벽면 센서노드의 방향정보를 이용한 파손감지 모니터링 시스템의 설계

전병찬,홍인식 한국지식정보기술학회 2023 한국지식정보기술학회 논문지 Vol.18 No.6

As with the recent earthquake in Turkiye, there is a lot of concern about earthquakes around the world. Korea is also on the rise in the frequency of earthquakes and is not an earthquake-safe country. Therefore, the importance of systems that monitor seismic design, damage, and internal cracks of buildings is increasing. Damage to the inner wall of a facility due to external factors such as earthquakes can be visually confirmed, but it is difficult to diagnose damage because it includes invisible inner wall or wall damage. One of the methods that can be monitored quickly when a building collapses due to a disaster or disaster is to use a smart sheet for damage detection. Damage to the structure of a building can be caused by various factors such as natural disasters, aging, or poor construction practices, which can lead to loss of life or property in the event of a building collapse. In addition, existing monitoring systems have limitations in detecting structural damage in the early stages, rely on sensors installed at specific locations, and can only detect damage in the area.Therefore, it may not be detected until damage to other parts of the building has progressed. In order to solve this problem, it is possible to collect and respond to the state information of the building in real time by using a sensor attached to the wall of the building and a crack detection tape to detect damage caused by natural disasters or structural defects of the building. It can also detect changes in the orientation of building walls or other physical parameters that may indicate damage, and provides accurate information to sensor nodes to detect damage in any part of the building. In this paper, we propose a damage detection monitoring system using the direction information of the wall sensor node to overcome these limitations.

Statistics based localized damage detection using vibration response

Dorvash, Siavash,Pakzad, Shamim N.,LaCrosse, Elizabeth L. Techno-Press 2014 Smart Structures and Systems, An International Jou Vol.14 No.2

Damage detection is a challenging, complex, and at the same time very important research topic in civil engineering. Identifying the location and severity of damage in a structure, as well as the global effects of local damage on the performance of the structure are fundamental elements of damage detection algorithms. Local damage detection is essential for structural health monitoring since local damages can propagate and become detrimental to the functionality of the entire structure. Existing studies present several methods which utilize sensor data, and track global changes in the structure. The challenging issue for these methods is to be sensitive enough in identifYing local damage. Autoregressive models with exogenous terms (ARX) are a popular class of modeling approaches which are the basis for a large group of local damage detection algorithms. This study presents an algorithm, called Influence-based Damage Detection Algorithm (IDDA), which is developed for identification of local damage based on regression of the vibration responses. The formulation of the algorithm and the post-processing statistical framework is presented and its performance is validated through implementation on an experimental beam-column connection which is instrumented by dense-clustered wired and wireless sensor networks. While implementing the algorithm, two different sensor networks with different sensing qualities are utilized and the results are compared. Based on the comparison of the results, the effect of sensor noise on the performance of the proposed algorithm is observed and discussed in this paper.

정적 변형률 데이터 기반 머신러닝에 의한 무도상 철도 판형교의 손상 탐지

문태욱 ( Taeuk Moon ),신수봉 ( Soobong Shin ) 한국구조물진단유지관리공학회 2020 한국구조물진단유지관리공학회 논문집 Vol.24 No.6

국내의 노후 철도교량이 증가함에 따라 노후화로 인한 유지관리비가 점점 증가하고 있으며, 지속적인 관리가 더욱 더 중요해지고 있다. 하지만 관리해야하는 노후 시설물은 증가하지만, 노후 시설물을 점검 및 진단을 할 수 있는 전문 인력은 부족해지고 있다. 이러한 문제를 해결하기 위해 본 연구는 정적 변형률 응답 데이터를 적용하여 AI 기술의 머신러닝 기법으로 구조물의 국부적인 손상을 탐지하는 개선된 학습모델을 제시하고자 한다. 손상탐지 머신러닝 학습 모델을 구성하기 위해 우선 무도상 철도 판형교의 설계도면을 참고하여 교량의 해석모델을 설정하였으며, 설정된 해석모델로 손상시나리오에 따른 정적변형률 데이터를 추출하여 통계적 기법을 이용해 교량의 신뢰도 기반의 Local 손상 지수를 제시하였다. 손상 탐지는 손상 유무 탐지, 크기 탐지, 위치 탐지 3단계의 과정을 수행하여 손상 크기 탐지에서 선형 회귀 모델을 추가로 고려해 임의의 손상을 탐지하였으며, 최종적으로 손상 탐지 머신러닝 분류 학습 모델과 회귀 모델을 이용한 임의의 손상 위치를 추정 및 검증하였다. As the number of aging railway bridges in Korea increases, maintenance costs due to aging are increasing and continuous management is becoming more important. However, while the number of old facilities to be managed increases, there is a shortage of professional personnel capable of inspecting and diagnosing these old facilities. To solve these problems, this study presents an improved model that can detect Local damage to structures using machine learning techniques of AI technology. To construct a damage detection machine learning model, an analysis model of the bridge was set by referring to the design drawing of a non-ballasted plate-girder railroad bridge. Static strain data according to the damage scenario was extracted with the analysis model, and the Local damage index based on the reliability of the bridge was presented using statistical techniques. Damage was performed in a three-step process of identifying the damage existence, the damage location, and the damage severity. In the estimation of the damage severity, a linear regression model was additionally considered to detect random damage. Finally, the random damage location was estimated and verified using a machine learning-based damage detection classification learning model and a regression model.

