• 한국BIM학회 논문집
• /
• 제9권4호
• /
• pp.31-40
• /
• 2019

With Building Information Modeling(BIM) becoming the de facto standard for data sharing in the AEC industry, additional needs have increased to ensure the data integrity of BIM models themselves. Although the Industry Foundation Classes provide an open and neutral data format, its generalized schema leaves it open to data loss and misclassifications This research applied deep learning to automatically classify BIM elements and thus check the integrity of BIM-to-IFC mappings. Multi-view CNN(MVCC) and PointNet, which are two deep learning models customized to learn and classify in 3 dimensional non-euclidean spaces, were used. The analysis was restricted to classifying subtypes of architectural walls. MVCNN resulted in the highest performance, with ACC and F1 score of 0.95 and 0.94. MVCNN unitizes images from multiple perspectives of an element, and was thus able to learn the nuanced differences of wall subtypes. PointNet, on the other hand, lost many of the detailed features as it uses a sample of the point clouds and perceived only the 'skeleton' of the given walls.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2002년도 가을 학술발표논문집 Vol.29 No.2 (2)
• /
• pp.232-234
• /
• 2002

EEG-based brain computer interface (BCI) provides a new communication channel between human brain and computer. The classification of EEG data is an important task in EEG-based BCI. In this paper we present methods which jointly employ principal component analysis (PCA) and linear dynamical system (LDS) modeling for the task of EEG classification. Experimental study for the classification of EEG data during imagination of a left or right hand movement confirms the validity of our proposed methods.

• 한국정밀공학회지
• /
• 제26권12호
• /
• pp.55-61
• /
• 2009

Machining systems have been evolved to produce more detailed products of high added value. This has been possible, in large part, due to the development of highly accurate multi-axis CNC machine tools. The conventional CNC of machine tools has individual axis controllers to maximize tracking performance. On the other hand, cross-coupling controllers can be integrated into the conventional CNC to enhance contouring performance. For this multi-axis cross-coupling control system, it is necessary to automatically adjust the controller gains depending on operating conditions and/or other external conditions from an optimization perspective. This paper proposes automatic modeling of feed drive systems that minimizes the difference in behavior between the system model and the actual system. Based on the modeling, an integrated auto-tuning method is also proposed to improve both tracking and contouring accuracy of a 3-axis cross-coupling control system as well as users' convenience. The proposed methods are evaluated by both simulation and experiments.

• Wind and Structures
• /
• 제27권5호
• /
• pp.293-309
• /
• 2018

The prediction of multimode flutter relies, to a larger extent than bimodal flutter, on accurate modeling of the self-excited forces since it is challenging to perform experimental validation by using aeroelastic tests for a multimode case. This paper sheds some light on the accuracy of predicted self-excited forces by comparing numerical predictions of self-excited forces with measured forces from wind tunnel tests considering the flutter vibration mode. The critical velocity and the corresponding flutter vibration mode of the Hardanger Bridge are first determined using the classical multimode approach. Then, a section model of the bridge is forced to undergo a motion corresponding to the flutter vibration mode at selected points along the bridge, during which the forces that act upon it are measured. The measured self-excited forces are compared with numerical predictions to assess the uncertainty involved in the modeling. The self-excited lift and pitching moment are captured in an excellent manner by the aerodynamic derivatives. The self-excited drag force is, on the other hand, not well represented since second-order effects dominate. However, the self-excited drag force is very small for the cross-section considered, making its influence on the critical velocity marginal. The self-excited drag force can, however, be of higher importance for other cross-sections.

• Earthquakes and Structures
• /
• 제14권4호
• /
• pp.297-303
• /
• 2018

Seismic behavior of Osmanli and Senyuva stone bridges was addressed in this study. A combination of FEM and DEM was employed for getting closer to the real behavior of the bridge. One of the unique features of this combinational method is simulation close to reality. Modal numerical analysis was also used to verify the modeling. At the end of earthquake, a part of two lateral walls of Osmanli bridge was broken. The growth of arch cracks also increased during the earthquake. A part of right-hand wall of Senyuva Bridge was destructed during the earthquake. The left-hand side of the bridge wall was damaged during the earthquake but was not destructed.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2003년도 추계학술대회
• /
• pp.207-214
• /
• 2003

3D Modeling Solution can reduce projecting error and also make possible to design visually, so that it can minimize human resources. On the other hand, it has defects of operating difficulties caused by too much complex function, thus it consequently prevent user from prompt handling. According to this paper, the developed party S/W will help that user can construct the parts easily, can let everybody use 3D modeling solution readily, and make fast development of new product. This software will carry out prompt development of new product and derive satisfaction of customers from high quality. Also, saving time and expense naturally raise competitive power in consequence.

