• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.77-91
• /
• 2022

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.

• Journal of Korean Society of Disaster and Security
• /
• v.12 no.2
• /
• pp.73-82
• /
• 2019

Recently, as the occurrence frequency of sudden floods due to climate change increased, the flood damage on riverside social infrastructures was extended so that there has been a threat of overflow. Therefore, a rapid prediction of potential flooding in riverside social infrastructure is necessary for administrators. However, most current flood forecasting models including hydraulic model have limitations which are the high accuracy of numerical results but longer simulation time. To alleviate such limitation, data driven models using artificial neural network have been widely used. However, there is a limitation that the existing models can not consider the time-series parameters. In this study the water surface elevation of the Hangang River bridge was predicted using the NARX model considering the time-series parameter. And the results of the ANN and RNN models are compared with the NARX model to determine the suitability of NARX model. Using the 10-year hydrological data from 2009 to 2018, 70% of the hydrological data were used for learning and 15% was used for testing and evaluation respectively. As a result of predicting the water surface elevation after 3 hours from the Hangang River bridge in 2018, the ANN, RNN and NARX models for RMSE were 0.20 m, 0.11 m, and 0.09 m, respectively, and 0.12 m, 0.06 m, and 0.05 m for MAE, and 1.56 m, 0.55 m and 0.10 m for peak errors respectively. By analyzing the error of the prediction results considering the time-series parameters, the NARX model is most suitable for predicting water surface elevation. This is because the NARX model can learn the trend of the time series data and also can derive the accurate prediction value even in the high water surface elevation prediction by using the hyperbolic tangent and Rectified Linear Unit function as an activation function. However, the NARX model has a limit to generate a vanishing gradient as the sequence length becomes longer. In the future, the accuracy of the water surface elevation prediction will be examined by using the LSTM model.

• Steel and Composite Structures
• /
• v.34 no.4
• /
• pp.499-509
• /
• 2020

The purpose of this paper is to investigate the degradation law of stiffness of steel-concrete composite beams after certain fatigue loads. First, six test beams with stud connectors were designed and fabricated for static and fatigue tests. The resultant failure modes under different fatigue loading cycles were compared. And an analysis was performed for the variations in the load-deflection curves, residual deflections and relative slips of the composite beams during fatigue loading. Then, the correlations among the stiffness degradation of each test beam, the residual deflection and relative slip growth during the fatigue test were investigated, in order to clarify the primary reasons for the stiffness degradation of the composite beams. Finally, based on the stiffness degradation function under fatigue loading, a calculation model for the residual stiffness of composite beams in response to fatigue loading cycles was established by parameter fitting. The results show that the stiffness of composite beams undergoes irreversible degradation under fatigue loading. And stiffness degradation is associated with the macrobehavior of material fatigue damage and shear connection degradation. In addition, the stiffness degradation of the composite beams exhibit S-shaped monotonic decreasing trends with fatigue cycles. The general agreement between the calculation model and experiment shows good applicability of the proposed model for specific beam size and fatigue load parameters. Moreover, the research results provide a method for establishing a stiffness degradation model for composite beams after fatigue loading.

• Computational Structural Engineering
• /
• v.10 no.3
• /
• pp.157-165
• /
• 1997

This research aims to develop an algorithm of optimal transducer placement for health monitoring of large structural system. The structural vibration response-based health monitoring is considered one of the best for the system which requires a long-term, continuous monitoring. In its experimental modal testing, however, it is difficult to decide on the measurement locations and their number, especially for complex structures, which have a major influence on the quality of the results. In order to minimize the number of sensing operations and optimize the transducer location while maximizing the accuracy of results, this paper discusses about an optimum transducer placement criterion suitable for the identification of structural damage for continuous health monitoring. As a criterion algorithm, it proposes the Kinetic Energy Optimization Technique (EOT), and then addresses the numerical issues which are subsequently applicable to actual experiment where a bridge model is used. By using the experimental data, it compares the EOT with the EIM(Effective Indefence Method) which is generally used to optimize the transducer placement for the damage identification and control purposes. The comparison conclusively shows that the EOT algorithm proposed in this paper is preferable when a structure is to be instrumented with fewer sensors for monitoring purpose.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.05b
• /
• pp.1138-1142
• /
• 2005

The tidal river is a river affected by tide, which causes the water level to rise and fall two times everyday periodically. The local velocity across the river could be very fast because of the cross-sectional characteristics of the river even though it's not a rainy season. Therefore extreme local scour could take place around hydraulic structures such as piers and caissons due to backward flow velocity. For the construction of pier foundation of Ilsan-bridge In the Han River, the field observations were performed to get the velocity and water level. The numerical analysis was performed by HEC-RAS(UNET). The relationship between measured maximum velocity and calculated mean velocity is achieved, which is used to estimate the velocity and water level as the construction is proceeding. Countermeasures for scour were designed with the results of the hydraulic analysis to avoid potential damage during construction work. According to the results of monitoring, the velocity increase after temporary road embankment was negligible, from which it is considered that the degradation of main channel compensated for the constriction of cross-section by embankment.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2289-2291
• /
• 2002

