• Smart Structures and Systems
• /
• v.16 no.2
• /
• pp.367-382
• /
• 2015

This paper aims to conduct the reliability-based assessment of the welded joint in the orthotropic steel bridge deck by use of a mesh-insensitive structural stress (MISS) method, which is an effective numerical procedure to determine the reliable stress distribution adjacent to the weld toe. Both the solid element model and the shell element model are first established to investigate the sensitivity of the element size and the element type in calculating the structural stress under different loading scenarios. In order to achieve realistic condition assessment of the welded joint, the probabilistic approach based on the structural reliability theory is adopted to derive the reliability index and the failure probability by taking into account the uncertainties inherent in the material properties and load conditions. The limit state function is formulated in terms of the structural resistance of the material and the load effect which is described by the structural stress obtained by the MISS method. The reliability index is computed by use of the first-order reliability method (FORM), and compared with a target reliability index to facilitate the safety assessment. The results achieved from this study reveal that the calculation of the structural stress using the MISS method is insensitive to the element size and the element type, and the obtained structural stress results serve as a reliable basis for structural reliability analysis.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.83-84
• /
• 2008

The primary water stress corrosion cracking (PWSCC) of dissimilar metal weld based on Alloy 82/182 is one of major issues in material degradation of nuclear components. It is well known that the crack initiation and growth due to PWSCC is influenced by material's susceptibility to PWSCC and distribution of welding residual stress. Therefore, modeling the welding residual stress is of interest in understanding crack formation and growth in dissimilar metal weld. Currently in Korea, a numerical round robin study is undertaken to provide guidance on the welding residual stress analysis of dissimilar metal weld. As a part of this effort, the present paper investigates distribution of welding resisual stress of a ferritic low alloy steel nozzle with dissimilar metal weld using Alloy 82/182. Two-dimensional thermo-mechanical finite element analyses are carried out to simulate multi-pass welding process on the basis of the detailed design and fabrication data. The present results are compared with those from other participants, and more works incorporating physical measurements are going to be performed to quantify the uncertainties relating to modelling assumptions.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.31 no.7
• /
• pp.450-454
• /
• 2018

Because of the rapidly changing environment and high uncertainties, the semiconductor industry is in need of appropriate forecasting technology. In particular, both the cost and time in the test process are increasing because the process becomes complicated and there are more factors to consider. In this paper, we propose a prediction model that predicts a final "good" or "bad" on the basis of preconditioning test data generated in the semiconductor test process. The proposed prediction model solves the classification and regression problems that are often dealt with in the semiconductor process and constructs a reliable prediction model. We also implemented a prediction model through various machine learning algorithms. We compared the performance of the prediction models constructed through each algorithm. Actual data of the semiconductor test process was used for accurate prediction model construction and effective test verification.

• Composites Research
• /
• v.20 no.4
• /
• pp.25-30
• /
• 2007

The advanced fiber-reinforced laminated composites are widely used in a variety of engineering applications such as aerospace, marine, mechanical and civil engineering for weight savings because of their high specific strength and stiffness. The material properties of ply is known to have larger variations than that of conventional materials and very sensitive to the loading direction. Therefore, it is important to consider the variations on designing the laminated composite. This paper demonstrates the importance of considering uncertainties through examining the effect of material properties variations on various design requirements such as tip deflection, natural frequency and buckling stress using COMSOL-MATLAB interface.

• Geomechanics and Engineering
• /
• v.33 no.5
• /
• pp.453-462
• /
• 2023

Intermediate Geomaterials (IGMs) are natural formation materials that exhibit the engineering behavior (strength and compressibility) between soils and rocks. The engineering behavior of such material is highly unpredictable as the IGMs are stiffer than soils and weaker/softer than rocks. Further, the characterization of such material needs exposure to both soil and rock mechanics. In most conventional designs of geotechnical structures, the engineering properties of the IGMs are either aligned with soils or rocks, and this assumption may end up either in an over-conservative design or under-conservative design. Hence, many researchers have attempted to evaluate its actual engineering properties through laboratory tests. However, the test results are partially reliable due to the poor core recovery of IGMs and the possible sample disturbance. Subsequently, in-situ tests have been used in recent years to evaluate the engineering properties of IGMs. However, the respective in-situ test finds its limitations while exploring IGMs with different geological formations at deeper depths with the constraints of sampling. Standard Penetration Test (SPT) is the strength-based index test that is often used to explore IGMs. Moreover, it was also observed that the coefficient of variation of the design parameters (which represents the uncertainties in the design parameters) of IGMs is relatively high, and also the studies on the probabilistic characterization of IGMs are limited compared with soils and rocks. With this perspective, the present article reviews the laboratory and in-situ tests used to characterize the IGMs and explores the shear strength variation based on their geological origin.

