• Journal of Wetlands Research
• /
• v.23 no.1
• /
• pp.74-84
• /
• 2021

Active stormwater management is essential to minimize the impact of urban development and improve the hydrological cycle system. In recent years, the Low Impact Development (LID) technique for urban stormwater management is attracting attention as a reasonable alternative. The Storm Water Management Model (SWMM) is actively used in urban hydrological cycle improvement projects as it provides simulation functions for various GI (Green Infra) facilities through its LID module. However, in order to simulate GI facilities using SWMM, there are many difficulties in setting up complex watersheds and deploying GI facilities. In this study, a model that can evaluate the performance of GI facilities is proposed while implementing the core hydrological process of GI facilities. Since the proposed model operates based on hydrological routing, it can not only reflect the infiltration, storage, and evapotranspiration of GI facilities, but also quantitatively evaluate the effect of improving urban hydrological cycle by GI facilities. The applicability of the proposed model was verified by comparing the results of the proposed model with the results of SWMM. In addition, a discussion of errors occurring in the SWMM's permeable pavement system simulation is included.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.1
• /
• pp.13-20
• /
• 2022

Aircraft weight is an important factor affecting performance and fuel efficiency. In the conceptual design stage of the aircraft, the process of balancing cost and weight is performed using empirical formulas such as fuel consumption cost per weight in estimating element weight. In addition, when an airline operates an aircraft, it promotes fuel efficiency improvement, fuel saving and carbon reduction through weight management activities. The relationship between changes in aircraft weight and changes in fuel consumption is called the cost of weight, and the cost of weight is used to evaluate the effect of adding or reducing weight to an aircraft on fuel consumption. In this study, the problems of the existing cost of weight calculation method are identified, and a new cost of weight calculation method is introduced to solve the problem. Using Breguet's Range Formula and actual flight data of the A350-900 aircraft, two weight costs are calculated based on take-off weight and landing weight. In conclusion, it was suggested that it is reasonable to use the cost of weight based on the take-off weight and the landing weight for other purposes. In particular, the cost of weight based on the landing weight can be used as an empirical formula for estimating element weight and optimizing cost and weight in the conceptual design stage of similar aircraft.

• Journal of the Korean Geotechnical Society
• /
• v.37 no.11
• /
• pp.107-118
• /
• 2021

Scan-to-BIM can be precisely mod eled by measuring structures with Light Detection And Ranging (LiDAR) and build ing a 3D BIM (Building Information Modeling) model based on it, but has a limitation in that it consumes a lot of manpower, time, and cost. To overcome these limitations, studies are being conducted to perform semantic segmentation of 3D point cloud data applying deep learning algorithms, but studies on how segmentation result changes depending on learning data are insufficient. In this study, a parametric study was conducted to determine how the size and track type of railroad tunnels constituting learning data affect the semantic segmentation of railroad tunnels through deep learning. As a result of the parametric study, the similar size of the tunnels used for learning and testing, the higher segmentation accuracy, and the better results when learning through a double-track tunnel than a single-line tunnel. In addition, when the training data is composed of two or more tunnels, overall accuracy (OA) and mean intersection over union (MIoU) increased by 10% to 50%, it has been confirmed that various configurations of learning data can contribute to efficient learning.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.263-278
• /
• 2022

With the recent advent of IoT technology, automatic pet feeders are being distributed so that owners can feed their companion animals while they are out. However, due to behaviors of pets, the method of measuring weight, which is important in automatic feeding, can be easily damaged and broken when using the scale. The 3D camera method has disadvantages due to its cost, and the 2D camera method has relatively poor accuracy when compared to 3D camera method. Hence, the purpose of this study is to propose a deep learning approach that can accurately estimate weight while simply using a 2D camera. For this, various convolutional neural networks were used, and among them, the ResNet101-based model showed the best performance: an average absolute error of 3.06 grams and an average absolute ratio error of 3.40%, which could be used commercially in terms of technical and financial viability. The result of this study can be useful for the practitioners to predict the weight of a standardized object such as feed only through an easy 2D image.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.3
• /
• pp.141-148
• /
• 2022

In this study, the structural behavior of reinforced concrete members under simultaneous axial and blast loads was analyzed. Nonlinear dynamic analysis verification was performed using the experimental data of panels under fundamental blast load as well as those of reinforced concrete columns subjected to axial and blast loads. Because Autodyn is a program designed only for dynamic analysis, an analysis process is devised to simulate the initial stress state of members under static loads, such as axial loads. A total of 80 nonlinear dynamic finite element analysis procedures were conducted by selecting parameters corresponding to axial load ratios and scaled distances ranging 0%~70% and 1.1~2.0 (depending on the equivalent of TNT), respectively. The structural behavior was compared and analyzed with the corresponding degree of damage and maximum lateral displacement through the changes in axial load ratio and scaled distance. The results show that the maximum lateral displacement decreases due to the increase in column stiffness under axial loads. In view of the foregoing, the formulated analysis process is anticipated to be used in developing blast-resistant design models where structural behavior can be classified into three areas considering axial load ratios of 10%~30%, 30%~50%, and more than 50%.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.10
• /
• pp.373-380
• /
• 2022

