• 대한조선학회논문집
• /
• 제59권4호
• /
• pp.192-199
• /
• 2022

With the advent of autonomous ships, it is emerging as one of the very important issues not only to operate with a minimum crew or unmanned ships, but also to secure the safety of ships to prevent marine accidents. On-site inspection of the hull is mainly performed by the inspector's visual inspection, and video information is recorded using a small camera if necessary. However, due to the shortage of inspection personnel, time and space constraints, and the pandemic situation, the necessity of introducing an automated inspection system using artificial intelligence and remote inspection is becoming more important. Furthermore, research on hardware and software that enables the automated inspection system to operate normally even under the harsh environmental conditions of a ship is absolutely necessary. For automated inspection systems, it is important to review artificial intelligence technologies and equipment that can perform a variety of hull failure detection and classification. To address this, it is important to classify the hull failure. Based on various guidelines and expert opinions, we divided them into 6 types(Crack, Corrosion, Pitting, Deformation, Indent, Others). It was decided to apply object detection technology to cracks of hull failure. After that, YOLOv5 was decided as an artificial intelligence model suitable for survey and a common hull crack dataset was trained. Based on the performance results, it aims to present the possibility of applying artificial intelligence in the field by determining and testing the equipment required for survey.

• 문화기술의 융합
• /
• 제10권2호
• /
• pp.511-516
• /
• 2024

최근 제정된 궤도시설의 성능평가에 관한 세부지침에서 궤도성능평가의 평가절차 및 실시방법 등에 관한 필요사항을 제시하였다. 그러나 외관조사(육안조사)에 의해 레일표면손상의 등급이 결정되며, 점검자의 주관적인 판단으로 정성적인 평가에만 의존할 수밖에 없는 실정이다. 따라서 본 연구에서는 레일표면손상을 이용하여 레일내부결함까지 진단할 수 있는 진단애플리케이션을 개발하고자 하였다. 현장조사에서는 레일표면손상을 조사하고 패턴을 분석하였다. 또한 실내시험에서는 레일내부손상 이미지 데이터를 구축하기 위하여 SEM 시험을 이용하였으며, 균열 길이, 깊이 및 각도를 정량화하였다. 본 연구에서는 현장조사와 실내시험에서 구축한 이미지 데이터를 적용한 딥러닝 모델(Fast R-CNN)을 애플리케이션에 적용하였다, 스마트기기에서 사용이 가능한 딥러닝 모델을 이용한 레일표면손상 진단 애플리케이션(App)을 개발하여 향후 궤도진단 및 성능평가 업무에 활용 가능한 레일표면손상 스마트 진단시스템을 개발하였다.

• 한국광학회지
• /
• 제29권1호
• /
• pp.19-27
• /
• 2018

일반적으로 광학계의 물체거리가 변하면 배율이 변하게 된다. 본 논문에서는 일반적인 이중 가우스(double-Gauss) 형태의 광학계에서 조리개를 기준으로 조리개 앞쪽에 위치한 렌즈군과 조리개 뒷쪽 렌즈군을 광축 방향으로 독립적으로 평행하게 이동하여 물체거리에 따라 배율과 상면이 고정되는 광학계를 제안하고 설계하였다. 이러한 광학계는 전방시현장치(head-up display, HUD), 두부장착디스플레이(head-mounted display, HMD) 등의 투사 광학계에 물체거리의 변화에 따라 상 크기가 변화하지 않도록 하여 전방시현장치 또는 두부장착디스플레이에서 초점 조절(focusing) 시에 화각이 변하지 않도록 하였다. 또한 반도체 칩과 IC 회로기판을 연결하는 와이어(wire)의 상태를 검사하는 과정에서 검사장비가 위 아래로 움직여서 물체거리가 변해도 광학계의 배율이 변하지 않도록 하여 고속검사가 가능할 수 있도록 별도 영상 처리를 시스템적으로 생략할 수 있었다. 본 논문에서 가우스 괄호법(Gaussian bracket method)을 이용하여 원하는 사양을 만족하도록 각 군의 이동량을 구해서 배율과 상면이 고정되도록 하였다. 초기 설계를 진행한 후, 최적화는 광학 설계 프로그램인 시놉시스(Synopsys)를 사용하였다.

