• 한국터널지하공간학회 논문집
• /
• 제18권1호
• /
• pp.13-30
• /
• 2016

본 연구에서는 해저 터널의 특수성을 고려한 TBM 세그먼트 라이닝의 최적 설계 시스템을 개발하였다. 해저 터널은 일반적으로 일정 수압 하의 토사나 암반 등으로 구성된 해저 지반 내에 시공된다. 본 설계 시스템은 특정 해저 터널 단면에서의 지반 조건, 시공 조건 및 터널 조건을 고려하여 인공신경망 기반의 세그먼트 라이닝 부재력 예측 시스템을 구축하고, 시공성이 확보된 단면 DB를 구축하여 해저터널에서 최적 단면 설계가 가능하도록 구성하였다. 결과적으로 본 시스템은 해저 터널 설계에 사용되는 BIM과 연동되어 자동으로 설계가 가능하도록 하였다. 단면 검토 및 설계에 사용되는 세그먼트 라이닝 부재력 예측은 유한요소해석을 토대로 구축한 인공신경망을 통해 일반화한 후 BIM 시스템에 접목시켜 별도의 추가 해석이 필요없이 유사 단면의 해저 터널 설계에 적용이 가능하도록 하였다.

• Geomechanics and Engineering
• /
• 제22권4호
• /
• pp.291-303
• /
• 2020

The deformation of the rock surrounding a tunnel manifests due to the stress redistribution within the surrounding rock. By observing the deformation of the surrounding rock, we can not only determine the stability of the surrounding rock and supporting structure but also predict the future state of the surrounding rock. In this paper, we used grey system theory to analyse the factors that affect the deformation of the rock surrounding a tunnel. The results show that the 5 main influencing factors are longitudinal wave velocity, tunnel burial depth, groundwater development, surrounding rock support type and construction management level. Furthermore, we used seismic prospecting data, preliminary survey data and excavated section monitoring data to establish a neural network learning model to predict the total amount of deformation of the surrounding rock during tunnel collapse. Subsequently, the probability of a change in deformation in each predicted section was obtained by using a Bayesian method for detecting change points. Finally, through an analysis of the distribution of the change probability and a comparison with the actual situation, we deduced the survey mark at which collapse would most likely occur. Surface collapse suddenly occurred when the tunnel was excavated to this predicted distance. This work further proved that the Bayesian method can accurately detect change points for risk evaluation, enhancing the accuracy of tunnel collapse forecasting. This research provides a reference and a guide for future research on the probability analysis of tunnel collapse.

• 경영정보학연구
• /
• 제12권2호
• /
• pp.75-88
• /
• 2010

정보기술의 발전과 더불어 기업과 고객간의 대부분의 정보가 축적되면서 기업은 거래고객의 자세한 정보를 활용하여 차별화된 마케팅을 제공할 수 있다. 본 연구는 기업이 제공하는 마케팅 전략을 보다 효과적으로 실행하기 위해 고객을 세분화하고, 세분화된 고객집단별 마케팅 프로모션에 대한 반응을 예측하는 모형을 제시하였다. 고객세분화에는 데이터마이닝 기법 중 SOM(Self-organizing Map)을 적용하였으며, 세분화된 집단별 프로모션 반응예측에는 로짓모형, 신경망 등의 단일모형과 k-최근접이웃법을 이용한 단일모형들의 통합모형을 적용하였다. 제시된 방법론으로 기업은 프로모션에 대한 고객반응을 예측할 뿐만 아니라 프로모션에 대한 반응을 쉽게 예측할 수 있는 고객집단과 반응예측이 어려운 고객집단으로 구분하여 프로모션의 효과를 극대화하고 각 집단에 맞는 프로모션 전략을 수립할 수 있다.

• 한국기계가공학회지
• /
• 제19권7호
• /
• pp.7-15
• /
• 2020

Recently, the fourth industrial revolution has been presented as a new paradigm and additive manufacturing (AM) has become one of the most important topics. For this reason, process monitoring for each cross-sectional layer of additive metal manufacturing is important. Particularly, deep learning can train a machine to analyze, optimize, and repair defects. In this paper, image classification is proposed by learning images of defects in the metal cross sections using the convolution neural network (CNN) image labeling algorithm. Defects were classified into three categories: crack, porosity, and hole. To overcome a lack-of-data problem, the amount of learning data was augmented using a data augmentation algorithm. This augmentation algorithm can transform an image to 180 images, increasing the learning accuracy. The number of training and validation images was 25,920 (80 %) and 6,480 (20 %), respectively. An optimized case with a combination of fully connected layers, an optimizer, and a loss function, showed that the model accuracy was 99.7 % and had a success rate of 97.8 % for 180 test images. In conclusion, image labeling was successfully performed and it is expected to be applied to automated AM process inspection and repair systems in the future.

