• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.107-114
• /
• 2021

Recently artificial intelligence technology has been introduced in various fields and various machine learning models have been operated in various frameworks as academic interest has increased. However, these frameworks have different data formats, which lack interoperability, and to overcome this, the open neural network exchange format, ONNX, has been proposed. In this paper we describe how to transform multiple machine learning models to ONNX, and propose algorithms and inference systems that can determine machine learning techniques in an integrated ONNX format. Furthermore we compare the inference results of the models before and after the ONNX transformation, showing that there is no loss or performance degradation of the learning results between the ONNX transformation.

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.8
• /
• pp.57-64
• /
• 1997

The process design of backward and forward extrusion of axisymmetric part has been studied in this paper. The important factors of cold forging process with complex geometry are the design of initial billet shape, the possibility of forming by one-stage operation and the determination of preform shapes, etc. Based on the systematic procedure of process sequence design, the forming operation of cold forged part is analyzed by the commercial finite element program, DEFORM. The design criteria are forming load, geo- metrical filling without defect and a sound distribution of effective strain in final product. It is noted that one step of preform operation is required to obtain the final product. Numerical result is compared with experi- mental one. It is found that the analyzed result is in good agreement with actual forming result.

• Journal of the Korean Institute of Gas
• /
• v.9 no.4 s.29
• /
• pp.11-16
• /
• 2005

In this paper, the hydrogen osmosis test and the mechanical tensile test were carried out to examine hydrogen cracking behavior of STS444 with welding conditions. In $0.5MH_2SO_4+0.001M \;As_2O_3$ solution, the hydrogen embrittlement characteristics of weld zone of STS444 added to load of $1,400 kg/cm^2$ together with hydrogen osmosis by current of $30mA/cm^2$ for 60 min. was considered. As a result of study on the hydrogen embrittlement and mechanical characteristics of STS444외와 with welding conditions the tensile stress and elongation of STS444 get lower by the absorption of oil or water before welding. Also, the reduction rate of tensile stress and elongation of STS444 is larger because of hydrogen embrittlement by the absorption of oil or water. STS444 by the absorption of water before welding is more sensitive to hydrogen embrittlement than oil.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.211-218
• /
• 2021

A recent spread of personal mobility devices such as electric kickboards has brought about a rapid increase in accident cases. Such vehicles are susceptible to falling accidents due to their low dynamic stability and lack of outer protection chassis. This paper presents the development of an automatic emergency braking system and a safe starting system as driving assistance devices for electric kickboards. The braking system employed artificial intelligence to detect nearby threaening objects. The starting system was developed to disable powder to the motor until when the driver's boarding is confirmed. This study is meaningful in that it proposes the convergence technology of advanced driver assistance systems specialized for personal mobility devices.

• The monthly packaging world
• /
• s.221
• /
• pp.130-132
• /
• 2011

국내 최대 규모로 펼쳐지는 인쇄인의 축제, 제18회 국제인쇄산업대전 KIPES 2011이 이번 달 28일부터 10월 1일까지 일산 KINTEX 전시장에서 개최된다. 이번 KIPES 2011 전시회는 KINTEX 제1전시장 Hall 5에서 1만1천$m^2$의 규모로 개최될 뿐만 아니라 KINTEX 제2전시장 개장과 함께 한국컨벤션산업 사상 가장 큰 규모인 10만8천$m^2$의 규모로 2011한국산업대전으로 개최된다. 따라서 KIPES 2011 비롯, 국제플라스틱 고무산업전, 한국기계전, 국제종합공구전, 금속산업대전, 국제종합전기기기전이 합동으로 개최됨에 따라 제1회 한국산업대전은 국내바이어 15만명, 해외바이어 5천명이 참관할 것으로 보이는 만큼 KIPES 2011의 발전 시너지 역시 클 것으로 기대되고 있다. 1977년 국내에서는 처음으로 전문산업전시회의 활성화를 위하여 "국제인쇄산업전시회(KIPES)"를 비롯하여 전문산업전시회를 130여 회 개최해 오고 있는 김충진 (주)한국이앤엑스 대표이사를 통해 KIPES 2011에 대한 기대 효과를 들어 보았다.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2018.06a
• /
• pp.131-131
• /
• 2018

