• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.1021-1024
• /
• 2001

Out of all metal-cutting processes, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. In this paper, the vision system of the sensing methods of drill flank wear on the basis of image processing is used to detect the wear pattern by non-contact and direct method and get the reliable wear information about drill. In image processing of acquired image, median filter is applied for noise removal. The vision flank wear area of the drill was measured. Backpropagation neural networks (BPns) were used for no-line detection of drill wear. The neural network consisted of three layers: input, hidden and output. The input vectors comprised of spindle rotational speed, feed rates, vision flank wear, thrust and torque signals. The output was the drill wear state which was either usable or failure. Drilling experiments with various spindle rotational speed and feed rates were carried out. The learning process was peformed effectively by utilizing backpropagation. The detection of the abnormal states using BPNs achieved 96.4% reliability even when the spindle rotational speed and feedrate were changed.

• Nuclear Engineering and Technology
• /
• 제47권2호
• /
• pp.219-225
• /
• 2015

In this paper, a three-dimensional model for the dynamic analysis of a flywheel based on the finite element method is presented. The static structure analysis for the model provides stress and strain distribution cloud charts. The modal analysis provides the basis of dynamic analysis due to its ability to obtain the natural frequencies and the vibration-made vectors of the first 10 orders. The results show the main faults are attrition and cracks, while also indicating the locations and patterns of faults. The harmonic response simulation was performed to gain the vibration response of the flywheel under operation. In this paper, we present a Hilbert-Huang transform (HHT) algorithm for flywheel vibration analysis. The simulation indicated that the proposed flywheel vibration signal analysis method performs well, which means that the method can lay the foundation for the detection and diagnosis in a reactor main coolant pump.

• 대한기계학회논문집
• /
• 제8권3호
• /
• pp.232-240
• /
• 1984

본 논문에서는 Stern이 제시한 경로적분식을 기본방정식으로 하여 예리한 임 의 노치내각을 가진(크랙의 경우 0˚), 즉 r$^{.lambda.}$ 형태의 특이점을 포함한 모우드-I 및 II 응력확대계수를 위한 특성해 및 보조해를 규정하고 선택모형문제로 예리한 노치 내각을 달리한 대칭 하중의 인장문제와 끝단 전단력하중하의 일단 고정보의 비대칭문 제의 응력확대계수를 기존의 재래식 유한요소법과 결합하여 계산하였다. 또한 각각 의 경우 적분경로 및 요소분할을 달리하여 수치해의 안정성 및 경로 독립성을 검토하 였다.

• 방송공학회논문지
• /
• 제13권3호
• /
• pp.289-298
• /
• 2008

본 논문은 특이값 분해에 기반하여 다양한 화질을 지원하는 3차원 메쉬 동영상의 SNR 계층 부호화 기법을 제안한다. SVD는 메쉬 동영상을 적은 수의 기저 벡터들과 특이값들로 표현하여 부호화 성능을 높일 수 있다. 본 논문에서는 비트 평면 부호화를 적용한 후 각 이진화 단계와 화질 사이의 관계를 정량적으로 유도한다. 유도된 관계식을 이용하여 비트량-왜곡 측면에서 최적화된 부호화 순서를 정의한다. 또한 시공간 영역의 잉여 정보를 효율적으로 제거하는 예측 기법을 제시한다. 모의 실험을 통하여 제안하는 알고리듬이 다양한 SNR을 지원하며 기존의 기법에 비해 향상된 비트량-왜곡 성능을 발휘함을 보인다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2011년도 춘계학술대회 논문집
• /
• pp.224-229
• /
• 2011

In this paper, in order to improve safety of hybrid electric vehicle a fault detection algorithm is introduced. The proposed algorithm uses SVDD techniques. Two methods for learning a lot of data are used in this technique. One method is to learn the data incrementally. Another method is to remove the data that does not affect the next learning. Using lines connecting support vectors selection of removing data is made. Using this method, lot of computation time and storage can be saved while learning many data. A battery data of commercial hybrid electrical vehicle is used in this study. In the study fault boundary via SVDD is described and relevant algorithm for virtual fault data is verified. It takes some time to generate fault boundary, nevertheless once the boundary is given, fault diagnosis can be conducted in real time basis.

• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2001년도 하계 학술대회 논문집(KISPS SUMMER CONFERENCE 2001
• /
• pp.285-288
• /
• 2001

본 논문에서는 신호간의 의존성과 관련성이 최소가 되도록 분리하는 독립성분분석 법을 이용하여 입력음성에서 변동량이 많은 방향으로 주축을 찾아 그 정보를 이용하여 데이터의 중복성을 제거한 후 음성특징벡터를 추출하는 방법을 제안한다. 학습 하고자하는 음성인식기의 음성에서 독립성분분석법을 이용하여 특징벡터를 추출하고 HMM 을 사용하여 기존의 음성특징벡터로 사용되는 mel-cepstrum과 비교하여 학습, 인식실험을 수행하였으며 제안한 방법에서 음성인식성능의 향상을 확인할 수 있었다. 또한, 인식시 주변여건에 따라 잡음에 의한 인식성능 저하에도 유연히 대처할 수 있음을 앞 수 있었다.

