• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.33 no.6
• /
• pp.359-366
• /
• 2020

In this study, a simulation-based damage estimation method for helidecks is proposed using an artificial neural network. The structural members that share a connecting node in the helideck are regarded as a damage group, and a total of 37,400 damage scenarios are numerically generated by applying randomly assigned damage to up to three damage groups. Modal analysis is then performed for all the damage scenarios, which are selectively used as either training or validation or verification sets based on the purpose of use. An artificial neural network with three hidden layers is constructed using a PyTorch program to recognize the patterns of the modal responses of the helideck model under both damaged and undamaged states, and the network is successively trained to minimize the loss function. Finally, the estimated damage rate from the proposed artificial neural network is compared to the actual assigned damage rate using 400 verification scenarios to show that the neural network is able to estimate the location and amount of structural damage precisely.

• International conference on construction engineering and project management
• /
• 2015.10a
• /
• pp.339-343
• /
• 2015

To rapidly recover ceased functionality of a facility after a catastrophic seismic event, critical decisions on facility repair works are made within a limited period of time. However, prolonged damage assessment of facilities, due to massive damage in the surrounding region and the complicated damage judgment procedures, may impede restoration planning. To assist reliable structural damage estimation without a deep knowledge and rapid interactive analysis among facility damage and restoration operations during the approximate restoration project planning phase, we developed a prototype of distributed facility restoration simulations through the use of high-level architecture (HLA) (IEEE 1516). The simulation prototype, in which three different simulations (including a seismic data retrieval technique, a structural response simulator, and a restoration simulation module) interact with each other, enables immediate damage estimation by promptly detecting earthquake intensity and the restoration operation analysis according to estimated damage. By conducting case simulations and experiments, research outcomes provide key insights into post-disaster restoration planning, including the extent to which facility damage varies according to disaster severity, facility location, and structures. Additional insights arise regarding the extent to which different facility damage patterns impact a project's performance, especially when facility damage is hard to estimate by observation. In particular, an understanding of required type and amount of repair activities (e.g., demolition works, structural reinforcement, frame installation, or finishing works) is expected to support project managers in approximate work scheduling or resource procurement plans.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.5D
• /
• pp.507-517
• /
• 2012

The main purpose of this study is to address flooding resilient land use management strategy based on the distributional characteristics of storm damage areas in Gyeonggi-do. The employed methods are 1) Exploratory Spatial Data Analysis (ESDA) to understand the spatial patterns of storm damage areas occurred from 2005 to 2009, 2) Local Indicator of Spatial Association (LISA) to examine spatial autocorrelation existed in storm damage areas for the year of 2009. The results show that 1) crop land damage is very sensitive to heavy precipitation, 2) damaged buildings are found in all over the Gyeonggi areas, but relatively more damages are in the regions closed to the City of Seoul, 3) damaged roads-bridges, streams, and reaches are found in mostly rural areas, 4) building and crop land damage occurs mostly in lowlands with different spatial patterns. These findings imply that 1) it will be useful to consider the average distances and slopes of damaged building and crop lands from streams for the decision making of land use management strategy, 2) further management efforts are required in the north, east, and south regions of Gyeonggi areas to prevent roads-bridge, stream, and reach damages, 3) the present land use pattern needs to be carefully investigated by considering the damage clustered areas for the year of 2009 based on watershed and municipality boundaries.

• Smart Structures and Systems
• /
• v.25 no.1
• /
• pp.1-22
• /
• 2020

The modal strain energy method is one of the efficient methods for detecting damage in the structures. Due to existing some limitations in real-world structures, sensors can only be located on a limited number of degrees of freedom (DOFs) of a structure. Therefore, the mode shape values in all DOFs of structures cannot be measured. In this paper, a modified modal strain energy based index (MMSEBI) is introduced to locate damaged elements of structures when a limited number of sensors are used. The proposed MMSEBI is based on the reconstruction of mode shapes using Improved Reduction System (IRS) method. Therefore, in the first step by employing IRS method, mode shapes in slave degrees of freedom are estimated by those of master degrees of freedom. In the second step, the proposed MMSEBI is used to located damage elements. In order to evaluate the efficiency of the proposed method, two numerical examples are considered under different damage patterns considering the measurement noise. Moreover, the universal threshold based on statistical hypothesis testing principles is applied to damage index values. The results show the effectiveness of the proposed MMSEBI for the structural damage localization when comparing with the available damage index named MESBI. The results demonstrate that the presented method can be used as a practical strategy for structural damage identification, especially when a limited number of sensors are installed on the structure. Finally, the combination of MMSEBI and IRS method can provide a reliable tool to identify the location of damage accurately.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.45 no.4
• /
• pp.98-106
• /
• 2003

In order to set up structural improvement strategy against meteorological disasters of the shading structures in ginseng fields, structural safety analyses as well as some case studies of structural damage patterns were carried out. According to the results of structural safety analysis, allowable safe snow depth for type B(wood frame with single span) was 25.9 cm, and those for type A(wood frame with multi span) and type C and D (steel frame with multi span) were 17.6 cm, 25.8 cm, and 20.0 cm respectively. So types of shading structures should be selected according to the regional design snow depth. An experiential example study on meteorological disasters indicated that a strong wind damage was experienced once every 20 years, and a heavy snow damage once every 9.5 years. The most serious disasters were caused by heavy snow and it was found that a half break and complete collapse of structures were experienced by about 70% of snow damage. In addition to maintenance, repair and reinforcement, it is also recommended that improved model of shading structures for ginseng cultivation should be developed as a long term countermeasures against meteorological disasters.