최근 절삭공구산업은 자동차, 항공기, IT, 선박, 에너지 등 첨단산업의 증가로 인해 CGI, CFRP, 내열합금 등 난삭재의 수요가 증가하고 있다. 난삭재는 고내열, 고경도, 초경량 같은 특성을 지니며 우수한 기계적 물성을 갖지만 가공의 어려움이 있어 산업에 적용하는데 한계가 있다. 이러한 한계를 극복하기 위해 개발된 가공기술 중 하드 코팅은 공구코팅비용 대비 공구의 표면경도와 수명을 효율적으로 향상시킬 수 있다고 알려져 있다. 대표적인 하드코팅으로는 AlN계, TiN계 코팅이 있다. 이러한 코팅의 경우 높은 기계적 물성과 우수한 내마모성으로 인해 절삭공구의 성능을 향상시킬 수 있기 때문에, 많은 연구가 진행되고 있으며 절삭공구산업에서 각광받고 있다. 기존 선행연구 결과에 따르면 질화물 코팅의 우수한 물성은 질화물(Nitride) 생성 및 질화 공정에 의한 코팅층의 고밀도화에 의해 나타난다고 알려져 있다. 그 중에서 AlCrN coating은 우수한 내마모성 및 향상된 고온경도를 갖고 있다. AlCrN based coating에 미량의 원소를 첨가하여 기존 AlCrN coating의 기계적 특성을 더욱 향상 시킨 coating은 일반적인 고성능 코팅 대비 공구수명이 길다고 알려져 있으며, 전반적으로 우수한 특성에 의해 전 세계적으로 습식 및 건식 기계 가공 용도로 사용되고 있다. 본 연구에서는 AlCrN based coating에 미량의 원소를 첨가한 coating의 우수한 기계적 특성의 원인을 규명하기 위해 텅스텐카바이드(WC) 기판 위에 아크 이온 플레이팅 장비를 이용하여 AlCrN based coating을 증착 시킨 sample을 분석하였다. 결정구조 및 상 분석을 위해 X선 회절분석(XRD)을 실시하였으며, 미세 구조를 분석하기 위해 전계방출형 주사전자현미경(FE-SEM), 투과 전자현미경(TEM) 분석을 실시하였다. 또한 코팅층의 화학적 성분 분석을 위해 EDX분석을 실시하였으며 기계적 특성 평가를 위해 나노압입시험(Nano-indentation test)을 진행하였다.

• Proceedings of the Korean Society of Postharvest Science and Technology of Agricultural Products Conference
• /
• 2003.04a
• /
• pp.109.1-109
• /
• 2003

중국, 유럽, 한국산 순무를 국내에서 파종.수확한 후 동치미를 담그어 발효 특성을 비교해 보고 김치 제조에 알맞은 품종을 알아보고자 하였다. 신선한 순무의 특성으로 중국산 순무는 가용성 고형물 함량, 경도 및 파쇄성이 가장 높았다. 한국산 순무는 안토시아닌 함량이가장 많았다. 원산지별 순무로 담근 동치미를 $0^{\circ}C$에서 70일까지 경시적으로 이화학적, 관능적 특성을 분석한 결과는 다음과 같다. 중국산 순무 동치미는 숙성 전 기간동안 산도, 가용성 고형물 함량, 탁도, 유산균수가 유럽산에 비하여 높아 발효속도가 가장 컸으며, 한국산 순무 동치미는 중간의 발효 속도를 나타내었다. 이같은 특성은 관능 검사 결과에서도 유사한 경향을 보여주었다. 안토시아닌 함량 및 적색도 a값은 한국산 순무 동치미가 가장 높았고 중국산 순무 동치미가 가장 낮았다. 조직감중 경도는 기계적 분석 결과와 관능 검사 결과 모두에서 한국산 순무 동치미와 중국산 순무 동치미가 숙성전 기간동안 높게 유지되었고 유럽산 순무 동치미는 낮았다. 그러나 전반적인 기호도는 숙성적기인 숙성 40 일에 한국산 순무 동치미가 6.7 점으로 가장 높았고, 그 다음이 중국산 동치미로 5.9 점이었고, 유럽산 동치미는 3.5 점으로 가장 낮았는데, 이 같은 경향은 숙성 전 기간 동안 유사한 경향을 나타내었다(p<0.05). 이상의 결과로부터 맛있는 동치미 제조를 위한 순무는 강화도 고유의 자색순무에 중국산 청색 순무가 지닌 높은 당도와 단단한 조직감을 나타내는 특성을 지닌 품종이 적합하다고 생각된다.

