• International Journal of Precision Engineering and Manufacturing
• /
• v.3 no.2
• /
• pp.5-14
• /
• 2002

Internet becomes very common tool far communication and data sharing. Virtual readily(VR) on web browser, and virtual prototyping and virtual manufacturing is widely used in many engineering fields. The reduction of overall development process and error minimization during data conversion becomes very crucial where sharing data via Internet and VR. This paper deals with the advantage and disadvantage of VRML format used in RP(Rapid Prototyping), and a software for RP data preparation is developed. If VRML format as an international standard for VR, is replaced with STL format, the weak points of STL formal can be overcome and the technique related to virtual prototyping and virtual manufacturing can be addressed more systematically by sharing the data. The system developed in this work shows a good window to get access to a mare realistic observation of an object for an RP system from a long remote sites in a more systematic way.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1753-1756
• /
• 2005

The sheet parts are formed with dies conventionally. But this conventional forming process is not suited to small volume and varied production for the reason of high cost. For the solution of this problem, a new forming process, which is called CNC incremental sheet forming, is being introduced. This process can form sheet parts without die, and is very well suited to small volume and varied production in space flight and automobile. In this paper, dieless CNC forming system based on a machining center is developed. A special device to grasp and pull the blank sheet built in the machining center and tool path generation S/W from STL file of 3-D model are developed. Several sheet parts are incrementally formed to verify the effectiveness of the developed system.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.1 s.94
• /
• pp.95-107
• /
• 1999

The design models of a new product in general are created using clay models or wooden mock-ups. The reverse engineering(RE) technology enables us to quickly create the CAD model of the new product by capturing the surface of the model using laser digitizers or coordinate measuring machines. Rapid prototyping (RP) is another technology that can reduce the product development time by fabricating the physical prototype of a part using a layered manufacturing technique. In reverse engineering process, however, the digitizer generates an enormous amount of point data, and it is time consuming and also inefficient to create surfaces out of these data. In addition, the surfacing operation takes a great deal of time and skill and becomes a bottleneck. In rapid prototyping, a faceted model called STL file has been the industry standard for providing the CAD input to RP machines. It approximates the CAD model of a part using many planar triangular patches and has drawbacks. A novel procedure that overcomes these problems and integrates RE with RP is proposed. Algorithms that drastically reduce the point clouds data have been developed. These methods will facilitate the use of reverse engineered geometric data for rapid prototyping, and thereby will contribute in reducing the product development time.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.04a
• /
• pp.724-727
• /
• 2000

Rapid Prototyping(RP)이란 3차원 솔리드 모델을 단면화한 뒤 하나씩 적층하는 가공방식을 총칭한다. 이때 단면화하는 방법에 따라서 uniform, adaptive slicing으로 나뉘며, 입력 모델에 따라서 direct slicing과 STL을 이용한 방식으로 나뉜다. 적층 방법에 따라서는 연속된 2D 윤곽을 기반으로 적층하는 vertical layer 방식과 인접한 두 개의 2D 윤곽들을 연결하며 만들어진 3D layer를 기반으로 가공하는 sloping layer방식으로 나뉠 수 있다. 현재 상용 RP 시스템들에서는 거의 모든 경우 vertical layer 방식이 채택되어 사용되고 있다. RP와 절삭 공정, 예를 들면 CNC 밀링의 장점을 효율적으로 결합하기 위해서는 임의의 복잡한 형상을 갖는 솔리드 모델을 정밀도에 제한이 없이 제조할 수 있어야 한다. 그러나 절삭 공정은 특별한 전문적 지식들을 필요로 한다 또한 상용 RP에서 사용하는 순차적인 적층 작업으로는 가공할 수 없는 형상들이 많다. 대표적인 것으로 지지대를 필요로 하는 형상들이 있다. 이러한 형상들을 지원하기 위해서는 복잡한 3D 형상을 절삭 가능한 형식으로 분할하는 것과 적층 가능한 순서대로 공정 계획하는 것이 필요하게 된다. 본 연구에서는 SDM에서 제시된 3D 분할 방법이 솔리드 모델을 기반으로 전개되어 STL file과 같은 삼각다면체 형식으로 근사화된 모델에 적용하기 어렵다는데 착안하여 STL file에서 읽어들인 삼각 다면체 모델을 가공 가능한 3D 형상으로 분할하는 알고리즘을 제시하고자 한다.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.75-78
• /
• 2002

SFFS(Solid Freeform Fabrication System) is commercializing to rapid prototyping concept in world-wide some corporations including the States, have much technological problems yet and need new mode for agile solid freeform fabrication as well as prototyping. In this paper, we design an automatic control algorithm that the cutting path of laser beam, on the SFFS, is controlled with constant speed. The designed algorithm for constant-speed path control is implemented and experimented in the $CAFL^{VM}$ (Computer Aided Fabrication of Lamination for Various Material) system, the new SFFS which is developed in this paper. Its process is an automated fabrication method in which a 3D object is constructed from STL(SToreoLithography) 2D data, derived from CAD 3D image, by sequentially laminating the part cross-sections. The constant-speed path control is started from the STL data. After STL file is modified in data format to be available for control. The fabrication of the 2D part is, with constant speed, conducted from the 23 position data by laser beam. we confirm its high-performance through experiment results from the application into $CAFL^{VM}$ system.