• 말소리와 음성과학
• /
• 제7권3호
• /
• pp.131-138
• /
• 2015

In this paper, we propose L1-norm regularization for state vector adaptation of subspace Gaussian mixture model (SGMM). When you design a speaker adaptation system with GMM-HMM acoustic model, MAP is the most typical technique to be considered. However, in MAP adaptation procedure, large number of parameters should be updated simultaneously. We can adopt sparse adaptation such as L1-norm regularization or sparse MAP to cope with that, but the performance of sparse adaptation is not good as MAP adaptation. However, SGMM does not suffer a lot from sparse adaptation as GMM-HMM because each Gaussian mean vector in SGMM is defined as a weighted sum of basis vectors, which is much robust to the fluctuation of parameters. Since there are only a few adaptation techniques appropriate for SGMM, our proposed method could be powerful especially when the number of adaptation data is limited. Experimental results show that error reduction rate of the proposed method is better than the result of MAP adaptation of SGMM, even with small adaptation data.

• Journal of Information Processing Systems
• /
• 제8권4호
• /
• pp.669-684
• /
• 2012

Discrete wavelet transforms are extensively preferred in biomedical signal processing for denoising, feature extraction, and compression. This paper presents a new denoising method based on the modeling of discrete wavelet coefficients of ECG in selected sub-bands with Kernel density estimation. The modeling provides a statistical distribution of information and noise. A Gaussian kernel with bounded support is used for modeling sub-band coefficients and thresholds and is estimated by placing a sliding window on a normalized cumulative density function. We evaluated this approach on offline noisy ECG records from the Cardiovascular Research Centre of the University of Glasgow and on records from the MIT-BIH Arrythmia database. Results show that our proposed technique has a more reliable physical basis and provides improvement in the Signal-to-Noise Ratio (SNR) and Percentage RMS Difference (PRD). The morphological information of ECG signals is found to be unaffected after employing denoising. This is quantified by calculating the mean square error between the feature vectors of original and denoised signal. MSE values are less than 0.05 for most of the cases.

• Tribology and Lubricants
• /
• 제36권6호
• /
• pp.324-331
• /
• 2020

Wedging in a frictional elastic system is defined if the state of stick exists after the external loading on the system is removed. This paper presents a method to determine the critical coefficient of wedging for an elastic frictional system by considering the separation state. Wedging is always possible if the coefficient of friction exceeds a critical value known as the critical wedging coefficient. This method requires two concepts: a necessary and sufficient condition for wedging, which can be interpreted as positive spanning sets of constraint vectors existing in the wedged system, and the minimal positive basis that enables a minimum wedging coefficient. The algorithm based on the positive spanning concept is repeatedly executed after eliminating nodes from the contact stiffness matrix, for which the separation states are impending. The simulation results show that once a node enters the separation state, it never returns to the contact state again and the critical wedging coefficient reduces during repeated algorithm execution. The benefit of this method is that the computation time permits handling models with large numbers of contact nodes. The algorithm can also numerically find the critical wedging coefficient, thereby contributing to fastening and assembly performance improvements in mechanical systems.

• 한국전산구조공학회논문집
• /
• 제24권3호
• /
• pp.329-335
• /
• 2011

유한요소법은 수학과 공학을 비롯한 다양한 분야에서 활용되고 있으나 해석대상을 유한 개의 다각형 요소로 분할하여 모델링하기 때문에 기하학적인 형상을 정확하게 기술하지 못하는 어려움이 있다. 그러나 최근에는 NURBS(Non-Uniform Rational B-Spline)를 기저함수로 사용하는 아이소-지오메트릭 해석법(Isogeometric analysis)이 개발되었는데 NURBS는 기하학적 모델을 정확하게 표현할 수 있을 뿐만 아니라 해석의 기저함수로서 응답해석에 사용될 수 있다. 그러나 NURBS 기저함수를 해석에서 따로 구성하는 일은 유한요소해석에서 요소망을 구성하는 만큼 시간과 노력이 많이 요구된다. 아이소-지오메트릭 해석법은 CAD(Computer-Aided Design)와 기하학적 정보를 공유할 수 있기 때문에 CAD 코드로 부터 해석모델의 정보를 직접 얻는 것이 가능하다. 본 논문에서는 상용 CAD 코드인 Rhinoceros 3D를 이용하여 CAD 모델을 작성하고 이를 STEP 파일로 출력하여 NURBS의 노트벡터와 조정점 등의 정보를 아이소-지오메트릭 해석법에 활용하는 기법을 소개한다. 몇몇 수치예제를 통하여 아이소-지오메트릭 해석법의 정확도를 유한요소해석 결과와 비교하여 검증하고, 상용 CAD와 CAE(Computer-Aided Engineering)가 결합된 아이소-지오메트릭 해석법의 효율성을 입증한다.