In maned airport, crews may have risks as they manage passenger control system in IPMS in damage situations such as fire in a airport. So the application of unmanned autonomous system can reduce the number of boarding crews and attribute to safe airplane transportation. PBBC model can be used to obtain control strategy, and airplane and enhance operators' skill by simulating the airport. The paper suggests an intelligent system of the pbbc control using microprocessor in integrated platform management system which can take measures against passenger situation of a airplane excluding unnecessary warnings with undamaged situations. The system here detected the passenger more accurately and adopted more appriate measures according to airplane status compared with conventional systems.

• Annual Conference of KIPS
• /
• 2002.04b
• /
• pp.1203-1206
• /
• 2002

In maned airport, crews may have risks as they manage passenger control system in IPMS in damage situations such as fire in a airport. So the application of unmanned autonomous system can reduce the number of boarding crews and attribute to safe airplane tranportation. PBBC model can be used to obtain control strategy, and airplane and enhance oprators' skill by simulating the airport. The paper suggests an intelligent system of the pbbc control using microprocessor in integrated platform management system which can take measures against passenger situation of a airplane excluding unnecessary warnings with undamaged situations. The system here detected the passenger more accurately and adopted more appriate measures according to airplane status compared with conventional systems.

• Smart Structures and Systems
• /
• v.30 no.1
• /
• pp.17-34
• /
• 2022

For steel structures, fatigue cracks are critical damage induced by long-term cycle loading and distortion effects. Vision-based crack detection can be a solution to ensure structural integrity and performance by continuous monitoring and non-destructive assessment. A critical issue is to distinguish cracks from other features in captured images which possibly consist of complex backgrounds such as handwritings and marks, which were made to record crack patterns and lengths during periodic visual inspections. This study presents a parametric study on image-based crack identification for orthotropic steel bridge decks using captured images with complicated backgrounds. Firstly, a framework for vision-based crack segmentation using the atrous convolution-based Deeplapv3+ network (ACDN) is designed. Secondly, features on crack images are labeled to build three databanks by consideration of objects in the backgrounds. Thirdly, evaluation metrics computed from the trained ACDN models are utilized to evaluate the effects of obstacles on crack detection results. Finally, various training parameters, including image sizes, hyper-parameters, and the number of training images, are optimized for the ACDN model of crack detection. The result demonstrated that fatigue cracks could be identified by the trained ACDN models, and the accuracy of the crack-detection result was improved by optimizing the training parameters. It enables the applicability of the vision-based technique for early detecting tiny fatigue cracks in steel structures.

• Journal of the Korean Society of Safety
• /
• v.38 no.2
• /
• pp.81-86
• /
• 2023

In this study, a one-dimensional numerical model was constructed to propose a flood control plan linked with the dam and river basin for the flood events of the Seomjin River in 2020. The flood level reduction of the downstream river was tested based on a scenario operation of the Seomjingang Dam and was also analyzed when a storage pocket was newly constructed as one of the river basin measures. It was confirmed that Seomjingang Dam's flood control capacity would be increased if the flood limit level was drastically lowered from the current EL. 196.5 m to EL. 188.0 m. In addition, if the upper area of the (old) Geumgok Bridge (which suffered great damage due to the loss of the levee) is used as a storage pocket, it would be effective in preventing floods in the lower area of it. In the era of the climate crisis, more integrated flood management is needed and basic river management must be observed.

• The Journal of Engineering Geology
• /
• v.3 no.2
• /
• pp.177-190
• /
• 1993

When soft soils are effected by unsymmetrical surcharge due to embankement and abutements of a bridge, large plastic sheraring deformations such as settlements, lateral displacements, upheavals and sliding shearing failure in the soils occurred and they have often damaged considerabily to the soils and structure. This study examines the existing theoretical background for the behavior of the displacement of soils by unsymmetrical surcharge on the soft soils and compares the analytical results to the actual measurements performed through the model test. The procedures of model test are that a model stock device is made and soft soils are filled in a container which fixes the soils. Then the displacements observed when surcharge load increa ses by regular interval at undrainage condition. It analyzes the relation of soil characteristics to displacement, critical surcharge and ultimate bearing capadty, condition of plastic flow and lateral flow pressure, comparing them with the existing theories. Understanding the causes of lateral displacement in soft soils due to unsymmetrical surchages will prevent a damage in advance.