• Journal of Sensor Science and Technology
• /
• v.31 no.2
• /
• pp.96-101
• /
• 2022

As the elderly population gradually increases, the risk of fatal fall accidents among the elderly is increasing. One way to cope with a fall accident is to determine the fall direction before impact using a wearable inertial measurement unit (IMU). In this context, a previous study proposed a method of classifying fall directions using a support vector machine with sensor velocity, acceleration, and tilt angle as input parameters. However, in this method, the IMU signals are processed through several processes, including a Kalman filter and the integration of acceleration, which involves a large amount of computation and error factors. Therefore, this paper proposes a machine learning-based method that classifies the fall direction before impact using IMU raw signals rather than processed data. In this study, we investigated the effects of the following two factors on the classification performance: (1) the usage of processed/raw signals and (2) the selection of machine learning techniques. First, as a result of comparing the processed/raw signals, the difference in sensitivities between the two methods was within 5%, indicating an equivalent level of classification performance. Second, as a result of comparing six machine learning techniques, K-nearest neighbor and naive Bayes exhibited excellent performance with a sensitivity of 86.0% and 84.1%, respectively.

• Journal of the Microelectronics and Packaging Society
• /
• v.13 no.3 s.40
• /
• pp.39-45
• /
• 2006

This study was investigated the thermal properties of underfill with various epoxy resins using thermal analysis methods such as differential scanning calorimetry (DSC), thermo gravimetry analysis (TGA), dynamic mechanical analysis (DMA) and thermo-mechanical analysis (TMA). And, the adhesion strength of the underfills/FR-4 substrate was evaluated. The glass transition temperature (Tg) of underfill which was composed the cycolaliphatic epoxy resin was lower than that of underfill which was not composed the cycolaliphatic epoxy resin. The thermal degradation of underfill was composed of two processes, which involved chemical reactions between the degrading polymer and oxygen from the air atmosphere. The coefficient of thermal expansion (CTE) of underfill which was composed the cycolaliphatic epoxy resin was higher than that of underfill which was not composed the cycolaliphatic epoxy resin. The excessive curing temperatures caused a weak boundary layer of epoxy resin, which resulted in a deterioration of mechanical properties in the epoxy resin and thus led to poor adhesion property between the underfill/FR-4 substrate.

• Polymer(Korea)
• /
• v.29 no.4
• /
• pp.375-379
• /
• 2005

The morphology and therma]/viscoelastic characteristics were investigated for PLLA/MMT nanocomposite manufactured by incorporating inorganic nanosized silicate nanoplatelets into biodeuadable poly(l-lactic acid) (PLLA). The XRD difiactogram and TEM image may be regarded as a formation of homogeneously dispersed nanocomposites. The melting energy(${\Delta}H_m$) was increased during hydrolysis process because of increase of crystallinity. As MMT played a role of reinforcing agent, the storage modulus was increase in case of PLLA/MMT nanocomposite, it was well coincided with our previous results. From SEM image, many tiny pinholes formed by spinodal decomposition were observed on the surface, and the shape of nanocomposite was maintained during hydrolysis process. In this study, it was shown that the control of biodegradation rate, thermal/mechnical property was possibile by incorporating MMT.