• Korean Chemical Engineering Research
• /
• v.58 no.2
• /
• pp.230-234
• /
• 2020

A system that uses deep-learning techniques to predict properties from molecular structures has been developed to apply to chemical, biological and material studies. Based on the database where molecular structure and property information are accumulated, a deep-learning model looking for the relationship between the structure and the property can eventually provide a property prediction for the new molecular structure. In addition, experiments on the actual properties of the selected molecular structure will be carried out in parallel to carry out continuous verification and model updates. This allows for the screening of high-quality molecular structures from large quantities of molecular structures within a short period of time, and increases the efficiency and success rate of research. In this paper, we would like to introduce the overall composition of the materiality prediction system using deep-learning and the cases applied in the actual excavation of new structures in LG Chem.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.33-39
• /
• 2019

In this paper, we propose deep convolutional neural network(CNN) with bottleneck structure which improves the performance of earthquake classification. In order to address all possible forms of earthquakes including micro-earthquakes and artificial-earthquakes as well as large earthquakes, we need a representation and classifier that can effectively discriminate seismic waveforms in adverse conditions. In particular, to robustly classify seismic waveforms even in low snr, a deep CNN with 1x1 convolution bottleneck structure is proposed in raw seismic waveforms. The representative experimental results show that the proposed method is effective for noisy seismic waveforms and outperforms the previous state-of-the art methods on domestic earthquake database.

• Journal of Korean Medical classics
• /
• v.35 no.2
• /
• pp.17-32
• /
• 2022

Objectives : To overcome limitations of previous diagnostic systems and constitutional medicine by suggesting a new perspective of constitutional medicine and a system of Yin Yang and Five Phases that is applicable to many diseases untreatable by biomedicine. Methods : The Life Structure Diagram which shows the different distribution of Yin Yang and Five Phases according to constitution, reflects the non-local and simultaneous characteristics of Yin Yang and Five Phases. It overcomes previous diagnostic systems and constitutional medicine which were local and segmented. Each constitutional types were determined through their defining deep fractal pulse patterns, for which appropriate acupuncture methods and formulas were suggested. Results & Conclusions : A more effective differentiation of constitution and treatment could be achieved through the Life Structure Diagram, which could overcome limitations of pre-existing diagnosis and treatments of Korean Medicine.

• Structural Engineering and Mechanics
• /
• v.83 no.3
• /
• pp.401-413
• /
• 2022

The collapse of civil infrastructure due to natural disasters results in financial losses and many casualties. In particular, the recent increase in earthquake activities has highlighted on the importance of assessing the seismic performance and predicting the seismic risk of a structure. However, the nonlinear behavior of a structure and the uncertainty in ground motion complicate the accurate seismic response prediction of a structure. Artificial intelligence can overcome these limitations to reasonably predict the nonlinear behavior of structures. In this study, a deep learning-based algorithm was developed to estimate the time-history seismic response of bridge structures. The proposed deep neural network was trained using structural and ground motion parameters. The performance of the seismic response prediction algorithm showed the similar phase and magnitude to those of the time-history analysis in a single-degree-of-freedom system that exhibits nonlinear behavior as a main structural element. Then, the proposed algorithm was expanded to predict the seismic response and fragility prediction of a bridge system. The proposed deep neural network reasonably predicted the nonlinear seismic behavior of piers and bearings for approximately 93% and 87% of the test dataset, respectively. The results of the study also demonstrated that the proposed algorithm can be utilized to assess the seismic fragility of bridge components and system.

• Economic and Environmental Geology
• /
• v.33 no.6
• /
• pp.525-536
• /
• 2000

Electromagnetic and electric surveys were performed to reveal the deep structure of the Yangsan fault in the Bong-gye and Unyang areas, Kyeongnam Province, Korea. Especially, high-frequency magnetotelluric (HFMT) method of EM survey was mainly employed to study the deep subsurface configuration of Yangsan fault. HFMT survey was performed at 25 points of spacing 50 m, making 1.3 km survey line in Unyang area and 13 points of spacing 50 m, making 0.6 km survey line in Bong-gye area. Two 2-D cross-sections (Unyang and Bong-gye areas) were achieved as results. Electric survey by dipole-dipole array was performed to study the structure of shallow subsurface and compare the results with HFMT surveys. The results of HFMT and electric surveys show that Yangsan fault is a geologic boundary. It is very narrow and steep (about $80^{\circ}C$), and extends to 1~1.5 km depth.

• Journal of Korean Association for Spatial Structures
• /
• v.21 no.1
• /
• pp.79-86
• /
• 2021

Control performance of a smart tuned mass damper (TMD) mainly depends on control algorithms. A lot of control strategies have been proposed for semi-active control devices. Recently, machine learning begins to be applied to development of vibration control algorithm. In this study, a reinforcement learning among machine learning techniques was employed to develop a semi-active control algorithm for a smart TMD. The smart TMD was composed of magnetorheological damper in this study. For this purpose, an 11-story building structure with a smart TMD was selected to construct a reinforcement learning environment. A time history analysis of the example structure subject to earthquake excitation was conducted in the reinforcement learning procedure. Deep Q-network (DQN) among various reinforcement learning algorithms was used to make a learning agent. The command voltage sent to the MR damper is determined by the action produced by the DQN. Parametric studies on hyper-parameters of DQN were performed by numerical simulations. After appropriate training iteration of the DQN model with proper hyper-parameters, the DQN model for control of seismic responses of the example structure with smart TMD was developed. The developed DQN model can effectively control smart TMD to reduce seismic responses of the example structure.

