• Journal of Korean Society of Environmental Engineers
• /
• v.34 no.3
• /
• pp.168-176
• /
• 2012

A total of 48 dust samples were collected from large shopping mall parking lots in Daegu metropolitan city in March 2011. Samples were sieved through a 100 ${\mu}m$ mesh and the concentration of 14 elements have been determined using by ICP after acid extraction. Results showed that Ca, Fe, K, Mg, Mn, Na and V were affected by natural sources while Cd, Cr, Cu, Ni, Pb and Zn were affected by anthropogenic sources. The measured values were remarkably higher in components from natural sources than in components from anthropogenic sources. Anthropogenic trace element concentrations of ground roof dust were higher than those of ground and underground indoor dust. A large percentage of trace elements came from natural sources rather than anthropogenic sources. The percentage composition of chemicals of ground roof dust were higher than those of ground and underground indoor dust. This study showed that investigated parking lots were rarely contaminated with hazardous heavy metals. The heavy metal pollution of ground roof were higher than those of ground and underground indoors. The correlation analysis among trace elements suggest that components in ground roof were more highly correlated than those in ground and underground indoor. Also anthropogenic trace element levels were well correlated with parking lot age and parking density.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.2
• /
• pp.77-84
• /
• 2021

A sampling-based approach was devised as a nuclear seismic probabilistic risk assessment (SPRA) method to account for the partially correlated relationships between components. However, since this method is based on sampling, there is a limitation that a large number of samples must be extracted to estimate the results accurately. Thus, in this study, we suggest an effective approach to improve the existing sampling method. The main features of this approach are as follows. In place of the existing Monte Carlo sampling (MCS) approach, the Latin hypercube sampling (LHS) method that enables effective sampling in multiple dimensions is introduced to the SPRA method. In addition, the degree of segmentation of the seismic intensity is determined with respect to the final seismic risk result. By applying the suggested approach to an actual nuclear power plant as an example, the accuracy of the results were observed to be almost similar to those of the existing method, but the efficiency was increased by a factor of two in terms of the total number of samples extracted. In addition, it was confirmed that the LHS-based method improves the accuracy of the solution in a small sampling region.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.2
• /
• pp.83-91
• /
• 2022

Seismic design and analysis of nuclear power plant components are performed based on an decoupled model. However, this decoupled analysis has a limitation in that it generates inaccurate results compared to the coupled analysis because it cannot simulate actual phenomena such as the interaction between structures and components. Thus, this study performed seismic coupled and decoupled analysis on an existing nuclear containment structure and related components, considering the mass and natural frequency ratios. And based on these results, comparative analyses of responses of components were conducted. Consequently, the seismic coupled analysis result generally gave a smaller value than the decoupled analysis result. These results were similar to the analysis results for the simple coupled model, which was an existing study, but the difference in component responses was much more pronounced. Also, this was influenced by the installation location of the component rather than the influence of the input frequency of the input seismic motions. Finally, the difference between the decoupled and coupled seismic analysis occurred in the region where the mass ratio of the components was large, and the natural frequencies were almost similar due to the considerable dynamic interaction between the structure and the component in this realm.

• Journal of the Society of Disaster Information
• /
• v.18 no.2
• /
• pp.252-260
• /
• 2022

Purpose: As the intensity and frequency of natural hazards are increasing due to climate change, external events that affecting nuclear power plants(NPPs) may increase. NPPs must be protected from external events such as natural hazards and human-induced hazards. External events that may occur in NPPs should be identified, and external events that may affect NPPs should be identified. This study introduces the methodology of identification and screening methods for external events by literature review. Method: The literature survey was conducted on the identification and screening methods of external events for probabilistic safety assessment of NPPs. In addition, the regulations on the identification and screening of external events were investigated. Result: In order to minimize the cost of external event impact analysis of nuclear power plants, research on identifying and screening external events is being conducted. In general, in the identification process, all events that can occur at the NPPs are identified. In the screening process, external events are selected based on qualitative and quantitative criteria in most studies. Conclusions: The process of identifying and screening external events affecting NPPs is becoming important. This paper, summarize on how to identify and screen external events for a probabilistic safety assessment of NPPs. It is judged that research on bounding analysis and conservative analysis methods performed in the quantitative screening process of external events is necessary.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.9-18
• /
• 2023

Predicting the compressive strength of high-performance concrete (HPC) is challenging because of the use of additional cementitious materials; thus, the development of improved predictive models is essential. The purpose of this study was to develop an HPC compressive-strength prediction model using an ensemble machine-learning method of combined bagging and stacking techniques. The result is a new ensemble technique that integrates the existing ensemble methods of bagging and stacking to solve the problems of a single machine-learning model and improve the prediction performance of the model. The nonlinear regression, support vector machine, artificial neural network, and Gaussian process regression approaches were used as single machine-learning methods and bagging and stacking techniques as ensemble machine-learning methods. As a result, the model of the proposed method showed improved accuracy results compared with single machine-learning models, an individual bagging technique model, and a stacking technique model. This was confirmed through a comparison of four representative performance indicators, verifying the effectiveness of the method.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.2
• /
• pp.111-118
• /
• 2024

An in-structure response spectrum (ISRS) is required to evaluate the seismic performance of a nuclear power plant (NPP). However, when a new ISRS is required because of the change in the unique spectrum of an NPP site, considerable costs such as seismic response re-analyses are incurred. This study provides several approaches to generate approximate methods for ISRS scaling, which do not require seismic response re-analyses. The ISRSs derived using these approaches are compared to the original ISRS. The effect of the ISRS of the approximate method on the seismic response and seismic performance of one of the main systems of an NPP is analyzed. The ISRS scaling approximation methods presented in this study produce ISRSs that are relatively similar at low frequencies; however, the similarity decreases at high frequencies. The effect of the ISRS scaling approximate method on the calculation accuracy of the seismic response/seismic performance of the system is determined according to the degree of similarity in the calculation of the system's essential mode responses for the method.