• 한국재난정보학회 논문집
• /
• 제17권3호
• /
• pp.556-567
• /
• 2021

연구목적: 본 연구에서는 무인이동체를 활용한 철도교량의 외관조사 점검을 보다 효율적이고 신뢰성 있게 점검을 위하여 무인이동체를 통해 촬영된 이미지를 바탕으로 다양한 방식의 딥러닝 기반 자동 손상 분석기술을 검토하였다. 연구방법: 취득된 이미지를 바탕으로 손상항목을 정의하고 학습데이터로 추출하여 딥러닝 분석 모델을 생성하였다. 그리고 철도교량의 외관 손상 중 균열, 콘크리트 박리·박락, 누수, 철근노출에 대한 손상 이미지를 학습한 모델을 적용하여 자동 손상 분석 결과로 테스트하였다. 연구결과: 분석 결과 평균 95%이상 검측 재현율을 도출하는 분석 기법을 검토할 수 있었다. 이와 같은 분석 기술은 기존 육안점검 결과 대비 보다 객관적이고 정밀한 손상 검측이 가능하다. 결론: 본 연구를 통해 개발된 기술을 통해 철도 유지관리 분야에서 무인이동체를 활용한 정기점검 시 자동손상분석을 통한 객관적인 결과도출과 기존 대비 소요시간, 비용저감이 가능할 것으로 기대된다.

• 소성∙가공
• /
• 제8권2호
• /
• pp.216-221
• /
• 1999

In this paper, an application-oriented approach to piercing analysis in automatic forging simulation by the rigid-viscoplastic finite element mehtod is presented. In the presented approach, the accumulated damage is traced and the piercing instant is determined when the accumulated damage reaches the critical damage value. A method of obtaining the critical damage value by comparing the tensile test result with the analysis one is given. The presented approach is verified by experiments and applied to automatic simulation of a sequence of 6-stage forging processes.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 춘계학술대회논문집A
• /
• pp.7-12
• /
• 2000

The automatic turbine startup system provides turbine control based on thermal stress. During the startup, control system monitors and evaluates main components of turbine using damage mechanism and life assessment. In case of valve chest, the temperature of inner/outer wall is measured by thermo-couples and the safety of these values are evaluated by using allowable ${\Delta}T$ limit curve during the startup. Because allowable ${\Delta}T$ limit curve includes life assessment, it is possible to apply this curve to turbine control system. In this paper, low cycle fatigue damage and combined rupture and low cycle fatigue damage criterion proposed for yielding the allowable ${\Delta}T$ limit curve of CV(control valve) chest. To calculate low cycle fatigue damage, the stress analysis of valve chest has peformed using FEM. Automatic turbine startup to assure service life of CV was achieved using allowable ${\Delta}T$ limit curve.

• 대한기계학회논문집A
• /
• 제26권1호
• /
• pp.1-6
• /
• 2002

The automatic turbine startup system provides turbine control based on thermal stress. During the startup, control system monitors and evaluates main components of turbine using damage mechanism and life assessment. In case of valve chest, the temperature of inner/outer wall is measured by thermo-couples and the safety of these values are evaluated by using allowable △T limit currie during the startup. Because allowable ΔT limit curve includes life assessment, it is possible to apply this curve to turbine control system. In this paper, low cycle fatigue damage, combined rupture and low cycle fatigue damage criterion were proposed for yielding the allowable ΔTf limit curve of CV(control valve) chest. To calculate low cycle fatigue damage, the stress analysis of valve chest has been performed using FEM. Automatic turbine startup to assure service life of CV was achieved using allowable ΔT limit curve.

• Smart Structures and Systems
• /
• 제2권3호
• /
• pp.253-274
• /
• 2006

The structural monitoring of multi-wire strands is of importance to prestressed concrete structures and cable-stayed or suspension bridges. This paper addresses the monitoring of strands by ultrasonic guided waves with emphasis on the signal processing and automatic defect classification. The detection of notch-like defects in the strands is based on the reflections of guided waves that are excited and detected by magnetostrictive ultrasonic transducers. The Discrete Wavelet Transform was used to extract damage-sensitive features from the detected signals and to construct a multi-dimensional Damage Index vector. The Damage Index vector was then fed to an Artificial Neural Network to provide the automatic classification of (a) the size of the notch and (b) the location of the notch from the receiving sensor. Following an optimization study of the network, it was determined that five damage-sensitive features provided the best defect classification performance with an overall success rate of 90.8%. It was thus demonstrated that the wavelet-based multidimensional analysis can provide excellent classification performance for notch-type defects in strands.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.1206-1208
• /
• 2003

The extraction of earthquake damage from remote sensed imagery requires high spatial resolution and temporal effectiveness of acquisition of imagery. The analog photographs and visual interpretation were taken traditionally. Now it is possible to acquire damage information from many commercial high resolution RS satellites. The key techniques are processing velocity and precision. The authors developed the automatic / semiautomatic image process techniques including feature enhancement, and classification, designed the emergency Earthquake Damage and Losses Evaluate System based on Remote Sensing (RSEDLES). The paper introduced the functions of RSEDLES as well as its application to the earthquakes occurred recently.