건물내벽의 자연재해 및 구조적 결함으로 인한 파손감지 모니터링 시스템의 설계

전병찬,홍인식 한국지식정보기술학회 2022 한국지식정보기술학회 논문지 Vol.17 No.6

Damage occurring on the inner wall of a building may be difficult to identify in relation to it, and it is difficult to diagnose because it includes not only visible damage but also invisible internal damage. The causes of damage to the inner walls of these buildings are various. Basically, there is damage due to aging of buildings due to sunlight, ultraviolet rays, moisture, wind, sulfate, harmful gases, mold, bacteria, moss, etc. This may cause damage. In addition, the inner wall of a building may appear in the form of externally identifiable defects such as cracks, water leakage, condensation, contamination, etc. Defects such as breakage of walls may occur. Methods for detecting damage occurring on the inner wall of a building include a method of using a thermal imaging camera, a method of detecting a leak occurring on a roof or basement, and an inner wall crack detection system using an embedded board that is being developed recently. Various systems have difficulty detecting blind spots or large areas. The RTD-1000 system is a system that can monitor leak detection in the water supply pipe network using a special pipe in which a detection wire called a leak detection pipe is inserted. RTD-1000 can monitor four paths sequentially, and TDR, the core of this system, can detect abnormalities in transmission lines using reflected waves. It is possible to use the smart sheet to adapt it for the detection of damage to the inner wall of a building. In this paper, we propose a building inner wall damage detection system that can monitor the damage of building inner wall using smart sheet and RTD-1000 system.

Statistics based localized damage detection using vibration response

Siavash Dorvash,Shamim N. Pakzad,Elizabeth L. LaCrosse 국제구조공학회 2014 Smart Structures and Systems, An International Jou Vol.14 No.2

Damage detection is a challenging, complex, and at the same time very important research topic in civil engineering. Identifying the location and severity of damage in a structure, as well as the global effects of local damage on the performance of the structure are fundamental elements of damage detection algorithms. Local damage detection is essential for structural health monitoring since local damages can propagate and become detrimental to the functionality of the entire structure. Existing studies present several methods which utilize sensor data, and track global changes in the structure. The challenging issue for these methods is to be sensitive enough in identifying local damage. Autoregressive models with exogenous terms (ARX) are a popular class of modeling approaches which are the basis for a large group of local damage detection algorithms. This study presents an algorithm, called Influence-based Damage Detection Algorithm (IDDA), which is developed for identification of local damage based on regression of the vibration responses. The formulation of the algorithm and the post-processing statistical framework is presented and its performance is validated through implementation on an experimental beam-column connection which is instrumented by dense-clustered wired and wireless sensor networks. While implementing the algorithm, two different sensor networks with different sensing qualities are utilized and the results are compared. Based on the comparison of the results, the effect of sensor noise on the performance of the proposed algorithm is observed and discussed in this paper.

Damage Detection at Welded Joint of Two-Dimensional Plane Model

Chang-Yong Chung,Hee-Chang Eun,Eun-Kyoung Seo 대한건축학회 2011 Architectural research Vol.13 No.4

Damage detection algorithms based on a one-dimensional beam model can detect damage within a beam span caused by flexure only but cannot detect damage at a joint with prescribed boundary conditions or at the middle part of a beam section where the neutral axis is located. Considering the damage at a welded joint of beam elements in steel structures and modeling the damage with twodimensional plane elements, this study presents a new approach to detecting damage in the depth direction of the joint and beam section. Three damage scenarios at the upper, middle, and lower parts of a welded joint of a rectangular symmetric section are investigated. The damage is detected by evaluating the difference in the receptance magnitude between the undamaged and damaged states. This study also investigates the effect of measurement locations and noise on the capability of the method in detecting damage. The numerical results show the validity of the proposed method in detecting damage at the beam’s welded joint.

