• 자원환경지질
• /
• 제28권5호
• /
• pp.469-480
• /
• 1995

Gravity and magnetic surveys were carried out to investigate the three-dimensional configuration and characteristics of the landfills at Nanjido waste disposal site. For terrain correction and three-dimensional density inversion of gravity data an algorithm, which calculates the gravity effect of a three-dimensional body by using the solid angle method, is developed. This algorithm has been proved to give more accurate terrain correction values for the small survey area having varied topography like Nanjido site as compared with widely used methods such as Hammer's method and multiquadric equation method. Density inversion of gravity anomaly data gives very useful information about the lateral and vertical variation of the landfills, which can be used to discriminate the kinds of wastes. The average density of filled materials appears to be $1.7\;g/cm^3$ which is much higher than the value $(0.8\;g/cm^3)$ estimated by Seoul City. The lateral variation of density shows high correlation with the pattern of ongoing depression of the landfills. The northern region of the landfill no. 1, which shows low density and high depression, is closely associated with the industrial waste and sludge filled area. The magnetic anomaly data provide information about relative concentration of magnetic materials, which is also very useful to investigate characteristics of the fills. Several high positive anomaly regions on the reduced-to-pole magnetic anomaly map are appeared to be associated with the industrial waste fills, but certain industrial waste fills show low negative anomalies. This kind of magnetic information can be used in selecting drilling locations over landfills away from buried metal products during the stabilization process.

• 응용통계연구
• /
• 제33권5호
• /
• pp.591-601
• /
• 2020

자기조직화지도는 고차원의 원자료를 노드들로 이루어진 저차원의 공간으로 투영하는 비지도학습 방법이다. 이 방법은 고차원의 자료를 노드들을 사용하여 2 또는 3차원의 공간에서 시각화할 수 있고, 이를 통해 자료의 특성을 탐색하는데 유용하다. 자료의 구조를 파악하기 위해 종종 노드들에 대한 군집분석을 시도하는데, 군집분석의 중요한 문제중 하나는 군집의 개수를 결정하는 것이다. 이 문제를 해결하기 위해 다양한 군집타당성지수들이 지금까지 개발되어 왔고, 이러한 지수들은 자기조직화지도의 노드들의 군집분석에 직접적으로 적용될 수 있다. 그러나, 자기조직화 지도가 원자료의 위상적 특성을 저차원 공간에 반영할 수 있다는 특징을 갖는데 반해, 이러한 일반적인 지수들은 이를 고려하지 않는 문제가 있다. 이에 본 연구에서는 원자료의 위상적 특성을 고려한 노드들 사이의 연결강도를 기반으로 하는 군집타당성지수를 제안한다. 이 새로운 군집타당성지수의 성능은 모의실험을 통해 기존의 군집타당성지수들과의 비교되고 검증된다.

• Journal of Advanced Marine Engineering and Technology
• /
• 제27권4호
• /
• pp.535-541
• /
• 2003

This paper describes numerical analysis of the impulse turbine with fixed guide vanes, a high performance hi-directional air turbine having simple structure for wane energy conversion. A 3-dimensional incompressible viscous flow numerical analysis based on the full Reynolds-averaged Wavier-Stokes equations was made to investigate the internal flow behavior Numerical results ate compared with experimental data. As a result, a suitable choice for the one of design factors has been clarified.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 2000년도 특별강연 및 추계학술발표대회 개요집
• /
• pp.40-42
• /
• 2000

This paper presents a new method of segmenting a one-dimensional signal into a set of features of type(line, Vee-groove, Lap-joint and etc.), A set of requirements for the segmentation process result from the application area, which in this case are laser welding, GMAW(Gas Metal Arc Welding), SAW(Submerged Arc Welding) and high speed tack welding. The algorithm is able to detect an exact welding position in the presence of noise and missing data, yet is reasonably economical to implement

• Journal of Information Technology Applications and Management
• /
• 제23권2호
• /
• pp.143-152
• /
• 2016

We can organize, manage, search, and process the documents efficiently by a document clustering. In general, the documents are clustered in a high dimensional feature space because the documents consist of many terms. In this paper, we propose a new method to cluster the documents efficiently in a low dimensional feature space by finding the core concepts from a domain ontology corresponding to the particular area documents. The experiment shows that our clustering method has a good performance.

