• Smart Structures and Systems
• /
• v.24 no.5
• /
• pp.567-585
• /
• 2019

In the past two decades, structural health monitoring (SHM) systems have been widely installed on various civil infrastructures for the tracking of the state of their structural health and the detection of structural damage or abnormality, through long-term monitoring of environmental conditions as well as structural loadings and responses. In an SHM system, there are plenty of sensors to acquire a huge number of monitoring data, which can factually reflect the in-service condition of the target structure. In order to bridge the gap between SHM and structural maintenance and management (SMM), it is necessary to employ advanced data processing methods to convert the original multi-source heterogeneous field monitoring data into different types of specific physical indicators in order to make effective decisions regarding inspection, maintenance and management. Conventional approaches to data analysis are confronted with challenges from environmental noise, the volume of measurement data, the complexity of computation, etc., and they severely constrain the pervasive application of SHM technology. In recent years, with the rapid progress of computing hardware and image acquisition equipment, the deep learning-based data processing approach offers a new channel for excavating the massive data from an SHM system, towards autonomous, accurate and robust processing of the monitoring data. Many researchers from the SHM community have made efforts to explore the applications of deep learning-based approaches for structural damage detection and structural condition assessment. This paper gives a review on the deep learning-based SHM of civil infrastructures with the main content, including a brief summary of the history of the development of deep learning, the applications of deep learning-based data processing approaches in the SHM of many kinds of civil infrastructures, and the key challenges and future trends of the strategy of deep learning-based SHM.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.876-879
• /
• 2009

As performance and feature set of mobile devices are getting better, the scope of things for which they can be used is also getting widened. More than simply browsing information using mobile web, now these devices can interact with users. It requires various features to manage a web site; supporting full-browsing alone is not enough since there are problems such as processing speed, user interface restriction. However, using mobile devices to manage a web site has its own merits since the can be used without restriction of time and location. In this paper, we introduce a mobile web site administration software that provides both online and offline features. It supports CRUD (Create, Read, Update, Delete) operations for news and users. These operations can be executed both offline and online. A user can do most of things offline; It connects to online only when it's absolutely necessary and does those in batch mode.

• Geomechanics and Engineering
• /
• v.11 no.1
• /
• pp.77-94
• /
• 2016

A forked tunnel, as a special complicated underground structure, is composed of big-arch tunnel, multi-arch tunnel, neighborhood tunnels and separate tunnels according to the different distances between two separate tunnels. Due to the complicated process of design and construction, surrounding jointed rock mass stability of the big-arch tunnel which belongs to the forked tunnel during excavation is a hot issue that needs special attentions. In this paper, an elasto-plastic damage constitutive model for jointed rock mass is proposed based on the coupling method considering elasto-plastic and damage theories, and the irreversible thermodynamics theory. Based on this elasto-plastic damage constitutive model, a three dimensional elasto-plastic damage finite element code (D-FEM) is implemented using Visual Fortran language, which can numerically simulate the whole excavation process of underground project and perform the structural stability of the surrounding rock mass. Comparing with a popular commercial computer code, three dimensional fast Lagrangian analysis of continua (FLAC3D), this D-FEM has advantages in terms of rapid computing process, element grouping function and providing more material models. After that, FLAC3D and D-FEM are simultaneously used to perform the structural stability analysis of the surrounding rock mass in the forked tunnel considering three different computing schemes. The final numerical results behave almost consistent using both FLAC3D and D-FEM. But from the point of numerically obtained damage softening areas, the numerical results obtained by D-FEM more closely approach the practical behaviors of in-situ surrounding rock mass.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.4
• /
• pp.72-79
• /
• 2015

A number of plating companies have been exposed to the risk of fire due to unexpected temperature increasing of water in a plating bath. Since the companies are not able to forecast the unexpected temperature increasing of water and most of raw materials in the plating process have low ignition temperature, it is easy to be exposed to the risk of fire. Thus, the companies have to notice the changes immediately to prevent the risk of fire from plating process. Due to this reason, an agile and systematic temperature monitoring system is required for the plating companies. Unfortunately, in case of small size companies, it is hard to purchase a systematic solution and be offered consulting from one of the risk management consulting companies due to an expensive cost. In addition, most of the companies have insufficient research and development (R&D) experts to autonomously develop the risk management solution. In this article, we developed a real time remote temperature monitoring system which is easy to operate with a lower cost. The system is constructed by using Raspberry Pi single board computer and Android application to release an economic issue for the small sized plating manufacturing companies. The derived system is able to monitor the temperature continuously with tracking the temperature in the batch in a short time and transmit a push-alarm to a target-device located in a remoted area when the temperature exceeds a certain hazardous-temperature level. Therefore, the target small plating company achieves a risk management system with a small cost.

