https://doi.org/10.11112/jksmi.2025.29.5.1

이승재(Seung-Jae Lee) ; 최경규(Kyoung-Kyu Choi) ; 박성호(Sung-Ho Park) ; 김승희(Seung-Hee Kim)

This study presents and validates a stereo vision-based non-contact system for measuring the dynamic responses of structures. A four-story steel frame specimen was tested on a uniaxial shaking table using five types of ground motions scaled to three peak ground acceleration (PGA) levels (0.1g, 0.3g, and 0.5g). The stereo vision measurement system, comprising synchronized high-resolution cameras, was calibrated through reprojection error minimization and coordinate alignment using physical reference points on the specimen. Vision-based displacement measurements were quantitatively compared with wire-type LVDTs and accelerometers, showing maximum errors within ±2 mm and an average root mean square error (RMSE) of 1.06 mm. Frequency response functions (FRFs) were derived from the vision data to extract natural frequencies and mode shapes, which closely matched accelerometer-based modal results, achieving modal assurance criterion (MAC) values above 0.9 in most modes. Damping ratios were also estimated using the half-power bandwidth method and compared with Rayleigh damping models. The findings demonstrate that the proposed stereo vision-based system provides sufficient accuracy and resolution for experimental modal analysis of multi-story structural frames and offers potential for application in structural health monitoring.

https://doi.org/10.11112/jksmi.2025.29.5.12

이용준(Yong-Jun Lee) ; 박경훈(Kyung-Hoon Park) ; 선종완(Jong-Wan Sun)

This study proposes an asset value evaluation method that takes into account the performance of pedestrian overpasses in South Korea. Conventional methods for asset valuation have limitations in that they depreciate assets by a fixed amount over the standard service life without considering the actual condition of the facilities, often resulting in undervaluation. To address this issue, this study derives performance indicators for pedestrian overpasses using the Weibull distribution and introduces two evaluation methods: the Estimated Service Life Depreciation (ESLD) method and the Performance-Based Depreciation (PBD) method. The ESLD method evaluates asset value based on the estimated average service life, while the PBD method uses a combination index derived from both reliability, based on service years, and condition grades. To estimate the performance indicators of domestic pedestrian overpasses, various datasets such as basic facility information, inspection records, and data on closure or demolition were processed into lifecycle data. The processed data were analyzed using the Weibull distribution to calculate indicators such as mean service life and reliability. These performance indicators were then applied to the proposed valuation models. The analysis results demonstrate that the proposed methods more accurately reflect the actual condition and remaining service life of in-service pedestrian overpasses, leading to more rational and realistic asset valuations compared to conventional methods. These methods are expected to provide meaningful indicators for facility managers, supporting more effective decision-making in maintenance and management.

https://doi.org/10.11112/jksmi.2025.29.5.21

홍성욱(Sung-Wook Hong) ; 서은아(Eun-A Seo) ; 이호재(Ho-Jae Lee)

This study investigates how ambient climate conditions influence the early age compressive strength of cast in place concrete and evaluates the feasibility of predicting strength from external weather data alone. Field compressive strength results from a construction site in Ho Chi Minh City, Vietnam, were paired with climate records for the curing period. The relationship between cumulative ambient temperature and the 7 day compressive strength was first examined. To account for other major climatic factors, K means clustering based on humidity, wind speed, and precipitation grouped the data into three clusters: general conditions, high wind conditions, and frequent rainfall conditions. Under rainfall dominant conditions, the correlation between temperature and strength weakened, likely due to surface cooling and delayed hydration. In contrast, partitioning by wind condition and using a maturity index based on ambient temperature produced stable regressions for the other clusters. The proposed approach enables early age quality control and scheduling without embedded sensors, supporting structural maintenance and site operations in hot and humid regions.

https://doi.org/10.11112/jksmi.2025.29.5.29

심승보(Seungbo Shim) ; 양엄지(Eomzi Yang)