• Smart Structures and Systems
• /
• v.31 no.5
• /
• pp.469-483
• /
• 2023

The dynamic characteristics of wind turbine blades are usually monitored by contact sensors with the disadvantages of high cost, difficult installation, easy damage to the structure, and difficult signal transmission. In view of the above problems, based on computer vision technology and the improved YOLOv5 (You Only Look Once v5) deep learning model, a non-contact dynamic characteristic monitoring method for wind turbine blade is proposed. First, the original YOLOv5l model of the CSP (Cross Stage Partial) structure is improved by introducing the CSP2_2 structure, which reduce the number of residual components to better the network training speed. On this basis, combined with the Deep sort algorithm, the accuracy of structural displacement monitoring is mended. Secondly, for the disadvantage that the deep learning sample dataset is difficult to collect, the blender software is used to model the wind turbine structure with conditions, illuminations and other practical engineering similar environments changed. In addition, incorporated with the image expansion technology, a modeling-based dataset augmentation method is proposed. Finally, the feasibility of the proposed algorithm is verified by experiments followed by the analytical procedure about the influence of YOLOv5 models, lighting conditions and angles on the recognition results. The results show that the improved YOLOv5 deep learning model not only perform well compared with many other YOLOv5 models, but also has high accuracy in vibration monitoring in different environments. The method can accurately identify the dynamic characteristics of wind turbine blades, and therefore can provide a reference for evaluating the condition of wind turbine blades.

• Journal of the Society of Naval Architects of Korea
• /
• v.40 no.6
• /
• pp.12-19
• /
• 2003

Corrugated bulkheads for a bulk carrier are divided into watertight bulkheads and deep tank bulkheads. Design of the watertight bulkheads is principally determined by the permissible limit of Classification and IACS requirements. But, the verification of strength through finite element analysis is indispensable for design of the deep tank bulkheads. A stage for stress evaluation of corrugated part is required for optimum structural design of the deep tank bulkheads. Since the finite element analysis for real model requires excessive amount of calculation time, in this study one corrugated structure is replaced with beam element and is idealized as 2 dimensional frame structure connected to upper and lower stool Minimum weight design of the deep tank bulkheads is performed through generalized sloped deflection method(GSDM) as direct calculation method. The purpose of this study is the development of design system for the minimization of steel weight of deep tank bulkheads as well as watertight bulkheads. Discrete variables are used as design variables for the practical design. Evolution strategies(ES) is used as an optimization technique.

• Ocean and Polar Research
• /
• v.27 no.2
• /
• pp.115-123
• /
• 2005

Recently, it has been observed in the East Sea that temperature increases below the thermocline, and dissolved oxygen increase in the intermediate layer but decrease below it. The layer of minimum dissolved oxygen deepens and the bottom homogeneous layer in oxygen becomes thinner. It emerges very probably that these changes are induced by the mode change of deep water formation associated with global warming. To further support this hypothesis, a one-dimensional model experiment is performed. First, a thermal profile is obtained by injecting a cold and high oxygen deep water into the bottom layer, say the bottom mode. Then, two thermal profiles are obtained from the bottom mode profile by assuming that either all the deep water introduce into the intermediate layer has been initiated, say the intermediate mode, or that only a part of the deep water has been initiated into the intermediate layer, say the intermediate-bottom mode. The results, from the intermediate-bottom mode experiment are closest to the observed results. They show quite well the tendency for oxygen to increase in the intermediate layer and the simultaneous thinning of the bottom homogeneous layer in oxygen. Therefore, it can be said that the recently observed slow variation of the thermal structure might be associated with changes in the deep water formation from the bottom mode to the intermediate-bottom mode.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.3
• /
• pp.337-347
• /
• 2020

In CRYPTO 2019, Gohr presents that Deep-learning can be used for cryptanalysis. In this paper, we verify whether Deep-learning can identify the structures of S-box. To this end, we conducted two experiments. First, we use DDT and LAT of S-boxes as the learning data, whose structure is one of mainly used S-box structures including Feistel, MISTY, SPN, and multiplicative inverse. Surprisingly, our Deep-learning algorithms can identify not only the structures but also the number of used rounds. The second application verifies the pseudo-randomness of and structures by increasing the nuber of rounds in each structure. Our Deep-learning algorithms outperform the theoretical distinguisher in terms of the number of rounds. In general, the design rationale of ciphers used for high level of confidentiality, such as for military purposes, tends to be concealed in order to interfere cryptanalysis. The methods presented in this paper show that Deep-learning can be utilized as a tool for analyzing such undisclosed design rationale.