to some extent. To eliminate numerical locking, the selective reduced one-point integration and $\bar{B}$ projection element based on stiffness ratio is devised to achieve locking free analysis for in-plane and out-of-plane models, respectively. The suggested integral schemes for moderately slender models obtain accurate results in both dominated and non-dominated regions of locking effect. Moreover, this strategy is effective for beam structures with different slenderness. Finally, the influence factors of structural parameters of curved beams on their natural frequency are scrutinized.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.10
• /
• pp.5974-5979
• /
• 2014

Lightweight parts with very uniform precision are manufactured by an indent method and the press technique has been improved. Upon assembly with an indent method, a deformation force due to the compressive force occurs between the pin and hole and the contact surface is affected by damage. Therefore, a 3 dimensional model was made using the CATIA program and the damage on the surface contacted with indent was estimated through the ANSYS program in this study. In the analysis result, the maximum load applied at the PCB plate was 21.3 N when the pin goes through the PCB plate. When PCB plate came out of the pin, the maximum load was 19.24 N. As the structural analysis result, the maximum equivalent stress of Pin 1 was 192.96MPa because the maximum stress occurs at Pin 1 among all parts of this study model. By examining the damage property of the contacted indent and applying this study result to the design of real indentation, the damage can be prevented and the durability can be estimated.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Computational Structural Engineering
• /
• v.4 no.3
• /
• pp.89-97
• /
• 1991

Conventional aseismic design methods of R/C frame all but disregard the state of damage over the entire building frame. This paper presents an automated damage-controlled design method for R/C frames which aims at an uniform energy dissipation rate throughout the building frame, so that the resulting damage is uniformly distributed as much as possible over all element. The accuracy of the basic hystertic model and the damage model for R/C members is verified by reproducing the experimental load-deformation curves of one-bay one-story frames. Application of this design method to various frame structures indicate that 1) regardless of the structural properties or input earthquake characteristics, damage-controlled frames generally survive more severe earthquake excitations and suffer less damage than conventionally designed frames, and 2) member yielding strength in the lower stories of damage-controlled frames is larger than that for conventionally designed frames, while the trend is opposite in the upper stories.