In recent construction sites, images of rebar and reinforcement often suffer from various degradation factors, such as low resolution, motion blur, and compression loss, as well as random noise introduced by the surrounding environment. These factors lead to the loss of structural information in the images, thereby reducing the accuracy and reliability of artificial intelligence-based analysis. In particular, due to the nature of construction sites, it is difficult to secure sufficient high-quality reference images, resulting in a lack of training data necessary for super-resolution algorithm development. To address this limitation, this study proposes a novel blind super-resolution method that utilizes the Stable Diffusion X4 model to generate sharp, high-quality synthetic images from low-resolution images, which are then used together with real rebar images as training data. The proposed method develops a super-resolution algorithm by simultaneously using low-resolution images and synthesized high-resolution images as inputs. Experiments were conducted by applying the method to three state-of-the-art super-resolution neural network models. As a result, experiments conducted by applying the proposed method to three state-of-the-art super-resolution neural network models demonstrated that, compared to the conventional training approach using only real low- and high-resolution image pairs without synthetic data, the proposed strategy achieved an average improvement of 0.45 dB in resolution metrics and 0.85% in structural similarity metrics, while more accurately restoring the shapes and fine structural details of rebar and reinforcement. This study demonstrates that even in construction site environments where image quality is degraded, synthetic data can be effectively utilized to improve super-resolution performance. Furthermore, it is expected to contribute to enhancing the reliability of rebar condition analysis and advancing automation technologies for image-based structural management in the future.

https://doi.org/10.11112/jksmi.2025.29.5.40

이종석(Jong-Suk Lee) ; 홍동의(Dong-Uie Hong) ; 이동훈(Dong-Hoon Lee) ; 박진완(Jin-Wan Park) ; 김민준(Min-Jun Kim)

Recent earthquake damage surveys in Korea indicate that nonstructural components often experience more than three times the damage of structural components, largely due to amplification of high-frequency seismic waves. Previous studies have focused mainly on structural responses, with limited experimental verification of nonstructural or internal equipment behavior. This study investigates the seismic response of a four-story steel cabinet model with thin plates simulating internal devices, tested on a shaking table with and without seismic isolation. Under non-isolated conditions, internal components showed acceleration responses up to four times greater than the exterior structure, suggesting potential functional failure. In contrast, seismic isolation reduced responses at all measurement points by 60?90%. FRF and FFT analyses confirmed a shift in resonant frequencies and suppression of high-frequency amplification. These findings experimentally demonstrate that seismic isolation tables can contribute not only to structural stability but also to the functional reliability of equipment installed in nonstructural components of nuclear power plants and other critical facilities.

https://doi.org/10.11112/jksmi.2025.29.5.50

지구철(Koochul Ji) ; 이승은(Seung Eun Lee) ; 장승환(Sung-Hwan Jang)

Intense snowfall and icing resulting from climate change provide a significant risk to railway operations, while traditional de-icing techniques employing calcium chloride lead to structural deterioration and environmental issues. This study assessed the feasibility of a high-performance heating concrete technique that incorporates carbon fibers as a heating element combined with functional elements. Following the selection of carbon fibers exhibiting optimal thermal performance, they were incorporated into plain, PS ball, and PCM concrete, respectively. Their thermal behaviors were then examined in comparison with the use of electric power. The experimental findings indicated that PS ball concrete using steel slag was beneficial for expedited snow melting owing to its enhanced heat diffusion rate and broader heating area. Conversely, PCM concrete exhibited superior thermal storage capabilities, maintaining a stable temperature after a power interruption, hence indicating potential for optimizing energy efficiency. This study concluded that the performance of heating concrete can be regulated for distinct goals (fast heating versus long-term thermal storage) by the utilization of functional components (PS balls, PCM). This indicates the possibility of creating an environmentally sustainable, tailored snow-melting system to substitute calcium chloride, thus improving the safety and longevity of railway infrastructure.

https://doi.org/10.11112/jksmi.2025.29.5.58

전준창(Jun-Chang Jeon) ; 이희현(Hee-Hyun Lee) ; 유(Hoon Yoo)

The field load test is an essential procedure for evaluating the load carrying capacity of bridges. However, during the test, public complaints due to traffic closures and safety concerns of workers working at night and high altitude have been pointed out. To address these issues, SISTech and CTC have proposed a simplified field load test method as an alternative. This study aims to assess the validity of the simplified method in terms of analyzing existing detailed safety inspection cases and conducting verification test on PSC I-girder bridge. A total of 20 detailed safety inspection reports for PSC I-girder bridges were employed for analyzing response ratio, impact factor, girder/bearing condition rating, and so on. Based on this analysis, response correction factors were estimated from both the existing and simplified field load test methods and compared. Additionally, the field load test using both the existing and simplified methods were carried out on PSC I-girder bridge to evaluate its load carrying capacity by each method. The above analysis and test results confirmed that the simplified field load test method can evaluate the load carrying capacity with an acceptable margin of error within approximately 5% in comparison with the existing method.