• Smart Structures and Systems
• /
• 제12권1호
• /
• pp.55-71
• /
• 2013

Structural health monitoring (SHM) is vital for detecting the onset of damage and for preventing catastrophic failure of civil infrastructure systems. In particular, piezoelectric transducers have the ability to excite and actively interrogate structures (e.g., using surface waves) while measuring their response for sensing and damage detection. In fact, piezoelectric transducers such as lead zirconate titanate (PZT) and poly(vinylidene fluoride) (PVDF) have been used for various laboratory/field tests and possess significant advantages as compared to visual inspection and vibration-based methods, to name a few. However, PZTs are inherently brittle, and PVDF films do not possess high piezoelectricity, thereby limiting each of these devices to certain specific applications. The objective of this study is to design, characterize, and validate piezoelectric nanocomposites consisting of zinc oxide (ZnO) nanoparticles assembled in a PVDF copolymer matrix for sensing and SHM applications. These films provide greater mechanical flexibility as compared to PZTs, yet possess enhanced piezoelectricity as compared to pristine PVDF copolymers. This study started with spin coating dispersed ZnO- and PVDF-TrFE-based solutions to fabricate the piezoelectric nanocomposites. The concentration of ZnO nanoparticles was varied from 0 to 20 wt.% (in 5 % increments) to determine their influence on bulk film piezoelectricity. Second, their electric polarization responses were obtained for quantifying thin film remnant polarization, which is directly correlated to piezoelectricity. Based on these results, the films were poled (at 50 $MV-m^{-1}$) to permanently align their electrical domains and to enhance their bulk film piezoelectricity. Then, a series of hammer impact tests were conducted, and the voltage generated by poled ZnO-based thin films was compared to commercially poled PVDF copolymer thin films. The hammer impact tests showed comparable results between the prototype and commercial samples, and increasing ZnO content provided enhanced piezoelectric performance. Lastly, the films were further validated for sensing using different energy levels of hammer impact, different distances between the impact locations and the film electrodes, and cantilever free vibration testing for dynamic strain sensing.

• Structural Monitoring and Maintenance
• /
• 제2권1호
• /
• pp.19-34
• /
• 2015

One of the main causes of a limited use of nondestructive evaluation (NDE) technologies in bridge deck assessment is the speed of data collection and analysis. The paper describes development and implementation of the RABIT (Robotics Assisted Bridge Inspection Tool) for data collection using multiple NDE technologies. The system is designed to characterize three most common deterioration types in concrete bridge decks: rebar corrosion, delamination, and concrete degradation. It implements four NDE technologies: electrical resistivity (ER), impact echo (IE), ground-penetrating radar (GPR), and ultrasonic surface waves (USW) method. The technologies are used in a complementary way to enhance the interpretation. In addition, the system utilizes advanced vision to complement traditional visual inspection. Finally, the RABIT collects data at a significantly higher speed than it is done using traditional NDE equipment. The robotic system is complemented by an advanced data interpretation. The associated platform for the enhanced interpretation of condition assessment in concrete bridge decks utilizes data integration, fusion, and deterioration and defect visualization. This paper concentrates on the validation and field implementation of two NDE technologies. The first one is IE used in the delamination detection and characterization, while the second one is the USW method used in the assessment of concrete quality. The validation of performance of the two methods was conducted on a 9 m long and 3.6 m wide fabricated bridge structure with numerous artificial defects embedded in the deck.

• Structural Engineering and Mechanics
• /
• 제54권2호
• /
• pp.257-289
• /
• 2015

Shear connectors are generally used to link the slab and girders together in slab-on-girder bridge structures. Damage of shear connectors in such structures will result in shear slippage between the slab and girders, which significantly reduces the load-carrying capacity of the bridge. Because shear connectors are buried inside the structure, routine visual inspection is not able to detect conditions of shear connectors. A few methods have been proposed in the literature to detect the condition of shear connectors based on vibration measurements. This paper proposes a different dynamic condition assessment approach to identify the damage of shear connectors in slab-on-girder bridge structures based on power spectral density transmissibility (PSDT). PSDT formulates the relationship between the auto-spectral densities of two responses in the frequency domain. It can be used to identify shear connector conditions with or without reference data of the undamaged structure (or the baseline). Measured impact force and acceleration responses from hammer tests are analyzed to obtain the frequency response functions at sensor locations by experimental modal analysis. PSDT from the slab response to the girder response is derived with the obtained frequency response functions. PSDT vectors in the undamaged and damaged states can be compared to identify the damage of shear connectors. When the baseline is not available, as in most practical cases, PSDT vectors from the measured response at a reference sensor to those of the slab and girder in the damaged state can be used to detect the damage of shear connectors. Numerical and experimental studies on a concrete slab supported by two steel girders are conducted to investigate the accuracy and efficiency of the proposed approach. Identification results demonstrate that damages of shear connectors are identified accurately and efficiently with and without the baseline. The proposed method is also used to evaluate the conditions of shear connectors in a real composite bridge with in-field testing data.