• Journal of the Society of Naval Architects of Korea
• /
• v.32 no.4
• /
• pp.38-47
• /
• 1995

The three-dimensional multi-layer computer simulation technology for tidal current developed in the previous study is updated to a new version. many improvements are achieved by following changes : (1) No-reflection condition is adopted instead of no-gradient condition as an open boundary condition. (2) Time marching algorithm is changed so that velocity and pressure(surface movement) might be salved in turn at different time step (3) Convection term in equation of motion is estimated by upwind differencing scheme instead of central differencing. The stability is improved considerably and the steady state is achieved within 2 tidal periods which is about 3 times shorter than that of the old version. Moreover, fluctuations in time disappeared by introducing the new time marching technique. An application to the real near shore area(near Inchon harbor) is performed by the new version. Simulated results are compared with those by the simulation total developed in the University of Tokyo. Validity and effectiveness of the two simulation technologies are chocked through the comparative research works.

• KSCI Review
• /
• v.14 no.1
• /
• pp.241-248
• /
• 2006

The recognition of a relatively big and rarely movable object. such as refrigerator and air conditioner, etc. is necessary because these objects can be crucial global stable features of Simultaneous Localization and Map building(SLAM) in the indoor environment. In this paper. we propose a novel method to recognize these big objects using a sequence of 3D scenes. The particles representing an object to be recognized are scattered to the environment and then the probability of each particles is calculated by the matching test with 3D lines of the environment. Based on the probability and degree of convergence of particles, we can recognize the object in the environment and the pose of object is also estimated. The experimental results show the feasibility of incremental object recognition based on particle filtering and the application to SLAM

• Nuclear Engineering and Technology
• /
• v.18 no.2
• /
• pp.78-84
• /
• 1986

An equation of material number density sensitivity coefficient is derived using first-order perturbation theory. The beginning of cycle of Super-Phenix I is taken as the reference system for this study. Effective multiplication factor of the reference system is defined as system response function and fuel enrichment and fuel effective density are chosen for the variation of reference input data since they are described by material number density which is a component of Boltzmann operator. The nuclear computational code system (KAERI-26 group cross section library/1DX/2DB/PERT-V) is employed for this calculation. Sensitivity coefficient of fuel enrichment on effective multiplication factor is 4.576 and sensitivity coefficient of effective fuel density on effective multiplication factor is 0.0756. This work shows that sensitivity methodology is lesser timeconsuming and gives more informations on important design parameters in comparison with the direct iterative calulation through large computer codes.

• Journal of the Optical Society of Korea
• /
• v.17 no.4
• /
• pp.344-349
• /
• 2013

This paper proposes a simulation algorithm to assess the gradual yellowing vision of the elderly, which refers to the predominance of yellowness in their vision due to aging of the ocular optic media. This algorithm employed the spectral transmittance property of a yellow filter to represent the color appearance perceived by elderly people with yellow vision, and modeled the changes in the color space through a spectrum change in light using the yellow filter effect. The spectral reflectivity data of 1269 Munsell matte color chips were used as reference data. Under the standard conditions of a D65 illuminant and a $10^{\circ}$ observer of 1964 CIE, the spectrum of the 1269 Munsell colors were processed through the yellow filter effect to simulate yellow vision. Various degrees of yellow vision were modeled according to the transmittance percentage of the yellow filter. The color differences before and after the yellow filter effect were calculated using the DE2000 formula, and the color pairs were selected based on the color difference function. These color pairs are distinguishable through normal vision, but the color difference diminishes as the degree of yellow vision increases. Assuming 80% of yellow vision effect, 17 color pairs out of $(1269{\times}1268)/2$ pairs were selected, and for the 90% of yellow vision effect, only 3 color pairs were selected. The result of this study can be utilized for the diagnosis system of gradual yellow vision, making various types of test charts with selected color pairs.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.5
• /
• pp.47-52
• /
• 2021

Recently, demand for artificial intelligence solutions for production process management has been increasing in the manufacturing industry. However, through the application of AI solutions in the manufacturing industry, there are limitations to legacy smart factory solutions such as POP and MES.Therefore, in order to overcome this, this paper aims to improve production management efficiency by applying guidance, an artificial intelligence concept, to image recognition systems. In the system flow, As_is To be separated and actual work flow was applied, and the process was improved for overall productivity efficiency. The pre-processing plan for AI guidance learning was established and the relevant AI model was designed, developed, and simulated, resulting in a 97% recognition rate.

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.479-484
• /
• 2021

This paper proposed and designed a structure to achieve high performance with a small number of cores in GPGPU with SIMT structure. GPGPU for application to mobile devices requires a structure to increase performance compared to power consumption. In order to reduce power consumption, the number of cores decreased, but to improve performance, the size of the warp scheduler for managing threads was set to 4, which was greatly reduced than 32 of general GPGPU. Reducing warp size can reduce the number of idle cycles in pipelines and efficiently apply memory latency to reduce miss penalty when accessing cache memory. The designed GPGPU measured computational performance using a test program that includes floating point operations and measured power consumption through a 28nm CMOS process to obtain 104.5GFlops/Watt as a performance per power. The results of this paper showed about four times better performance per power compared to Tegra K1 of Nvidia