https://doi.org/10.11112/jksmi.2025.29.5.66

한상석(Sangseok Han) ; 정호성(Hoseong Jeong) ; 김재민(Jae Min Kim) ; 윤영택(Yeongtaek Yoon) ; 김강수(Kang Su Kim)

This study presents a fire detection algorithm that integrates image processing techniques with an autonomous mobile robot system. To enable real-time object detection and segmentation within images, the You Only Look Once (YOLO) framework was employed. Multiple versions of the YOLO model were trained and evaluated to analyze their performance variations. Additionally, a stereo vision approach combining the Oriented FAST and Rotated BRIEF (ORB) algorithm with Semi Global Block Matching (SGBM) was utilized to estimate the actual distance to fire. To validate the accuracy of distance estimation, a series of experiments were performed. The integrated image processing techniques were employed together with an autonomous driving algorithm to support real-time execution on a mobile robotic platform. The proposed algorithm and the developed platform demonstrated effective real-time fire detection and notification capabilities within a simulated environment.

https://doi.org/10.11112/jksmi.2025.29.5.76

진현철(Hyun-Chul Jin) ; 노병철(Byeong-Cheol Lho)

The steel-composite rigid frame method maximizes structural efficiency by combining the high tensile strength of steel with the excellent compressive strength of concrete. Continuous development has been driven by its advantage of reducing construction time through the factory fabrication of steel girders. With the growing demand for low-profile, long-span structures in projects like river maintenance, this method is increasingly being applied to bridges with spans exceeding 40 meters. This study proposes and evaluates the T-wing PS (Prestressed) steel-composite rigid frame method, which enhances structural efficiency by installing a transverse T-shaped member (T-wing) at the girder ends and introducing prestress using PS steel rods. To analyze its structural behavior, a finite element analysis was conducted for comparison with conventional methods, and a static loading test was performed to verify its actual performance. The numerical analysis of a 40-meter span bridge showed that the T-wing method (with four PS rods) reduced the positive moment at mid-span by approximately 29% compared to a non-prestressed frame and by 17% compared to a conventional method using two PS rods. The static loading test recorded a maximum load of 523 kN and a maximum deflection of 58 mm, with the PS rods exhibiting uniform linear strain, confirming that no eccentric behavior occurred in the T-wing member. Through this numerical and experimental verification, it was demonstrated that the T-wing PS steel-composite rigid frame method effectively enhances structural performance through prestressing and exhibits stable behavior under external loads, proving its suitability for rigid frame structures.

https://doi.org/10.11112/jksmi.2025.29.5.84

조연상(Younsang Cho) ; 강만성(Man-Sung Kang) ; 김호진(Hojin Kim) ; 김재환(Jaehwan Kim) ; 안윤규(Yun-Kyu An)

This study proposes an automated structural event (ship collision, typhoons, earthquakes, etc.) classification model that applies a statistical-based probabilistic data augmentation technique to overcome the learning limitations of rare events by utilizing data collected via IoT sensors for real-time safety diagnosis of cable-stayed bridges. Representative data were extracted from actual measurement data to analyze abnormal signals and rare events occurring on real bridges. Adaptive thresholds were selected for each sensor based on the extracted representative data per sensor channel to determine the presence of abnormal signals. Based on the values classified according to the presence of abnormal signals, a Sensor Spatial Matrix (SSM) was constructed, and the patterns of the SSM observed for event were analyzed. For the SSM patterns, a Gumbel distribution-based probabilistic modeling was used to statistically reproduce the active patterns of rare events, generating synthetic data for training. Generating synthetic training data confirmed the ability to address the problem of sparse data for rare events on bridges. Applying this data augmentation technique to train a CNN-based event classification model ensured stability in model training and reduced the tendency for the trained model to overgeneralize. Field experiments conducted on the Jindo Bridge in Jindo, South Korea, achieved a classification accuracy of 97.46%. This research enhances the real-time automatic diagnostic capability of Structural Health Monitoring (SHM) and presents a methodology to compensate for insufficient data on rare events.

https://doi.org/10.11112/jksmi.2025.29.5.93

윤수민(Su Min Yoon) ; 이윤정(Yoon Jung Lee) ; 정호성(Hoseong Jung) ; 할리오나(Khaliunaa Darkhanbat) ; 김강수(Kang Su Kim)