• International Journal of Highway Engineering
• /
• v.15 no.6
• /
• pp.11-15
• /
• 2013

PURPOSES : This study is to compare sound transmission loss(STL) value depending on the four kinds of materials, PC(Polycarbonate), PMMA(Polymethyl mathacrylate), PE(Polyethlyene), PP(Polypropylene), and two types of structure, single layer and double with vacuum layer, of soundproof panel. METHODS : With four sorts of polymer material, the specimens were made as various structures, 4 mm and 8 mm of single soundpoof panel and vacuum layered 4 mm of one. The experimental condition and procedures were complied with authorized process test, KS F 2808. RESULTS : STL of single panel made of PC were the greatest followed by PMMA, PE, PP regardless of the thickness of panel, However, STL of PMMA panel began to decrease around 2500 Hz and reached the lowest value among others in 5000 Hz. Vacuum layer soundproof panel showed good performance in more than 2000 Hz. Only vacuum layer panel made of PC presented resonance frequency at 800 Hz while that of other vacuum ones at 1000 Hz. CONCLUSIONS : According to results of single layer, it was found that single panel functioned as the theorical way we expected in terms of surface density. That trends were blurred as the panel got thicker. And it was suggested also that vacuum layer panel performed well at high frequency, more than 2000 Hz.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.6
• /
• pp.188-196
• /
• 2004

The objective of this paper is to propose a new technology of the orthopaedic surgery using the combination of reverse engineering (RE) based on CT data and rapid prototyping (RP). The proposed technology utilizes symmetrical characteristics of the human body and capability of the combination of RE and RP, which rapidly manufactures three-dimensional parts from CT data. The original .stl data of injured extents are generated from the mirror transformation of .stl file fur uninjured extents. The physical shape before injuring is manufactured from RP using the original .stl data. Subsequently, pre-operative planning, such as a selection of proper implants, preforming of the implant, a decision of fixation locations and an insert position for the implant, an estimation of the invasive size, and pre-education of operators are performed using the physical shape. In order to examine the applicability and the efficiency of the proposed surgical technology, various case studies, such as a distal tibia commented fracture, a proximal tibia plateau fracture and an iliac wing fracture of pelvis, are carried out. From the results of case studies, it has been shown that the proposed technology is an effective surgical tool of the orthopaedic surgery reducing the operational time, the operational cost, the radiation exposure of the patient and operators, and morbidity. In addition, the proposed technology could improve the accuracy of operation and the speed of rehabilitation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.180-194
• /
• 2021

For small image registration, feature-based approaches are likely to fail as feature detectors cannot detect enough feature points from low-resolution images. The classic FFT approach's prediction accuracy is high, but the registration time can be relatively long, about several seconds to register one image pair. To achieve real-time and high-precision rigid registration for small images, we apply deep neural networks for supervised rigid transformation prediction, which directly predicts the transformation parameters. We train deep registration models with rigidly transformed CIFAR-10 images and STL-10 images, and evaluate the generalization ability of deep registration models with transformed CIFAR-10 images, STL-10 images, and randomly generated images. Experimental results show that the deep registration models we propose can achieve comparable accuracy to the classic FFT approach for small CIFAR-10 images (32×32) and our LSTM registration model takes less than 1ms to register one pair of images. For moderate size STL-10 images (96×96), FFT significantly outperforms deep registration models in terms of accuracy but is also considerably slower. Our results suggest that deep registration models have competitive advantages over conventional approaches, at least for small images.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.95-98
• /
• 2021

최근 비정상적인 네트워크 활동 감지 및 네트워크 서비스 프로비저닝과 같은 다양한 분야에서 응용되는 네트워크 트래픽 예측 기술이 네트워크 통신 문제에 의한 트래픽의 결측 및 네트워크 유저의 불규칙한 활동에 의한 비선형 특성 때문에 발생하는 성능 저하를 극복하기 위해 딥러닝 신경망에 대한 연구가 활성화되고 있다. 이 딥러닝 신경망 중 시계열 딥러닝 신경망은 단기 네트워크 트래픽 볼륨을 예측할 때 낮은 오류율을 보인다. 하지만, 시계열 딥러닝 신경망은 기울기 소멸 및 폭발과 같은 비선형성, 다중 계절성 및 장기적 의존성 문제와 같은 한계를 보여준다. 이 논문에서는 계절성 임베딩을 고려한 주의 신경망 기반 트래픽 예측 기법을 제안한다. 제안하는 기법은 STL 분해 기법을 통해 분해된 트래픽 트랜드, 계절성, 잔차를 이용하여 일별 및 주별 계절성을 임베딩하고 이를 주의 신경망을 기반으로 향후 트래픽을 예측한다.

• Journal of Korean Society of Environmental Engineers
• /
• v.36 no.9
• /
• pp.597-603
• /
• 2014

The aim of the present study is to compare sound performance depending on thickness, materials, and structure of polymer soundproof panels consisting of PC, PMMA, HDPE, and PP, respectively. As a result of comparing sound transmission loss (STL) of single layer panel made of four types of polymer, the better sound transmission loss was obtained in order of PC, PMMA, HDPE, and PP, which was obviously followed mass law. 8 mm of single panel showed 5~6 dB(A) greater STL than that of 4 mm panels and lower frequency for coincidence effect so that STL of 8 mm panels decreased around 4,000~5,000 Hz, indicating less STL of 4 mm panels than those of 8 mm. When it comes to structure, 4 mm panels with air layer appeared similar value of STL with 8 mm single panels under 300 Hz. In range of high frequency above 2,000 Hz, 4 mm panels with air layer performed better than 8 mm of single layer panel while resonance effects were observed at 500~630 Hz. It was found that these results could be practically utilized as fundamental data for noise barriers design considering the change to each condition.