• Steel and Composite Structures
• /
• v.19 no.5
• /
• pp.1055-1075
• /
• 2015

In order to investigate the accuracy and applicability of steel equivalent constitutive model, the calculated results were compared with typical tests of steel frames under static and dynamic loading patterns firstly. Secondly, four widely used models for time history analysis of steel frames were compared to discuss the applicability and efficiency of different methods, including shell element model, multi-scale model, equivalent constitutive model (ECM) and traditional beam element model (especially bilinear model). Four-story steel frame models of above-mentioned finite element methods were established. The structural deformation, failure modes and the computational efficiency of different models were compared. Finally, the equivalent constitutive model was applied in seismic incremental dynamic analysis of a ten-floor steel frame and compared with the cyclic hardening model without considering damage and degradation. Meanwhile, the effects of damage and degradation on the seismic performance of steel frame were discussed in depth. The analysis results showed that: damages would lead to larger deformations. Therefore, when the calculated results of steel structures subjected to rare earthquake without considering damage were close to the collapse limit, the actual story drift of structure might already exceed the limit, leading to a certain security risk. ECM could simulate the damage and degradation behaviors of steel structures more accurately, and improve the calculation accuracy of traditional beam element model with acceptable computational efficiency.

• Journal of the Korean Institute of Rural Architecture
• /
• v.12 no.4
• /
• pp.1-11
• /
• 2010

This study is a research examining reshaped pattern in the perspective of repairing parts and materials through repairing renovation of registered cultural properties constructed in modern times and results are as followings. Repairing construction of registered cultural properties of Jollanamdo according to parts is shown numerously in the order of roof, walls and windows, and mostly outworn as time passes on and leakage were the main cause of repair. Also when original shape was damaged by previous repair in the past, and this was another reason of repairing. It was surveyed that among the repairing job repair of damage occurred on the part of roofing area including water leakage, corrosion and damage of roofing material, and damage of groove channel were the most main cause of repair. Especially when roof leakage is occurred by outworn of roofing materials, this cause corrosion and damage of materials due to the damage of leakage parts and this cause repeating cycles of worse leakage again and again. Main repairing materials which deform the original shape of registered cultural properties were confirmed as copper plate used on the roof. Copper plate showing high frequency of application which replaces groove slate or cement roofing tile used on the roof before has high durability and anti-corrosion but it is considered improper material to recover original shape.

• Smart Structures and Systems
• /
• v.30 no.4
• /
• pp.397-410
• /
• 2022

It is essential to monitor the working conditions of bolt-connected joints, which are widely used in various kinds of steel structures. The looseness of bolts may directly affect the stability and safety of the entire structure. In this study, a guided wave-based method for bolt looseness detection and localization is presented for a joint structure with multiple bolts. SH waves generated and received by a small number (two pairs) of magnetostrictive transducers were used. The bolt looseness index was proposed based on the changes in the reconstructed responses excited by the time reversal signals of the measured unit impulse responses. The damage locations and local damage severities were estimated using the damage indices from several wave propagation paths. The back propagation neural network (BPNN) technique was employed to identify the local damages. Numerical and experimental studies were conducted on a lap joint with eight bolts. The results show that the total damage severity can be successfully detected under the effect of external force and measurement noise. The local damage severity can be estimated reasonably for the experimental data using the BPNN constructed by the training patterns generated from the finite element simulations.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.16 no.2
• /
• pp.87-94
• /
• 2012

A damage estimation method of wind turbine tower using natural frequency and mode shape is presented for effective condition monitoring. Dynamic analysis for a wind turbine was carried out to obtain the response of tower from which modal properties were identified. A neural network was learned based on training patterns generated by the changes of natural frequency and mode shape due to various damages. The changes of modal property were calculated using a program for modal parameter estimation. Damage locations and severities could be successfully estimated for 10 damage cases including multi-damage cases using the trained neural network. The damage severities for very small damages generally tends to be slightly under-estimated however, the identified damage locations agreed reasonably well with the accurate locations. Enhancement of the estimation result for very small damage and verification of the proposed method through experiment will be carried out by further study.

• Journal of the Mineralogical Society of Korea
• /
• v.31 no.3
• /
• pp.149-159
• /
• 2018

The Gyeongju Bunhwangsa Stone Brick Pagoda, which was built with bricks of andesite, is the oldest brick stone pagoda of Silla period. The damage patterns in the stone pagoda are pollutants such as white crust, black crust, discoloration, soil adsorption, and microorganisms, and repair materials. The damage pattern of structural factors in the Stone Brick Pagoda is a bulging phenomenon. According to the X-ray diffraction analysis, white crust are mainly consist of calcite ($CaCO_3$) and thermonatrite ($Na_2CO_3{\cdot}H_2O$) that evaporite finds in nature. Damage pattern varies depending on location of stone pagoda. The pollutants are first story body of pagoda. The microorganisms are confirmed at base, lion statues, first and second story capstone, and repair materials observed at base. The bulging phenomenon appeared on the first story body of the pagoda. Occupancy rates by damage type were higher in the order of microorganisms, pollutants, repair material, bulging phenomenon, and peeling. The highest percentage of individual damage patterns were black microorganisms (39.3%), followed by lichen (17.9%), discoloration (8.0%), white crust (5.5%), cement mortar (5.1%) and peeling (3.1%).