To ensure structural stability and buildability in 3D concrete printing (3DCP), it is essential to measure and evaluate the rheological properties of fresh concrete in real time. However, conventional rheometer-based tests, although highly precise, have limited applicability in practice due to the high cost of equipment, long measurement time, and reliance on operator expertise. In this study, the rheological properties (yield stress and viscosity) and ultrasonic signals of ultra-high-performance concrete (UHPC) for 3D printing were measured, and their correlations were analyzed. The experimental variables were set as the water-to-binder ratio (W/B) and the dosage of superplasticizer (SP). Ultrasonic measurements were performed using a transmitter and receiver operating at 54 kHz, and the attenuation coefficient was calculated from the acquired signals. The results showed that the attenuation coefficient increased as the ultrasonic signal was weakened by particle dispersion and pore structure during the early stage of hydration. In particular, mixtures with lower W/B and lower SP content exhibited a rapid increase in attenuation coefficient over time, whereas those with higher W/B and higher SP content showed a more gradual increase. These variations in attenuation coefficient were consistent with the time-dependent changes in yield stress and viscosity. From these results, an artificial neural network (ANN) model was developed using attenuation coefficient, mixture parameters, and initial rheological properties as input variables to predict the yield stress and viscosity of fresh concrete. The proposed model achieved coefficient of determination (R²) of 0.81 and 0.82 for yield stress and viscosity, respectively, in the test set.

https://doi.org/10.11112/jksmi.2025.29.5.103

권유진(U-Jin Kwon) ; 백종하(Jong-Ha Paik) ; 백은림(Eun-Rim Baek) ; 이상호(Sang-Ho Lee)

Low-rise RC piloti buildings, which are prevalent in Korea, are particularly vulnerable to seismic load when torsional irregularity is present. Current seismic design codes evaluate torsional moment using static eccentricity based on elastic stiffness. However, this approach does not fully evaluate the actual torsional response once the structure shows the inelastic range, due to translates in the center of rigidity. This study quantitatively investigates the influence of rigidity and strength ratios between wall and column in the soft first story on the nonlinear torsional behavior of irregular piloti buildings. Analytical models with various rigidity and strength ratios are analyzed non-linear static(push-over) analysis. The results of this study show that inelastic behavior was primarily governed by the strength ratio rather than the rigidity ratio. When the wall-to-column strength ratio was low, wall yielded first, causing the center of rigidity to translate toward the center of mass and reducing torsional effects. In such cases, the current codes tended to overestimate torsional moments. Otherwise, with a high strength ratio, weaker columns yielded first, translation the center of rigidity further away from the center of mass and amplifying eccentricity. This led to torsional moments up to 10% higher than those predicted by elastic analysis, suggesting that the current design approach may be unresonable. For the safe seismic designs of torsionally irregular piloti buildings, both rigidity and strength ratios between vertical members should be considered to accurately evaluate torsional moment variations arising from inelastic translates of the center of rigidity.

https://doi.org/10.11112/jksmi.2025.29.5.114

권유진(U-Jin Kwon) ; 백종하(Jong-Ha Paik) ; 이상호(Sang-Ho Lee) ; 백은림(Eun-Rim Baek)

Low-rise reinforced concrete (RC) piloti buildings are highly vulnerable to torsion due to the eccentric placement of core walls, a fact confirmed by numerous failures during the 2017 Pohang earthquake. This study aims to identify the actual inelastic collapse mechanism of piloti buildings and to propose key design parameters for ensuring their seismic safety. To achieve this, a three-dimensional non-linear static analysis (push-over) is performed with the wall-to-column strength ratio (S) and the planar aspect ratio (AR) as the main variables. The results of this study showed that the collapse mechanism was dominantly governed by the strength ratio rather than the initial static eccentricity. When the strength ratio was low (S≤4), a stable, ductile failure was induced by the wall yielding first, which effectively controlled the torsional behavior. In contrast, when the strength ratio was high (S≥6), an unstable, brittle failure occurred due to the premature yielding of weak-side columns, which amplified the torsional response. This difference in collapse mechanisms was clearly identified by the post-yield movement direction of the center of strength (CS). In models with a low strength ratio, the CS moved inward toward the center of mass (CM), indicating a balanced strength distribution. Conversely, in models with a high strength ratio, the CS moved outward, indicating a worsening of the strength imbalance. These models with excessively high strength ratio and aspect ratio failed prematurely without achieving their designed strength. Therefore, it is essential to move beyond static indices like initial eccentricity. Instead the design should focus on controlling the failure sequence by maintaining a wall-to-column strength ratio of 2.0 or less.