• Nuclear Engineering and Technology
• /
• v.56 no.3
• /
• pp.873-879
• /
• 2024

A sodium-cooled fast reactor (SFR) core has a potential of prompt criticality due to a change of core material distribution during a severe accident, and the resultant energy release has been one of the safety issues of SFRs. In this study, the safety assessment of an unprotected loss-of-flow (ULOF) in a small SFR (SSFR) has been performed using the SIMMER-IV computer code, which couples the models of space- and time-dependent neutronics and multi-component, multi-field thermal hydraulics in three dimensions. The code, therefore, is applicable to the simulations of transient behaviors of extended disrupted core material motion and its reactivity effects during the transition phase (TP) of ULOF, including a potential of prompt-criticality power excursions driven by fuel compaction. Several conservative assumptions are used in the TP analysis by SIMMER-IV. It was found out that one of the important mechanisms that drives the reactivity-inserting fuel motion was sodium vapor pressure resulted from a fuel-coolant interaction (FCI), which itself was non-energetic local phenomenon. The uncertainties relating to FCI is also evaluated in much conservative way in the sensitivity analysis. From this study, the ULOF characteristics in an SSFR have been understood. Occurrence of recriticality events under conservative assumptions are plausible, but their energy releases are limited.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.1
• /
• pp.97-106
• /
• 2011

In this study, viscoplastic material properties of solder alloy which is used in the electronics packages are inversely estimated. A specimen is fabricated to this end, and an experiment is conducted to examine deformation by Moir$\acute{e}$ interferometry. As a result of the experiment, bending displacement of the specimen and shear strain of the solder are obtained. A viscoplastic finite element analysis procedure is established, and the material parameters are determined to match closely with the experiments. The uncertainties which include inherent experimental error and insufficient data of experiments are addressed by using the method of computer model calibration. As a result, material parameters are identified in the form of confidence interval, and the displacements and strains using these parameters are predicted in the form the prediction interval.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.5 no.2
• /
• pp.51-57
• /
• 1985

In the design of civil engineering structures, the designers are invariably faced to the uncertainties and the randomness of the design parameters such as material properties and loads. Even when the structures are built, the actual geometries of the structures are also subject to their random variations from their nominal design values. Thus, the reliability of a structure in terms of these uncertainties and variations becomes a matter of great concern to the structural designers. This study employs the First Order Second Moment Method to evluate numerically the reliability of a simple tension member and discusses the influence on the final failure probability of that structure due to: 1) use of equivalent normal distribution in place of non-normal distribution, 2) linearization of non linear limit state equation. A discussion is also made on the necessity of fundamental studies on the distrubution characteristics of the strength of locally produced construction materials and those of the loads frequently encountered in the structural design.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.4
• /
• pp.336-345
• /
• 2012

In crack growth life, uncertainties are caused by variance of geometry, applied loads and material properties. Therefore, the reliability estimation for these uncertainties is required to keep the robustness of calculated life. The stress intensity factors are the most important variable in crack growth life calculation, but its equation is hard to know for complex geometry, therefore they are processed by the finite element analysis which takes long time. In this paper, the response surface is considered to increase efficiency of the reliability analysis for crack growth life of a turbine wheel. The approximation model of the stress intensity factors is obtained by the regression analysis for FEA data and the response surface of crack growth life is generated for selected factors. The reliability analysis is operated by the Monte Carlo Simulation for the response surface. The results indicate that the response surface could reduce computations that need for reliability analysis for the turbine wheel, which is hard to derive stress intensity factor equation, successfully.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2008.04a
• /
• pp.293-296
• /
• 2008

Creep and shrinkage in concrete structures are very complex phenomena in which various uncertainties exist with regard to inherent material variations as well as modeling uncertainties. The creep and shrinkage models which are capable of predicting long-term structural response are specified in design codes such as ACI 209-92, CEB-FIP Model Code 90, etc. However, in the prediction formulas of creep and shrinkage effects of concrete, various kinds of parameters are involved to express the characteristics of concrete under consideration (i.e. the proportion of concrete, the shape of the structure, relative humidity, etc.). And the predicted values from each design code under same environment differ from each other. To predict the characteristics of concrete, the long-term experiments of creep and shrinkage is necessary but this is not suitable for a construction field. In this study, adjustment method of creep coefficient using sensitivity analysis is proposed to predict creep coefficient of concrete exactly and it is checked up on the validity of the predicting method by comparing to the assumed value and predicted one.