• Advances in Computational Design
• /
• 제3권3호
• /
• pp.269-288
• /
• 2018

180 different 2D numerical analyses have been carried out to estimate the factor of safety (FOS) for rooted slopes. Four different types of vegetated coverage and a variety of slope geometry considering three types of soil have been evaluated in this study. The highly influenced parameters on the slope's FOS are determined. They have been chosen as the input parameters for developing a new practical relationship to estimate the FOS with an emphasis on the roots effects. The dependency of sliding mode and shape considering the soil and roots-type has been evaluated by using the numerical finite element model. It is observed that the inclination and height of the slope and the coverage type are the most important effective factors in FOS. While the soil strength parameters and its physical properties would be considered as the second major group that affects the FOS. Achieved results from the developed relationship have shown the acceptable estimation for the roots slope. The extracted R square from the proposed relationship considering nonlinear estimation has been achieved up to 0.85. As a further cross check, the achieved R square from a multi-layer neural network has also been observed to be around 0.92. The numerical verification considering different scenarios has been done in the current evaluation.

• 한국지능시스템학회논문지
• /
• 제23권5호
• /
• pp.479-486
• /
• 2013

항공기의 제어 시스템 설계에 있어 두 가지 어려움이 있다. 즉 항공기의 동적 특성이 비선형 특성을 갖고 있고 그 파라미터 값들이 시간 혹은 비행 조건에 따라 변화하는 시변 특성을 갖고 있다는 점이다. 그럼에도 불구하고 고전적인 제어 이론을 활용한 신뢰성 높고 효율적인 제어 기법들이 계속 개발되어 왔으나 정확한 이론적 분석이 수반되지 않으면 항공기의 성능, 강건성, 그리고 안전성조차도 확보하기 어려운 문제점을 갖는다. 이에 최근에는 퍼지 논리, 신경망, 유전자 알고리즘으로 대표되는 지능 제어 기법을 활용한 항공기 제어 시스템 개발이 시도 되고 있다. 본 논문에서는 기존의 퍼지 논리가 갖고 있는 불확실성에 대한 취약점들을 크게 감소시킬 수 있는 Interval Type-2 퍼지 논리 이론을 기반으로 고속 항공기의 지능형 비행 횡 제어 시스템을 개발하고 컴퓨터 모의실험에 의해 그 효용성을 입증한다.

• Journal of Acupuncture Research
• /
• 제35권1호
• /
• pp.28-36
• /
• 2018

Background: This study investigated the effects of acupuncture at Sobu (HT8) and Haenggan (LR2) on scopolamine-induced, cognitively impaired rats. Methods: Scopolamine-treated Sprague-Dawley rats were divided into 6 groups; normal, control, HT8, LR2, HT8 + LR2 and sham group. Cognitive impairment was induced by scopolamine, in control, and then in HT8, LR2, HT8 + LR2 and sham groups. Acupuncture treatment was performed at HT8, LR2, HT8 + LR2, and a random acupoint, respectively, every other day for 2 weeks. After each treatment, behavior change was observed and the rats were sacrificed. The change in brain-derived neurotrophic factor, glutathione peroxidase, and superoxide dismutase activity was evaluated by polymerase chain reaction. Results: Latency time to target in Morris Water-Maze test for the HT8 + LR2 group showed a significant decrease compared with control (p<0.05). Target crossing times and time zone ratios in Morris Water-Maze test for HT8 + LR2 group showed a significant increase compared with control (p<0.01). In the Y-Maze test the HT8 + LR2 group showed a significant increase compared with control (p<0.05). Brain-derived neurotrophic factor, glutathione peroxidase, and superoxide dismutase, in the HT8 + LR2 group, showed a significantly increased level compared with control (p<0.05). Neural activity of acetylcholine esterase in HT8 + LR2 group showed a significant decrease compared with the control group (p<0.01), choline acetyltransferase activity in the HT8 + LR2 group showed a significant increase compared with control (p<0.05). Conclusion: Acupuncture at HT8 + LR2 restored scopolamine-induced cognitive impairment, suggesting acupuncture could be an alternative to improve cognitive function.

