• Journal of the Society of Naval Architects of Korea
• /
• v.33 no.2
• /
• pp.96-110
• /
• 1996

Aging ships can suffer structural damage due to corrosion, fatigue crack etc., and possibility of catastrophic failure of seriously damaged ships is very high. To reduce the risk of loss of ships due to hull collapse, it is essential to evaluate ultimate hull strength of aging ships taking into account various uncertainties associated with structural damages. In this paper, ultimate strength-based reliability analysis of ship structures considering wear of structural members due to corrosion is described. A corrosion rate estimate model for structural members is introduced. An ultimate limit state function of a ship hull is formulated taking into account corrosion effects. The model is applied to an existing oil tanker, and reliability index associated with hull collapse is calculated by using the second-order reliability method (SORM). Discussions on structure safety of corroded ships are made.

• Composites Research
• /
• v.23 no.2
• /
• pp.40-47
• /
• 2010

In this paper, the hybrid response surface method(HRSM) is proposed and examined. Hybrid response surface method calculate a approximate model repeatedly based on MPP coordinates. To verify the performance, probability of failure, MPP(Most Probable failure Point) and reliability index are calculated for nonlinear function and composite laminated plate by using reliability analysis method and compared with results by using typical response surface method(RSM). Probability of failure is calculated under the assumption of the nonlinear limit state equation and given failure criterion. The results of proposed method shows performance improvement in estimating the probability of failure.

• Journal of Ocean Engineering and Technology
• /
• v.29 no.5
• /
• pp.342-350
• /
• 2015

A seismic reliability analysis of the jacket-type support structure for an offshore wind turbine was performed. When defining the limit state function using the dynamic response of the support structure, numerous dynamic calculations should be performedin an approach like the FORM (first-order reliability method). This causes a substantial increase in the analysis cost. Therefore, in this paper, a new reliability analysis approach using the static response is used. The dynamic effect of the response is considered by introducing a new parameter called the peak response factor (PRF). The probability distribution of the PRF could be estimated using the peak value of the dynamic response. The probability distribution of the PRF was obtained for a set of ground motions. A numerical example is considered to compare the proposed approach with the conventional static-response-based approach.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.12 no.2
• /
• pp.217-227
• /
• 1992

The first and second-order reliability method(FORM and SORM) is presented using one dimensional finite difference and two dimensional finite element transport models. FORM and SORM can be used without any restrictive assumptions about the properties of the media, and the sensitivity information obtained as part of these analyses is used to identify the parameters which have major influence on the estimate of probability. The reliability analysis of transport in a one-dimensional domain is used to test the robustness of the reliability code and to evaluate the accuracy of the reliability method. A continuous source 2-D example with a concentration threshold limit state function is used to evaluate the influence of the parameters in the location of interest on the reliability solution.

• Journal of Korean Library and Information Science Society
• /
• v.35 no.1
• /
• pp.303-317
• /
• 2004

The purpose of this study was to develop a framework for predicting searching performance through an understanding of how cognitive ability relates to searching process and outcome. Specifically, this study examined the relationship between spatial visualization, logical reasoning, integrative reasoning, and information searching process and outcome. Information searching process was assessed by seven search process indicators: (1) search command selection: (2) combination of search commands; (3) application of Boolean logic: (4) application of truncation; (5) use of limit search function; (6) number of search statements; and (7) number of search errors made. Searching outcome was assessed by the number of correct answers to search questions. Subjects first took three standardized cognitive tests that measured cognitive abilities, and performed online catalog searching in response to seven information search questions. The searches were logged using Lotus ScreenCam, and reviewed for the analysis. Factor analysis was used to find underlying structures of the seven search process variables. Multiple regression analysis was applied to examine the predictive power of three cognitive variables on three extracted factors, and search outcome. Results of the data analysis showed that individual differences in logical reasoning could predict information searching process and outcome.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.3
• /
• pp.224-229
• /
• 2009

Partial safety factors(PSFs) for Level I reliability based design of caisson type quay walls were calculated. First order reliability method(FORM) based PSFs are the functions of sensitivities of limit state function with respect to design random variables, target reliability index, characteristic values and first moment of random variables. Modified PSFs for water level and resilient water level are newly defined to keep consistency with the current design code. In the numerical example, PSFs were calculated by using a target reliability index. Seismic coefficient is defined to show extreme distribution. It was found that PSFs for seismic coefficient becomes smaller as the return period for design seismic coefficient grows longer.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.55 no.3
• /
• pp.75-84
• /
• 2013

The seepage quantity analysis of reservoir embankment is very important for assessment of embankment safety. However, the conventional analysis does not consider uncertainty of soil properties. Permeability is known that the coefficient of variation is larger than other soil properties and seepage quantity is highly dependent on the permeability of embankment. Therefore, probabilistic analysis should be carried out for seepage analysis. To designers, however, the probabilistic analysis is not an easy task. In this paper, the method that can be performed probabilistic analysis easily and efficiently through the numerical analysis based commercial program is proposed. Stochastic response surface method is used for approximate the limit state function and when estimating the coefficients, the moving least squares method is applied in order to reduce local error. The probabilistic analysis is performed by LHC-MCS through the response surface. This method was applied to two type (homogeneous, core zone) earth dams and permeability of embankment body and core are considered as random variables. As a result, seepage quantity was predicted effectively by response surface and probabilistic analysis could be successfully implemented.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.18 no.3
• /
• pp.69-75
• /
• 2014

The priority of repair areas are chosen with the probability distribution of 0.3mm wide crack and carbonation induced corrosion. Data is analyzed and evaluated based on the 28 section of Precise Inspection for Safety and Diagnosis (PISD) in seoul. As the crack is distributed in log-normal, the carbonation and cover are in normal distribution. To have rational in repair sections among 503 sheets of underground structure, it is adopted the reliability index as well as the environment factors: strength, sonic speed, $CO_2$ concentration, corrosion, and content of chloride.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.5
• /
• pp.397-404
• /
• 2012

Fragility curves show the probability of a system reaching a limit state as a function of some measure of seismic intensity. To obtain fragility curves of six and twelve story staggered wall structures with middle corridor, incremental dynamic analyses were carried out using twenty two pairs of earthquake records, and their failure probabilities for various intensity of seismic load were investigated. The performances of staggered wall structures with added columns along the central corridor and the structures with their first story walls replaced by columns were compared with those of the regular staggered wall structures. Based on the analysis results it was concluded that staggered wall structures with central columns have the largest safety margin for the same level of seismic load.

• Journal of the Korean institute of surface engineering
• /
• v.37 no.4
• /
• pp.226-233
• /
• 2004

The residual stresses on the surfaces of low carbon 12Cr steels used as a nuclear steam turbine blade material have been studied by controlling the flame hardening surface treatments. The temperature cycles on the surfaces of 12Cr steel were controlled precisely as a function of both the surface temperature and cooling rate. The final residual stress state generated by flame hardening was dominated by two opposite competitive contributions; one is tensile stress due to phase transformation and the other is compressive stress due to thermal contraction on cooling. The optimum processing temperatures required for the desirable residual stress and hardness were in the range of $850^{\circ}C$ to $960^{\circ}C$ on the basis of the specification of GE power engineering. It was also observed that the high residual tensile stress generated by flame hardening induced the cracks on the surfaces, especially across the prior austenite grain boundaries, and the material failure virtually, which might limit practical use of the surface engineered parts by flame hardening.