• 한국기계가공학회지
• /
• 제9권4호
• /
• pp.14-25
• /
• 2010

Small cracks or chips of a product appear very frequently in the course of continuous production of an automatic press process system. These phenomena became the cause of not only defective product but also damage of a press mold. In order to solve this problem AE(Acoustic emission) system was introduced. AE system was expected to be very effective to real time detection of the defective product and for the prevention of the damage in the press molds In this study, for the pick and analysis of AE signals generated from the press process, AE sensors/pre-amplifier/analysis and processing board were used as frequently found in the other similar cases. For the analysis and processing the AE signals picked in real time from the normal or the detective products, specialized software called AE-win(software for processing AE signal from Physical Acoustics Corporation) was used. As a result of this work it was conformed that intensity and shape of the various AE signals differ depending on the weight of the press and thickness of sheet and process type.

• 대한토목학회논문집
• /
• 제28권6A호
• /
• pp.873-880
• /
• 2008

구조물의 열화 또는 손상상태 점검 시 정기적인 외관조사는 매우 중요한 역할을 담당한다. 이때 실시되는 외관조사는 주로 조사자의 스케치나 일반 카메라를 이용한 사진촬영에 의해 이루어지고 있다. 이럴 경우, 조사에 많은 노력과 시간이 소요된다. 따라서 본 연구에서는 보다 효율적이고 효과적인 조사를 위하여, 터널 외관상의 결함을 자동인식 할 수 있는 시스템을 개발하고자 한다. 이 과정에서 결함 손상인식을 위한 화상처리기술과 데이터 관리 및 분석 시스템이 중점적으로 개발 응용되었다. 이를 통해 콘크리트의 균열, 누수, 백화, 박리, 박락, 손상 등의 결함을 인식하여 데이터를 획득하고, 지속적이고 체계적인 데이터 관리시스템을 구축하여, 터널의 안전성 확보, 기능성 유지, 수명연장 등의 효과를 얻고자 한다.

• 한국ITS학회 논문지
• /
• 제6권1호
• /
• pp.27-37
• /
• 2007

과속단속시스템은 지능형교통체계(ITS)의 한 영역인 첨단교통관리시스템(ATMS)으로서 중요한 역할을 수행하고 있다. 그러나 과속사고를 방지하기 위해 설치된 시스템이 시거가 확보되지 않은 구간이나 급커브 지역 등에 설치됨으로써 추가적인 교통사고 위험을 내포하고 있다. 이에 본 연구에서는 고속도로의 기하구조별로 시스템 설치위치를 구분하고, 각 위치에서 발생한 교통사고 자료를 시스템 설치 전 후로 나누어 기하구조별 과속단속시스템의 설치 전 후의 사고건수(Frequency), 대물피해환산법(EPDO, Equivalent Property Damage Only)을 통한 사고심각도와 사고비용(Accident Cost)을 분석하였다. 이와 같은 연구수행 절차를 통해 시거가 충분히 확보되지 않는 구간에서의 과속단속시스템의 설치가 사고에 미치는 영향을 알아보고 최종적으로 과속단속시스템의 설치위치에 따른 영향을 분석하고자 한다.

• 한국자동차공학회논문집
• /
• 제21권2호
• /
• pp.63-71
• /
• 2013

The estimation of occupant injury risk at crash accident is one of the most important assessments for the vehicle crashworthiness performance. The design of safety devices such as occupant restraining system also depend on the kinematics of occupant and its injury risk. The real world in-depth accident investigation provides detailed and realistic information of vehicle damage and occupant injury as well as the accident conditions. This paper introduces a statistical analysis of NASS/CDS database and domestic accident data to correlate speed change, vehicle damage extend, and occupant injury at frontal crash. The maximum crush extend shows a linear relationship with the effective impact speed. The injury risks of the occupant with and without restraining were also respectively quantified with the crush extend. This result can be effectively used for the emergent rescue of crash victims with automatic crash notification system.