• Design & Manufacturing
• /
• 제13권2호
• /
• pp.6-11
• /
• 2019

In this study, we constructed the mathematical model to evaluate the roundness for plastic injection mold parts with complicated 3D curvatures. Mathematically we started off from the equation of circle and successfully derived an analytical solution so as to minimize the area of the residuals. On the other hand, we employed the numerical method the similar optimization process for the comparison. To verify the mathematical models, we manufactured and used a ball valve type plastic parts to apply the derived model. The plastic parts was fabricated under the process conditions of 220-ton injection mold machine with a raw material of polyester. we experimentally measured (x, y) position using 3D contact automated system and applied two mathematical methods to evaluated the accuracy of the mathematical models. We found that the analytical solution gives better accuracy of 0.4036 compared to 0.4872 of the numerical solution. The numerical method however may give adaptiveness and versatility for optional simulations such as a fixed center.

• International Journal of Advanced Culture Technology
• /
• 제10권4호
• /
• pp.420-426
• /
• 2022

Developing a machine learning model and putting it into production goes through a number of steps. Automated Machine Learning(AutoML) appeared to increase productivity and efficiency by automating inefficient tasks that occur while repeating this process whenever machine learning is applied. The high degree of automation of AutoML models allows non-experts to use machine learning models and techniques without the need to become machine learning experts. Automating the process of applying machine learning end-to-end with AutoML models has the added benefit of creating simpler solutions, generating these solutions faster, and often generating models that outperform hand-designed models. In this paper, the AutoML data is collected and AutoML's Color Petri net model is created and analyzed based on it.

• 정보처리학회논문지B
• /
• 제17B권4호
• /
• pp.287-294
• /
• 2010

TOF(Time of Flight) 기술은 물체의 3차원 깊이 정보 추출을 가능케 하는 기술 중의 하나이다. 하지만, TOF의 카메라의 출력인 깊이 영상을 이용한 물체의 3차원 위치 추출은 몸이나 손 등 크기가 큰 물체의 경우에는 비교적 신뢰성 있는 결과를 얻을 수 있는데 비하여 크기가 작은 손 끝의 경우에는 신뢰할 수 있는 값을 얻기 힘들다. 본 논문에서는 TOF 카메라에서 육면체 손 모델을 이용하여 수정된 손의 영상에서의 위치 정보와 팔 모델을 이용하여 손 끝의 3차원 좌표를 추정한다. 제안된 방법으로 실험을 한 결과 TOF 카메라의 깊이 영상만을 사용하여 인식한 손 끝의 3차원 위치정보와 비교하여 훨씬 더 향상된 결과를 얻을 수 있었다.

• 지구물리와물리탐사
• /
• 제6권4호
• /
• pp.165-170
• /
• 2003

최근 전기비저항탐사는 산악과 같은 지형의 기복이 심한 지역에서도 널리 수행되고 있다. 그러나 이러한 지형의 기복은 전기비저항 탐사자료의 왜곡을 유발하며, 잘못된 해석의 원인이 될 수도 있다. 본 논문에서는 전기비저항탐사에서의 지형효과를 유한요소법에 의한 수치모델링과 축소모형의 두 종류의 모형실험을 통하여 분석하였다. 한편 축소모형실험을 연못에서 수행함으로써, 실내 수조 모형실험에서 문제가 될 수 있는 가장자리 효과를 피할 수 있었다. 다양한 경사를 갖는 능선과 계곡 지형 모형에 대하여 모형실험을 수행하였으며, 두 종류의 실험결과가 실험한 모든 지형모형에 대하여 서로 잘 부합됨을 확인하였다. 분석결과, 계곡모델은 계곡의 중심부에 낮은 겉보기비저항대가 분포하고 그 좌우에 높은 겉보기비저항대가 나타나는 양상을 보이며, 능선모델은 그 반대의 양상을 보인다. 또한 능선 및 계곡 모두 그 경사가 심할 경우에는 음의 겉보기비저항을 나타낼 수도 있음을 확인하였다. 따라서 전기비저항 탐사 자료의 해석시 지형효과를 꼭 고려해야 할 것으로 판단된다.