Sang Woo Kim
포항공과대학교 기계공학과 · 공학
Sang Woo Kim 교수의 연구실은 첨단 센서 기반 이상 진단 및 고도화된 머신러닝 기법을 융합한 스마트 제조 및 에너지 시스템 진단 기술을 주요 연구 분야로 삼고 있습니다. 특히 강판 표면의 결함 검출, 리튬이온 배터리의 동반 고장 진단, 제어 케이블의 소프트 결함 탐지 등 실생활 산업 문제 해결을 위한 혁신적 알고리즘 개발에 집중하고 있습니다. 연구는 비정상 상태 탐지, 자동 특징 추출, 계산 효율성 향상 기반의 실시간 진단 기술을 핵심으로 하며, 딥러닝과 제어 이론을 융합한 지능형 진단 시스템을 구축하고 있습니다.
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
In this paper a novel filtering scheme combined with a lighting method is proposed for defect detection in steel surfaces. A steel surface has non-uniform brightness and various shaped defects, which cause difficulties in defect detection. To solve this problem we propose a sub-optimal filtering that is combined with a switching-lighting method. First, dual-light switching lighting (DLSL) is explained, which decreases the effect of non-uniformity of surface brightness and improves the detection
Most of the prototype reduction schemes (PRS), which have been reported in the literature, process the data in its entirety to yield a subset of prototypes that are useful in nearest-neighbor-like classification. Foremost among these are the prototypes for nearest neighbor classifiers, the vector quantization technique, and the support vector machines. These methods suffer from a major disadvantage, namely, that of the excessive computational burden encountered by processing all the data. In thi
Accurate health diagnostics of lithium-ion batteries are indispensable for efficient utilization. A decrease in battery capacity not only diminishes the energy efficiency but also causes several detrimental effects , such as an internal short circuit (ISC) fault; these fault can lead to thermal runaway. However, the simultaneous impact of aging and ISC faults complicates the ability to distinguish between two factors within a singular discharging or charging process. This study focuses on the co
Deep neural networks have been used in various fields, but their internal behavior in how they understand images is not well known. In this study, we discuss two counterintuitive properties of convolutional neural networks (CNNs). First, we evaluated the size of the receptive field of CNNs with their classification accuracy. Previous studies have attempted to increase the size of the receptive field for performance gain. However, we observed that some CNNs with a smaller receptive field can achi
Two lower bounds for the trace of the solution of the discrete algebraic Riccati equation (DARE) are presented. It is shown that in many cases, these trace bounds are tighter than those in the literature and greater than the trace of the state weighting matrix even when the system matrix is singular. The results are illustrated by an example.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