One-Class Convolutional Neural Network (OC-CNN) Model for Rapid Bridge Damage Detection Using Bridge Response Data

Fadel Yessoufou,Jinsong Zhu 대한토목학회 2023 KSCE Journal of Civil Engineering Vol.27 No.4

This study proposes a numerical investigation for rapid bridge damage detection based on a semi-supervised deep learning (DL) model and a damage index (DI)-based Gaussian process. The proposed damage detection method uses bridge response data (acceleration and displacement data) from various damage scenarios within a simply supported girder bridge subjected to a two-axle moving vehicle load. As for semi-supervised learning, we used a one-class convolutional neural network (OC-CNN) model. This model combines a one-class (OC) classification algorithm with a simple one-dimensional convolutional neural network (1D CNN) configuration. The performance of the proposed OC-CNN model was evaluated through a numerical example of a vehicle-bridge coupling system. The proposed OC-CNN model trained using acceleration data showed promising results for different vehicle weights and speeds. These results offer confidence in using the prediction error loss of the proposed OC-CNN model as an ideal damage-sensitive feature for rapid bridge damage detection. In addition, the Gaussian process used in the DI can classify the prediction error losses resulting from the change induced by different damage severities (10%, 20%, and 30%) and different types of damage scenarios (single damage, double damages, and multiple damages). These results emphasize the potential of the proposed damage detection method to monitor the state of bridges in practical engineering.

Numerical evaluation for vibration-based damage detection in wind turbine tower structure

Nguyen, Tuan-Cuong,Huynh, Thanh-Canh,Kim, Jeong-Tae Techno-Press 2015 Wind and Structures, An International Journal (WAS Vol.21 No.6

In this study, the feasibility of vibration-based damage detection methods for the wind turbine tower (WTT) structure is evaluated. First, a frequency-based damage detection (FBDD) is outlined. A damage-localization algorithm is visited to locate damage from changes in natural frequencies. Second, a mode-shape-based damage detection (MBDD) method is outlined. A damage index algorithm is utilized to localize damage from estimating changes in modal strain energies. Third, a finite element (FE) model based on a real WTT is established by using commercial software, Midas FEA. Several damage scenarios are numerically simulated in the FE model of the WTT. Finally, both FBDD and MBDD methods are employed to identify the damage scenarios simulated in the WTT. Damage regions are chosen close to the bolt connection of WTT segments; from there, the stiffness of damage elements are reduced.

Numerical evaluation for vibration-based damage detection in wind turbine tower structure

Tuan-Cuong Nguyen,현탄칸,김정태 한국풍공학회 2015 Wind and Structures, An International Journal (WAS Vol.21 No.6

In this study, the feasibility of vibration-based damage detection methods for the wind turbine tower (WTT) structure is evaluated. First, a frequency-based damage detection (FBDD) is outlined. A damage-localization algorithm is visited to locate damage from changes in natural frequencies. Second, a mode-shape-based damage detection (MBDD) method is outlined. A damage index algorithm is utilized to localize damage from estimating changes in modal strain energies. Third, a finite element (FE) model based on a real WTT is established by using commercial software, Midas FEA. Several damage scenarios are numerically simulated in the FE model of the WTT. Finally, both FBDD and MBDD methods are employed to identify the damage scenarios simulated in the WTT. Damage regions are chosen close to the bolt connection of WTT segments; from there, the stiffness of damage elements are reduced.

비선형 파라메트릭 사영필터를 이용한 2차원 트러스 구조물의 손상추정에 적용된 손상지표의 유효성과 수렴성에 관한 연구

문효준,오창희,서일교 제주대학교 공과대학 첨단기술연구소 2005 尖端技術硏究所論文集 Vol.16 No.2

A study on the effectiveness and convergency of damage measures for structural damage detection using the nonlinear parametric projection filter algorithm is presented in this paper. Damage measures are associated with the change in mode shape and displacement due to structural damage. So, five damage measures are presented in this paper. They are static displacement, curvature of static displacement, natural frequency, mode shape and curvature of mode shape. The static data are obtained by loading static load to the structure. And modal and dynamic data of a vibrating structure can be gained without dismantling the structure, these data are very useful in various aspects. But the results of detection are affected by the kind of damage measures to be used. The effectiveness and convergency of these damage measures on the structural damage detection using the nonlinear parametric projection filtering algorithm are demonstrated with the numerical examples.