• Journal of Biosystems Engineering
• /
• 제40권1호
• /
• pp.19-27
• /
• 2015

Purpose: A system to measure the visibility of agricultural tractor operators was designed and evaluated according to ISO standards, and a blind area diagram around the tested tractor was created based on the manual method recommended by the National Institute for Occupational Safety and Health (NIOSH). Methods: A visibility measurement system was designed and evaluated based on the ISO 5006 and ISO 5721-1 standards. Two bulbs used to simulate the operator's eyes were mounted on a bar with a supporting frame. A wooden frame was used to determine the seat index point position. The 12-m visibility test circle was divided into six sectors of vision, and the test tractor was placed at the center of the circle. Artificial light was supplied in the darkened environment, and shadow or masking effects were measured manually around the 12-m circle. Results: When the bulbs were placed at the operator's eye level, front visibility was good; no masking was found in the "A" vision sector, but larger masking widths were found in the "B" and "C" vision sectors. Since the masking width exceeded 700 mm, additional tests, such as movement of the light sources to both sides of the operator's eye level, were performed. Less than six masking effects were found in the semi-circle of vision to the front, and more than one masking was found in the "B" and "C" visual fields. The minimum distance between the centers of two masking effects exceeded 2500 mm when measured as a chord on the semi-circle of vision. A blind area diagram was created to define the exact nature of the blind spots and mirror visibility. Conclusions: Visibility evaluation is an effective way to enable proper and safe operation for agricultural tractor operators. Inclusion of this visibility evaluation test in the general testing process might aid tractor manufacturers.

• 한국산학기술학회논문지
• /
• 제15권10호
• /
• pp.6282-6288
• /
• 2014

본 논문은 용접 산업현장에서 많이 사용 되는 $CO_2$ 용접에서는 차량내부 및 탱크내부 용접에서 제어 패널을 볼 수 없는 곳, 작업장이 먼 곳 등에서 용접사가 용접을 하면서 용접조건에 맞는 전류와 전압을 조절할 수가 없고, 전류와 전압을 조절하기 위해 용접을 중단하고 잦은 이동으로 용접 구조물의 순간적인 냉각에 의해 용접결함이 발생된다. 본 연구에서는 기존의 $CO_2$ 용접기 3종류를 각각 사용하여 원격제어 토치의 용접에 대해 SS400 용접구조용 압연강재를 사용하여 원격제어 토치와 기존 $CO_2$ 용접 토치를 V형 맞대기 수직자세로 용접실험을 실시하고 용접 부의 표면비드 상태의 형상을 육안검사 관찰하고 또한 이를 침투탐상검사 및 굽힘 시험을 통해 용접부의 외관품질에 대하여 중점적으로 수행하여 용접결함 감소 및 기존 상용용접기에 교체사용에 대한 성능 및 호환성여부에 미치는 영향에 대해 평가하였다.

• 한국물환경학회지
• /
• 제37권4호
• /
• pp.275-285
• /
• 2021

Algal bloom is an important issue in maintaining the safety of the drinking water supply system. Fast detection and classification of algae images are essential for the management of algal blooms. Conventional visual identification using a microscope is a labor-intensive and time-consuming method that often requires several hours to several days in order to obtain analysis results from field water samples. In recent decades, various deep learning algorithms have been developed and widely used in object detection studies. YOLO is a state-of-the-art deep learning algorithm. In this study the third version of the YOLO algorithm, namely, YOLOv3, was used to develop an algae image detection model. YOLOv3 is one of the most representative one-stage object detection algorithms with faster inference time, which is an important benefit of YOLO. A total of 1,114 algae images for 30 genera collected by microscope were used to develop the YOLOv3 algae image detection model. The algae images were divided into four groups with five, 10, 20, and 30 genera for training and testing the model. The mean average precision (mAP) was 81, 70, 52, and 41 for data sets with five, 10, 20, and 30 genera, respectively. The precision was higher than 0.8 for all four image groups. These results show the practical applicability of the deep learning algorithm, YOLOv3, for algae image detection.

• 한국산업보건학회지
• /
• 제29권2호
• /
• pp.217-225
• /
• 2019

Objectives: The aim of this study is to evaluate the quality management systems of domestic asbestos survey and monitoring service providers and the relationships with the number of licenses or designations and sales performances. Methods: Data on quality management systems were collected by assessors who were assigned by the Korea Occupational Safety and Health Agency(KOSHA) during a pilot evaluation program for designated asbestos survey and monitoring service providers in 2016 using evaluation criteria developed by KOSHA. Basic characteristics, evaluated scores, and sales performance were gathered and statistically analyzed. Results: The median and arithmetic mean of the total scores were 0.64 and 0.66. Evaluation fields that scored highly with the highest percentages were sales performance, installation and availability of equipment, compliance with the mandatory minimum number of airborne samples, laboratory independence, and results of proficiency analytical testing, in that order. Evaluation fields that received low marks with the highest percentages were the training of personnel, blank field samples, calibration of flow rates, preliminary check and visual inspection of the work area prior to the clearance test, and review and approval of final reports, in that order. Comparison of normalized scores between service providers registered for asbestos and other tasks and those designated for only asbestos showed significant differences in their evaluated scores. Sales performance did not show a positive correlation with evaluated scores. Conclusions: The quality management systems of domestic asbestos survey and monitoring service providers were poor. High scores were recorded mostly in evaluation fields related to regulatory requirements. Low scores were recorded mostly in evaluation fields related to documentation and recordkeeping. Considering the low influence of quality on sales performance, the government needs to evaluate the quality management of asbestos survey and monitoring service providers and provide the results to public in order to address their low levels of quality management.