This paper presents a deep learning based group synchronization that supports networked immersive interactions between remote users. The goal of group synchronization is to enable all participants to synchronously interact with others for increasing user presence Most previous methods focus on NTP-based clock synchronization to enhance time accuracy. Moving average filters are used to control media playout time on the synchronization server. As an example, the exponentially weighted moving average(EWMA) would be able to track and estimate accurate playout time if the changes in input data are not significant. However it needs more time to be stable for any given change over time due to codec and system loads or fluctuations in network status. To tackle this problem, this work proposes the Deep Group Synchronization(DeepGroupSync), a group synchronization based on deep learning that models important features from the data. This model consists of two Gated Recurrent Unit(GRU) layers and one fully-connected layer, which predicts an optimal playout time by utilizing the sequential playout delays. The experiments are conducted with an existing method that uses the EWMA and the proposed method that uses the DeepGroupSync. The results show that the proposed method are more robust against unpredictable or rapid network condition changes than the existing method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.7-15
• /
• 2022

Since the direct and indirect damage caused by the fire in the underground utility tunnel will cause great damage to society as a whole, it is necessary to make efforts to prevent and control it in advance. The most of the fires that occur in cables are caused by short circuits, earth leakage, ignition due to over-current, overheating of conductor connections, and ignition due to sparks caused by breakdown of insulators. In order to find the cause of fire at an early stage due to the characteristics of the underground utility tunnel and to prevent disasters and safety accidents, we are constantly managing it with a detection system using image analysis and making efforts. Among them, a case of developing a fire detection system using CCTV-based deep learning image analysis technology has been reported. However, CCTV needs to be supplemented because there are blind spots. Therefore, we would like to develop a high-performance acoustic-based deep learning model that can prevent fire by detecting the spark sound before spark occurs. In this study, we propose a method that can collect sound in underground utility tunnel environments using microphone sensor through development and experiment of prototype module. After arranging an acoustic sensor in the underground utility tunnel with a lot of condensation, it verifies whether data can be collected in real time without malfunction.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.429-440
• /
• 2021

With the recent development of deep convolutional neural network learning, deep learning techniques applied to single image super-resolution are showing good results. One of the existing deep learning-based super-resolution techniques is RDN(Residual Dense Network), in which the initial feature information is transmitted to the last layer using residual dense blocks, and subsequent layers are restored using input information of previous layers. However, if all hierarchical features are connected and learned and a large number of residual dense blocks are stacked, despite good performance, a large number of parameters and huge computational load are needed, so it takes a lot of time to learn a network and a slow processing speed, and it is not applicable to a mobile system. In this paper, we use the residual dense structure, which is a continuous memory structure that reuses previous information, and the residual dense channel attention block using the channel attention method that determines the importance according to the feature map of the image. We propose a method that can increase the depth to obtain a large receptive field and maintain a concise model at the same time. As a result of the experiment, the proposed network obtained PSNR as low as 0.205dB on average at 4× magnification compared to RDN, but about 1.8 times faster processing speed, about 10 times less number of parameters and about 1.74 times less computation.

• Journal of Convergence for Information Technology
• /
• v.11 no.7
• /
• pp.111-117
• /
• 2021

This research work is to suggest the experimental results capable of solving an initial unsuitability of combustion and environment in a constant volume combustion chamber by using LFG(Land Fill Gas) which consists of 40% CO₂ and 60% CH₄. The experimental condition is set as 0.9~1.6 of air-fuel ratio, 3bar of combustion pressure, 25℃ of room temperature, methane for using gas, and 2.5~4.5 of Pre-chamber hole sizes. As a result, it can be seen that diffusion of initial flame is significantly increased by M3.0 model comparing with other one. The reason for the characteristics is that orifice effect is extremely improved by 0.9, 1.0, and 1.2 of air-fuel ratio comparing with other one. Consequently, this experiment is shown that M3.0 model is partially capable of improving combustion performance than a conventional ignition plug in case of applying to LFG with Pre-chamber design.

• Journal of the Korea Institute of Building Construction
• /
• v.22 no.6
• /
• pp.565-575
• /
• 2022

Thanks to the latest developments in digital architectural technologies, free-form designs that maximize the creativity of architects have rapidly increased. However, there are a lot of difficulties in forming various free-form curved surfaces. In panelizing to produce free forms, the methods of mesh, developable surface, tessellation and subdivision are applied. The process of applying such panelizing methods when producing free-form panels is complex, time-consuming and requires a vast amount of manpower when extracting production data. Therefore, algorithms are needed to quickly and systematically extract production data that are needed for panel production after a free-form building is designed. In this respect, the purpose of this study is to propose mathematical algorithms for the automatic generation of production data of free-form panels in consideration of the building model, performance of production equipment and pattern information. To accomplish this, mathematical algorithms were suggested upon panelizing, and production data for a CNC machine were extracted by mapping as free-form curved surfaces. The study's findings may contribute to improved productivity and reduced cost by realizing the automatic generation of data for production of free-form concrete panels.