https://doi.org/10.11112/jksmi.2025.29.5.124

나상일(Sangil Na) ; 최웅규(Woonggyu Choi) ; 김승우(Seungwoo Kim) ; 박승희(Seunghee Park)

Recent lightweight footbridges can experience excessive vibrations even under pedestrian loads. Conventional vibration measurements primarily rely on accelerometers, which require direct attachment to the structure. This approach is cumbersome, and multi-point measurements increase wiring complexity and costs. This study proposes a practical and efficient non-contact method for evaluating footbridge vibration serviceability using smartphone video and computer vision techniques. The recorded video was processed using Eulerian Video Magnification based on the avoidance frequency range suggested in domestic design guidelines, enabling visualization of subtle motions. The Lucas Kanade Pyramid Optical Flow algorithm was then applied to estimate time-series displacements, which were converted to real world units using the handrail as a reference length. Welch Power Spectrum Density analysis was performed to obtain the maximum vertical acceleration and natural frequency, and the damping ratio was calculated using logarithmic linear regression. Experiments on a 72 m single span footbridge clearly showed that the proposed method achieved an average error rate below 2.05% for maximum vertical acceleration and natural frequency compared to reference accelerometer data, with similar damping ratios. The natural frequency of approximately 1.875 Hz matched the sensor measurements, confirming that the proposed non-contact video-based method accurately captures the essential dynamic behavior of footbridges.

https://doi.org/10.11112/jksmi.2025.29.5.132

전의익(Eui-Ik Jeon) ; 장다혜(Da-Hye Jang) ; 박현수(Hyun-Soo Park)

While Uncrewed Aerial Vehicle inspections enhance data acquisition, subsequent analysis is often limited to simple crack detection, lacking quantitative analysis and validation against on-site measurements. To address this gap, this study proposes and validates an integrated workflow that automatically detects and quantifies micro-cracks as fine as 0.1 mm from facade orthomosaics to create an exterior damage map. The methodology enhances image quality via a super-resolution model, detects cracks using semantic segmentation, and calculates their length and width through vectorization post-processing. The reliability of these quantitative results was validated against on-site measurements performed by an expert. The results demonstrated a high crack detection precision of 90%. The AI-estimated crack lengths showed an average similarity of 91% to expert measurements, and the crack width estimation yielded a low mean error of 0.0175 mm. This study proves the practicality of using automated quantitative analysis from drone imagery to create reliable exterior damage maps, thereby enhancing the efficiency and objectivity of bridge safety inspections.

https://doi.org/10.11112/jksmi.2025.29.5.141

송성휘(Seonghwi Song)

This study analyses the axial compressive behavior of prefabricated concrete-filled steel tube (CFT) columns considering various parameters of internal ribs, and quantitatively evaluates their structural performance. To address the inherent weaknesses associated with welding in conventional CFT columns, a prefabricated CFT column that can be assembled without welding was proposed. Finite element analyses (FEA) were conducted using ABAQUS based on experimental data from previous studies to analyze the influence of key parameters such as rib angle, rib length, and rib height on the axial compressive behavior. The FEA results revealed that a rib angle of 45° provided the highest ultimate compressive strength and axial stiffness, while longer and lower ribs exhibited superior axial compressive performance. Furthermore, a parametric study was performed to determine the optimal rib aspect ratio that ensures adequate compressive capacity.

https://doi.org/10.11112/jksmi.2025.29.5.149

이문석(Moon-Seok Lee) ; 최창식(Chang-Sik Choi) ; 배백일(Baek-Il Bae)

In precast concrete (PC) structures, hooked bars have attracted attention as an alternative to straight bars, as they can achieve sufficient anchorage performance with shorter lap lengths, thereby improving constructability and cost efficiency. When determining lap lengths between PC members, the development length equations for hooked bars specified in current design codes (ACI 318-25, KDS 14 20 52) are derived from beam? column joint tests, and thus their direct applicability to hooked bar lap splices is limited. This study experimentally investigates the performance of 180° hooked bar lap splices by testing 10 precast slab specimens under four-point bending. The key variables considered were splice length, bar diameter, concrete compressive strength, and yield strength of reinforcement, which were determined based on the ACI 318-25 hooked bar development length equation. Splice lengths were set to 0.3?1.0 times the code-based development length to examine the effective lap length of hooked bars required in practice. Test results indicated that the bar stresses measured in spliced specimens were 1.07?1.08 times greater than the stresses predicted by the KDS 14 20 52 equations, suggesting that the current provisions can be reasonably applied to estimate lap splice lengths of 180° hooked bars.