• 한국추진공학회지
• /
• 제10권2호
• /
• pp.102-109
• /
• 2006

본 논문에서 항공기용 터보 축 엔진의 결함 진단 알고리즘을 개발하기 위해 Support Vector Machine(SVM)과 인공신경망(ANN)을 이용하였다. 신경망을 이용한 시스템은 비선형성이 과도한 데이터를 학습할 때 지역 최소점(Local Minima)에 빠져 분류 정확률이 낮아질 수 있다. 이러한 위험성을 보안하기 위해 SVM에 의한 ANN의 분할 학습 알고리즘(SLA)을 제안하였다. 이것은 SVM을 이용하여 결함 위치를 판별 한 후 신경망이 선택적으로 학습을 하는 방법으로 학습 데이터의 비선형성을 줄여 분류 정확률을 높이기 때문에 신경망을 단독으로 사용할 때보다 개선된 성능을 보여주었다.

• Structural Engineering and Mechanics
• /
• 제76권3호
• /
• pp.395-412
• /
• 2020

In this article, the values of internal force and deformation of a curved beam under any action with the firm or elastic supports are determined by using structural matrices. The article presents the general differential formulation of a curved beam in global coordinates, which is solved in an orderly manner using simple integrals, thus obtaining the transfer matrix expression. The matrix expression of rigidity is obtained through reordering operations on the transfer notation. The support conditions, firm or elastic, provide twelve equations. The objective of this article is the construction of the algebraic system of order twenty-four, twelve transfer equations and twelve support equations, which relates the values of internal force and deformation associated with the two ends of the directrix of the curved beam. This final algebraic system, expressed in matrix form, is divided into two subsystems: twelve algebraic equations of internal force and twelve algebraic equations of deformation. The internal force and deformation values for any point in the curved beam directrix are determined from these values in the initial position. The five examples presented show how to apply the matrix procedures developed in this article, whether they are curved beams with the firm or elastic support.

• The Journal of Korean Physical Therapy
• /
• 제31권4호
• /
• pp.222-227
• /
• 2019

Purpose: This study examined the influence of longitudinal arch on the strength and muscle activity of the abductor hallucis in the standing position in subjects with and without navicular drop signs. Methods: A sample of 34 subjects with and without navicular drop signs between 22 and 28 years of age were enrolled in this study. The strength and muscle activity of the abductor hallucis was measured using a tensiometer. The Smart KEMA System and electromyography device was used on the subjects with and without navicular drop signs. Two groups were classified using the navicular drop test to identify the longitudinal arch of the foot. The strength of the abductor hallucis was evaluated in standing, both with and without an external arch support condition. The two-way mixed ANOVA was used. The level of statistical significance was set to ${\alpha}=0.05$. Results: The strength and muscle activity of the abductor hallucis in standing was significantly higher with external arch support than that without the external arch support in the group with navicular drop signs. There was no significant difference in the abductor hallucis strength and muscle activity with and without external arch support in the subjects without navicular drop signs. Conclusions: The strength and muscle activity of the abductor hallucis in standing can be influenced by the external arch support in the group with navicular drop signs. The strength measurement of the abductor hallucis in standing should be separately performed in conditions with and without longitudinal arch of foot.

• 재활복지공학회논문지
• /
• 제9권4호
• /
• pp.285-291
• /
• 2015

본 논문에서는 IMU(inertial measurement unit)의 데이터를 이용하여 사람의 앉은 자세를 분류하는 알고리즘을 제안한다. 제안하는 알고리즘은 IMU의 데이터를 주성분 분석법(principle component analysis: PCA)을 이용하여 특징 벡터를 3개로 축소시켰고, RBF(radial basis function) 커널을 적용한 서포트 벡터 머신(support vector machine: SVM)을 이용하여 자세를 분류하였다. 데이터의 측정을 위하여 건강한 성인 3명을 대상으로 실험을 실시하였고, 데이터의 수집을 위하여 넥밴드 형태의 이어폰에 IMU를 내장한 장치를 개발하여 착용하였다. 피험자는 각각 neutral position, smartphoning, writing의 세 가지 앉은 자세에 대하여 실험을 진행하였다. 실험 결과 제안하는 PCA-SVM 알고리즘은 특징 벡터의 차원을 25%로 축소시키면서도 95%의 신뢰를 보였다.

• Structural Engineering and Mechanics
• /
• 제17권5호
• /
• pp.611-626
• /
• 2004

On the basis of a structural analysis of radial gate (i.e. Tainter gate), the current paper focuses on weight minimization according to the location of the arms on a radial gate. In spite of its economical significance, there are hardly any previous studies on the optimum design of radial gate. Accordingly, the present study identifies the optimum position of the support point for a radial gate that guarantees the minimum weight satisfying the strength constraint conditions. This study also identifies the optimum position for 2 or 3 radial arms with a convex cylindrical skin plate relative to a given radius of the skin plate curvature, pivot point, water depth, ice pressure, etc. These optimum designs are then compared with previously constructed radial gates. Local genetic and hybrid-type genetic algorithms are used as the optimum tools to reduce the computing time and enhance the accuracy. The results indicate that the weights of the optimized radial gates are appreciably lower than those of previously constructed gates.

