http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
센서 데이터 변곡점에 따른 Time Segmentation 기반 항공기 엔진의 고장 패턴 추출
백수정(Sujeong Baek) 한국산업경영시스템학회 2021 한국산업경영시스템학회지 Vol.44 No.3
As mechatronic systems have various, complex functions and require high performance, automatic fault detection is necessary for secure operation in manufacturing processes. For conducting automatic and real-time fault detection in modern mechatronic systems, multiple sensor signals are collected by internet of things technologies. Since traditional statistical control charts or machine learning approaches show significant results with unified and solid density models under normal operating states but they have limitations with scattered signal models under normal states, many pattern extraction and matching approaches have been paid attention. Signal discretization-based pattern extraction methods are one of popular signal analyses, which reduce the size of the given datasets as much as possible as well as highlight significant and inherent signal behaviors. Since general pattern extraction methods are usually conducted with a fixed size of time segmentation, they can easily cut off significant behaviors, and consequently the performance of the extracted fault patterns will be reduced. In this regard, adjustable time segmentation is proposed to extract much meaningful fault patterns in multiple sensor signals. By considering inflection points of signals, we determine the optimal cut-points of time segments in each sensor signal. In addition, to clarify the inflection points, we apply Savitzky-golay filter to the original datasets. To validate and verify the performance of the proposed segmentation, the dataset collected from an aircraft engine (provided by NASA prognostics center) is used to fault pattern extraction. As a result, the proposed adjustable time segmentation shows better performance in fault pattern extraction.
CNN 기반의 상자 표면 편차 이미지 분석을 통한 상자 포장공정에서의 실시간 진공 그리퍼 파지 작업 모니터링
백수정(Sujeong Baek),전승호(Seung Ho Jeon),송의찬(Eui Chan Song) 대한산업공학회 2020 대한산업공학회지 Vol.46 No.2
Due to smart factories that can produce various types of products quickly, the ability to pick up various objects within a single manufacturing process has become important. Accordingly, many general-purpose vacuum grippers have been used, however, the performance of the pick-up operation depends on the degree of uniformness in a contact surface of a target object. Therefore, we aim to prevent abnormal conditions in the manufacturing process by real-time inspecting a surface of a target product before a pick-up/gripping process is performed. In particular, in order not to increase the manufacturing cost, the surface deviation data of a product was collected using an ultrasonic sensor, and a quality inspection was performed in real time by applying a convolution neural network to the collected data. The performance of the proposed method is verified through pick-up operation using the 6-axis robot with a vacuum gripper in the box packaging process.
최동희(Donghee Choi),백수정(Sujeong Baek),이송일(Songil Lee),경규형(Gyouhyung Kyung) 한국HCI학회 2013 한국HCI학회 학술대회 Vol.2013 No.1
컴퓨터 마우스는 다양한 형태의 컴퓨터와의 상호작용을 위해 키보드와 함께 가장 일반적으로 사용되고 있는 입력장치다. 상당수의 VDU (Visual display unit) 작업자들은 컴퓨터 작업 시 마우스의 반복적, 장시간 사용으로 인하여 수근관 증후근과 같은 근골격계질환의 위험에 노출 되어 있다. 본 연구의 목적은 손목과 그 주변의 불편도를 감소시키기 위한 마우스 개발이다. 첫 번째 실험을 통해 마우스 사용시의 손바닥 압력과 주관적 불편도의 관계를 통해 인체공학적 마우스 설계를 위한 주요 인자들을 정하였다. 이 실험을 통하여 손바닥 압력과 주관적 불편도 사이에는 양의 상관관계 (Adjusted R2=0.624)가 있으며, 주요인자는 손목 측면 기울기 각임을 발견하였다. 두 번째 실험에서는 일정수준의 사용성을 유지할 수 있는 인체공학적인 마우스 설계를 위해 찾아낸 인자의 수준을 결정하는 실험을 수행했다. 실험결과, 측면 각도조절이 가능한 마우스의 경우, 일정 수준 이상의 편안함을 제공함을 발견하였다. 손목 내외전 자세에 다양성을 주는 마우스를 사용할 경우 근골격계질환의 예방에 도움을 줄 것으로 예상된다. The computer mouse is one of the most commonly used input devices to interact with a computer. There are a significant number of VDU (visual display unit) workers exposed to the potential musculoskeletal disorders (MSDs) such as the carpal tunnel syndrome due to repetitive and continuous mouse usage. First, the current study investigated major factors for designing an ergonomic mouse by investigating the relationship between pressure and perceived discomfort at the right hand. In the first experiment, a positive correlation between these two (Adjusted R2=0.624) was found, and wrist pronation and supination was determined as a major factor for an ergonomics mouse. In the second experiment, a laterally tiltable mouse appeared to facilitate wrist supination and pronation while providing comparable usability. Hence, a laterally tiltable mouse is expected to prevent MSDs around the wrist.
자동화 공정의 에러 상태 감지를 위한 영상처리 기반 디지털 제어 신호의 데이터 전처리 알고리즘 개발
송용욱(Yong-Uk Song),백수정(Sujeong Baek) 대한기계학회 2020 대한기계학회 춘추학술대회 Vol.2020 No.12
Since many data-driven fault detection studies have been done, data pre-processing becomes important. However, it is difficult to determine healthy states and to obtain additional information on the healthy states in automatic manufacturing systems operated only with digital control signals. In this paper, we develop a pre-processing algorithm which detects system’s healthy states in real-time and collects datasets during healthy states of a manufacturing system automatically. Image data is obtained by a webcam, and the proposed algorithm detects the process movement by analyzing the image. When such operation movement is identified, we export the raw dataset to a pre-processing database and we additionally determine the current operation as healthy or not. This method was applied to a smart factory testbed and the performance is verified. As a result, we expect that the proposed pre-processing study will improve the accuracy of process error detection in automatic manufacturing systems using CCTV.