• Journal of Power Electronics
• /
• 제13권4호
• /
• pp.516-527
• /
• 2013

This paper presents a new overall system for state-of-available-power (SoAP) prediction for a lithium-ion battery pack. The essential part of this method is based on an adaptive network architecture which utilizes both fuzzy model (FIS) and artificial neural network (ANN) into the framework of adaptive neuro-fuzzy inference system (ANFIS). While battery aging proceeds, the system is capable of delivering accurate power prediction not only for room temperature, but also at lower temperatures at which power prediction is most challenging. Due to design property of ANN, the network parameters are adapted on-line to the current battery states (state-of-charge (SoC), state-of-health (SoH), temperature). SoC is required as an input parameter to SoAP module and high accuracy is crucial for a reliable on-line adaptation. Therefore, a reasonable way to determine the battery state variables is proposed applying a combination of several partly different algorithms. Among other SoC boundary estimation methods, robust extended Kalman filter (REKF) for recalibration of amp hour counters was implemented. ANFIS then achieves the SoAP estimation by means of time forward voltage prognosis (TFVP) before a power pulse occurs. The trade-off between computational cost of batch-learning and accuracy during on-line adaptation was optimized resulting in a real-time system with TFVP absolute error less than 1%. The verification was performed on a software-in-the-loop test bench setup using a 53 Ah lithium-ion cell.

• 한국유체기계학회 논문집
• /
• 제13권5호
• /
• pp.43-53
• /
• 2010

This study investigates a design optimization of a rotating two-pass rectangular cooling channel with staggered arrays of pin-fins. The radial basis neural network method is used as an optimization technique with Reynolds-averaged Navier-Stokes analysis of fluid flow and heat transfer with shear stress transport turbulent model. The ratio of the diameter to height of the pin-fins and the ratio of the streamwise spacing between the pin-fins to height of the pin-fin are selected as design variables. The optimization problem has been defined as a minimization of the objective function, which is defined as a linear combination of heat transfer related term and friction loss related term with a weighting factor. Results are presented for streamlines, velocity vector fields, and contours of Nusselt numbers, friction coefficients, and turbulent kinetic energy. These results show how fluid flow in a two-pass square cooling channel evolves a converted secondary flows due to Coriolis force, staggered arrays of pin-fins, and a $180^{\circ}$ turn region. These results describe how the fluid flow affects surface heat transfer. The Coriolis force induces heat transfer discrepancy between leading and trailing surfaces, having higher Nusselt number on the leading surface in the second pass while having lower Nusselt number on the trailing surface. Dean vortices generated in $180^{\circ}$ turn region augment heat transfer in the turning region and in the upstream region of the second pass. As the result of optimization, in comparison with the reference geometry, thermal performance of the optimum geometry shows the improvement by 30.5%. Through the optimization, the diameter of pin-fin increased by 14.9% and the streamwise distance between pin-fins increased by 32.1%. And, the value of objective function decreased by 18.1%.

• 대한화학회지
• /
• 제45권5호
• /
• pp.422-428
• /
• 2001

기체 크로마토그래피에서 PAH와 그것의 유도체들의 상대적 용리시간을 인공신경망분석과 다변량 선형 회귀분석을 사용하여 학습한 후, 시험세트들의 상대적 용리시간을 예측하였다. QSRR에서 PAH와 그것의 유도체의 주요한 설명인자는 분자량의 제곱근, 분자의 연결지수($^1{\chi}_v$), 분자 쌍극자모멘트 및 분자의 길이와 폭의 비율(L/B)이었다. 다변량선형회귀분석에 의하면 큰 분자일수록 용리시간은 길어지며 또한 L/B의 값이 커지면 용리시간이 증가하는 것으로 보아 슬롯이론을 따르고 있음을 알 수 있었다. 반면에 설명인자 사이의 선형 독립성에 영향을 받지 않는 인공신경망 분석결과에 의하면 분자량과 분자 쌍극자 모멘트가 주요한 인자로 작용하고 있었다. 시험세트의 예측 정확도를 나타내는 분산은 선형회귀분석에서는 1.860, 인공신경망분석법에서 0.206으로서 인공신경망 분석법이 다변량회귀분석보다 더 좋은 예측방법임을 알 수 있었다.