• International Journal of Aeronautical and Space Sciences
• /
• 제18권3호
• /
• pp.414-424
• /
• 2017

This paper introduces an experimental device which can measure accurate aerodynamic forces without support interference in wide experimental region for wind tunnel test of micro aerial vehicles (MAVs). A stereo pattern recognition (SPR) method was introduced to a magnetic suspension and balance system (MSBS), which can eliminate support interference by levitating the experimental model, to establish wider experimental region; thereby MSBS-SPR integrated system was developed. The SPR method is non-contact, highly accurate three-dimensional position measurement method providing wide measurement range. To evaluate the system performance, a series of performance evaluations including SPR system measurement accuracy and 6 degrees of freedom (DOFs) position/attitude control of the MAV model were conducted. This newly developed system could control the MAV model rapidly and accurately within almost 60mm for translational DOFs and 40deg for rotational DOFs inside of $300{\times}300mm$ test section. In addition, a static wind tunnel test was conducted to verify the aerodynamic force measurement capability. It turned out that this system could accurately measure the aerodynamic forces in low Reynolds number, even for the weak forces which were hard to measure using typical balance system, without making any mechanical contact with the MAV model.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2008년 영문 학술대회
• /
• pp.72-79
• /
• 2008

The weak point of the artificial neural network(ANN) is that it is easy to fall in local minima when it learns too much nonlinear data. Accordingly, the classification ratio must be low. To overcome this weakness, the hybrid method has been proposed. That is, the ANN learns data selectively after detecting the defect position by the support vector machine(SVM). First, the SVM has been used for determination of the defect position and then the magnitude of the defect has been measured by the ANN. In off-design condition, the operation region of the engine is wide and the nonlinearity of learning data increases. The module system, dividing the whole operating region into reasonably small-size sections, has been suggested to solve this problem. In this study, the proposed algorithm has diagnosed the defects of triple components as well as single and dual components of the gas turbine engine in off-design condition.

• 디지털융복합연구
• /
• 제12권12호
• /
• pp.427-439
• /
• 2014

본 연구는 부산지역에 소재하는 미용실 종사자들을 대상으로 직무스트레스, 직무소진 및 사회적 지지와 이직의도와의 상관관계를 알아보고 이직의도에 미치는 변인들의 영향을 분석하고자 하였다. 첫째, 연령이 낮을수록 감정노동이 높으며, 연령, 학력, 경력, 직위가 낮을수록 직무스트레스와 직무소진이 높고, 연령과 학력이 낮고 휴무일이 적을수록 이직의도가 많았다. 둘째, 원장/디자이너는 감정절제, 감정부조화, 심리적 스트레스, 역할스트레스가 이직의도가 유의미하고 스텝은 감정부조화, 심리적 스트레스가 유의미한 것으로 나타났다. 셋째, 이직의도전체는 원장/디자이너는 감정절제, 역할스트레스, 상사지지와 동료지지가, 직장관련 이직의도에는 감정절제, 역할스트레스, 동료지지, 비인격화가 영향을 미친다. 스텝은 이직의도전체와 직장관련 이직의도에는 상사지지, 동료지지, 비인격화가, 직업관련 이직의도는 상사지지가 영향을 미친다.

• 대한심미치과학회지
• /
• 제8권1호
• /
• pp.46-59
• /
• 1999

A fixed bridge is preferred as a prosthetic option supported by oral implants. However, it is very difficult to re tore edentulous maxilla with fixed prosthesis in cases with improper position and angulation of fixtures, abnormal jaw relation, and need for proper lip support. Six Br${\aa}$nemark implants were installed in edentulous maxilla opposing mandible with natural dentition. A removable hybrid prosthesis attached to a bar milled by 6 degrees was therefore designed to overcome such disadvantages of fixed prosthesis. Lateral stabilization of removable prosthesis was obtained by framework closely fitting the milled bar. Support for the prosthesis was ensured by three elevated areas on the bar. Two precision attachments(CEKA REVAX) provided appropriate direct and indirect retention without influencing support. A clinical and laboratory procedure was presented.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2006년도 제26회 춘계학술대회논문집
• /
• pp.209-212
• /
• 2006

본 논문에서는 항공기용 터보 축 엔진의 결함 진단 알고리즘을 개발하지 위해 Support Vector Machine(SVM)과 인공신경망(ANN)을 이용하였다. SVM을 이용하여 결함 위치를 판별한 후 인공신경망이 선택적으로 학습하는 분할 학습 알고리즘(SLA)을 제안하였으며 이를 고도 변화에 따른 가스 터빈 엔진의 결함 진단에 적용하여 분류 속도 및 예측 정확률 개선 가능성을 확인하였다.