• 산업공학
• /
• 제11권1호
• /
• pp.199-205
• /
• 1998

This paper describes the development of on-the-machine measuring(OMM) system which can directly measure the two and three dimensional machined accuracy using a scanning probe in milling machine. Two algorithms, NC program based continuous path(CP) measurement and CAD data assisted point to point(PTP) measurement, are developed for three dimensional measurements, with consideration of the characteristics of the scanning probe. The algorithms are used to develop an auto measuring system. The delveloped system is compared with the CMM (Coordinate Measuring Machine) in terms of accuracy and repeatability. The OMM system is expected to realize measurement time reduction and hence result in high productivity.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1994년도 추계학술대회 논문집
• /
• pp.336-342
• /
• 1994

This paper presents an useful technique for assessing the volumetric error in multi_axis machine tools using the kinematic double ball bar and 3 dimensional spherical contouring. The developed system proposes the 3 dimensional spherical contour for the error analysis. The developed system input the measured radial data, analysing the volumetric errors such as positional, strightness, angle, and squareness errors, etc. The developed system has been tested in a practical machine tool, and showed high

• 한국정밀공학회지
• /
• 제17권9호
• /
• pp.202-209
• /
• 2000

In recent years the concern of designing multi-stage gear drives increases with the more application of gear drives in high-speed and high-load. until now however research on the gear drive design has been focused on single gear pairs and the design has been depended on experiences and know-how of designers and carried out commonly by trial and error. We propose the automation of the dimensional design of gears and the configuration design for gear arrangement of two-and three-stage cylindrical gear drives. The dimensional design is divided into two types of design processes to determine the dimensions of gears. The first design process(Process I) uses the total volume of gears to determine gear ratio and uses K factor unit load and aspect ratio to determine gear dimensions. The second one(Process II) makes use of Niemann's formula and center distance to calculate gear ratio and dimensions. Process I and II employ material data from AGMA and ISO standards respectively. The configuration design determines the positions of gears to minimize the volume of gearbox by simulated annealing algorithm. Finally the availability of the design algorithm is validated by the design examples of two-and three-stage gear drives.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2008년 영문 학술대회
• /
• pp.723-728
• /
• 2008

We performed numerical analysis of base drag phenomenon, when a projectile with backward step flies into atmosphere at supersonic speed. We compared with other researchers. From our previous studies that were 2-dimensional simulation, we found out from sophisticated simulations that need dense mesh points to compare base pressure and velocity profile after from base with experimental data. Therefore, we focus on high order spatial disceretization over 3rd order with TVD such as MUSCL TVD 3rd, 5th, and WENO 5th order, and Limiters such as minmod, Triad. Moreover, we enforce to flux averaging schemes such as Roe, RoeM, HLLE, AUSMDV. In present, one dimensional result of Euler tests, there are Sod, Lax, Shu-Osher and interacting blast wave problems. AUSMDV as a flux averaging scheme with MUSCL TVD 5th order as spatial resolution is good agreement with exact solutions than other combinations. We are carrying out the same approaches into 3-dimensional base flow only candidate flux schemes that are Roe, AUSMDV. Additionally, turbulence models are used in 3-dimensional flow, one is Menter s SST DES model and another is Sparlat-Allmaras DES/DDES model in Navier-Stokes equations.

• ETRI Journal
• /
• 제44권2호
• /
• pp.286-299
• /
• 2022

Human activity recognition in real time is a challenging task. Recently, a plethora of studies has been proposed using deep learning architectures. The implementation of these architectures requires the high computing power of the machine and a massive database. However, handcrafted features-based machine learning models need less computing power and very accurate where features are effectively extracted. In this study, we propose a handcrafted model based on three-dimensional sequential skeleton data. The human body skeleton movement over a frame is computed through joint positions in a frame. The joints of these skeletal frames are projected into two-dimensional space, forming a "movement polygon." These polygons are further transformed into a one-dimensional space by computing amplitudes at different angles from the centroid of polygons. The feature vector is formed by the sampling of these amplitudes at different angles. The performance of the algorithm is evaluated using a support vector machine on four public datasets: MSR Action3D, Berkeley MHAD, TST Fall Detection, and NTU-RGB+D, and the highest accuracies achieved on these datasets are 94.13%, 93.34%, 95.7%, and 86.8%, respectively. These accuracies are compared with similar state-of-the-art and show superior performance.