• Smart Structures and Systems
• /
• 제30권6호
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• 대한화학회지
• /
• 제48권1호
• /
• pp.94-98
• /
• 2004

• 대한기계학회논문집A
• /
• 제31권12호
• /
• pp.1208-1213
• /
• 2007

Some structural problems and temperature difference of the pyrolysis equipment were improved by using the structure analysis and the optimal design of torch. The pyrolysis equipment developed in this study is expected to the excellent pyrolysis effect. To modify user-friendly the dimension of a part, we developed the feature modeling system that all of the related parts automatically change applying to the three-dimensional modeling method.

• Journal of Photoscience
• /
• 제6권1호
• /
• pp.29-36
• /
• 1999

Chlorophasts are a central structural feature of stomatal guard cells. Guard cell chloroplasts have both photosystems I and II (PS I and II), carry out O2 evoluation , cyclic and noncyclic photophosporylation, and possess the Calvin-Benson cycle enzymes involved in CO2 fixation. These imply that guard cell chloroplasts have a normal photosynthetic carbon reduction pathway just like their mesophyll counterparts, indicating similar fuctional organization of thylakoid membranes in both types of mesophyll and guard cell chloroplasts. It has been, however, found that guard cell chloroplasts have distinctive and comparative properties in their photosynthetic performance. In this article, I review the intrinsic features on the light reaction of and carbon reduction by guard cell chloroplasts.

• 한국진공학회지
• /
• 제6권S1호
• /
• pp.66-69
• /
• 1997

Porosity is one important characteristic feature and structural index of sprayed coatings. A method of measurement of porosity, EPMA-EDS image processing is developed in the paper. The characteristics of pores can be determined by processing of the image obtained from an electron microscope via VISTA, Not only the porosity can be presented but also the statistical result of pore size distribution. Finally it can be drawn from this paper that EPMA-EDS is a quite effective method to completely characterize the pores in plasma sprayed coatings.

• 기술사
• /
• 제34권2호
• /
• pp.14-18
• /
• 2001

The primary cooling piping of HANARO is classified as safety class 3, seismic class 1 and quality class Q. This piping as safety related feature has been designed, manufactured and tested in accordance with ASME SEC. Ⅲ, DIV 1, Class 3. In October of 2000, the first step of the in-service inspection for this piping was carried out in accordance with ASME SEC. XI. This describes the results of the Inspection including the preparation of inservice inspection plan and inspection method. It is verified through the results that the safety related piping is maintained the mechanical and structural Integrities.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.138.2-138
• /
• 2001

This paper proposes the technique for recognizing the human´s emotion by using the CCD color image. To do this, we first acquire the color image from the CCD camera. And then propose the method for recognizing the expressing to be represented the structural correlation of man´s feature points(eyebrows, eye, nose, mouse), In the proposed method. Human´s emotion is divided into four emotion(surprise, anger, happiness, sadness). Finally, we have proven the effectiveness of the proposed method through the experimentation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.476-481
• /
• 1991

This paper investigates the relationship between the two problems, supervisor reduction and observation function (projection) design, which arise in supervisory control of DEDS. It is shown through an example that a reduced supervisor of minimal size does not necessarily result in a maximal projection when a projection design method which uses the transition structure of a supervisor is applied. Also, if an L-realizable projection P is available and if a supervisor has a special structural feature, a cover C for supervisor reduction can be easily obtained. By investigating the control-compatibility of states of the reduced supervisor based on C, we can also check maximality of P in a simple manner.

• 대한전자공학회논문지
• /
• 제27권4호
• /
• pp.590-597
• /
• 1990

To represent shape characteristics of digital closed curve, many algorithms, mainly based on local properties, have been proposed. In this paper, we propose a new algorithm for detecting local curvature maxima which reflects context, i.e., structural or surrounding regional characteristics. The algorithm does not require the value of k as an input parameter which is the major problem in k-curvature method in digital curve, but calculates it at each point depening on the context. The algorithm has been applied to two dimensional image boundaries. The efficiency of the algorithm is addressed by comparing the result of existing contest dependent algorithm.

• Structural Engineering and Mechanics
• /
• 제15권4호
• /
• pp.451-462
• /
• 2003

In this paper, a new block iterative algorithm is presented by using the special feature of the continuous Riccati equation in the optimal shape control. Because the real-time control require that the CPU time should be as short as possible, an appropriate modal control algorithm is sought. The computing cost is less than the one of the all state feedback control. A numerical example is given to illustrate the algorithm.