• Communications for Statistical Applications and Methods
• /
• 제31권2호
• /
• pp.235-246
• /
• 2024

In high-dimensional data analysis, sufficient dimension reduction (SDR) has been considered as an attractive tool for reducing the dimensionality of predictors while preserving regression information. The principal support vector machine (PSVM) (Li et al., 2011) offers a unified approach for both linear and nonlinear SDR. This article comprehensively explores a variety of SDR methods based on the PSVM, which we call principal machines (PM) for SDR. The PM achieves SDR by solving a sequence of convex optimizations akin to popular supervised learning methods, such as the support vector machine, logistic regression, and quantile regression, to name a few. This makes the PM straightforward to handle and extend in both theoretical and computational aspects, as we will see throughout this article.

• 한국멀티미디어학회논문지
• /
• 제20권2호
• /
• pp.134-143
• /
• 2017

Machine fault diagnosis recovers all the studies that aim to detect automatically faults or damages on machines. Generally, it is very difficult to diagnose a machine fault by conventional methods based on mathematical models because of the complexity of the real world systems and the obvious existence of nonlinear factors. This study develops an automatic machine fault diagnosis system that uses pattern recognition techniques such as principal component analysis (PCA) and artificial neural networks (ANN). The sounds emitted by the operating machine, a drill in this case, are obtained and analyzed for the different operating conditions. The specific machine conditions considered in this research are the undamaged drill and the defected drill with wear. Principal component analysis is first used to reduce the dimensionality of the original sound data. The first principal components are then used as the inputs of a neural network based classifier to separate normal and defected drill sound data. The results show that the proposed PCA-ANN method can be used for the sounds based automated diagnosis system.

• 펄프종이기술
• /
• 제31권5호
• /
• pp.73-85
• /
• 1999

Yield and fracture are separated in the tensile failure of paper. Failure in the machine direction of photocopy paper is contrasted with failure in the cross-machine direction . The ratios of distortion (shape change) to dilatation (volume change) for individual elements at yield and fracture are described. The ratios of distortion to dilatation are measured and compared to predicted values of the strain energy density theory. To evaluate the effect of the angle from the principal material direction on the strain energy density theory. To evaluate the effect of the angle from the principal material direction on the strain energy density factor, samples are prepared from machine direction to cross-machine direction in 15 degree intervals. the strain energy density of individual elements are obtained by the integration of stress from finite element analysis with elastic plus plastic strain energy density theory. Poison's ratio and the angle from the principal material direction have a great effect ion the ratio fo distortion to dilatation in paper. During the yield condition, distortion prevails over dilatation . At fracture, dilatation is at a maximum.

• 한국수산과학회지
• /
• 제31권3호
• /
• pp.317-322
• /
• 1998

Generally, the conventional fish size grading methods just adopt the mechanical technique. So the grading methods have a problem such that the graded fish is easy to hurt on the skin and in the internal organs. In this paper, a fish size grading machine is developed using an image processing method. The grading method is based on the principal axis theorem. The length and projected area of a fish are obtained by getting the principal axis and the product of inertia moment on the captured image of a target fish. The developed machine uses an uncontact technique that the target fishes go through the front side of the CCD camera. So the above stated problem can be improved. The performance of this method is discussed with the experimental results.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.453-456
• /
• 2022

전력망의 역할은 안정적인 전력공급이 최우선이다. 예고 없는 불안정한 상황에 대한 여러 가지 대비에 대한 방안이 필요하다. 기상 데이터를 활용하여 탐구적 데이터 분석을 통한 피처 간의 관계를 파악하여 머신러닝 기반의 에너지 생산 예측 모형을 모델링한다. 본 연구에서는 주성분분석을 사용하여 에너지 생산 예측 시 영향을 미치는 피처를 추출하였으며 머신러닝 모델에 적용함으로써 예측 신뢰도를 높였다. 제안한 모형을 사용하여 특정 기간을 대상으로 생산 에너지를 예측하고 해당 시점의 실제 생산 값과 비교함으로써 주성분분석을 적용한 에너지 생산 예측에 대한 성능을 확인하였다.

• Korean Chemical Engineering Research
• /
• 제60권2호
• /
• pp.223-228
• /
• 2022

화학공정에서 의도되지 않게 발생하는 이상은 큰 사고를 유발할 수 있다. 이러한 문제를 해결하기 위해, 신속하게 이상의 원인을 감지하고 판별하는 이상 진단 모델이 필요하다. 하지만, 이상 진단을 연구하는 대부분 연구의 경우, 상용프로그램에서 공정 시뮬레이션을 이용하여 이상 데이터를 생성하고 이를 이용하여 연구한 방법론을 적용하고 있다. 이는 실제 공정상에서 이상을 포함하는 실제 데이터를 얻는 데 많은 제약이 있음을 의미한다. 본 연구에서는 실제 폴리스티렌 반응기에서 얻은 이상 데이터와 정상 데이터를 분석하여 적절한 이상 진단 모델을 설계하고자 하였다. 먼저, 정상 데이터를 분석하여 세 가지의 조업 모드가 존재함을 확인하였으며, 모드 판별을 위한 모델을 SVM (Support Vector Machine)을 이용하여 만들었다. 각 조업 모드 별로 PCA (Principal Component Analysis)를 이용하여 이상 진단 모델을 만들었으며, 실제 이상 데이터를 이용하여 계산한 결과 신속하게 이상을 진단할 수 있음을 확인하였다. 본 연구에서 제안한 모델을 통해, 실제 사고가 발생하는 경우 신속한 대처가 가능하며, 이는 잠재적인 손실의 감소에 기여할 수 있음을 의미한다.

• 한국기계가공학회지
• /
• 제15권4호
• /
• pp.40-45
• /
• 2016

Interrupted cutting has different cutting characteristics compared with continuous cutting. In interrupted cutting, the workpiece has a groove that regularly impacts the cutting tool and workpiece. Therefore, tool damage occurs rapidly, and this increases the cutting force and surface roughness. In this study, we performed interrupted cutting of carbon steel for machine structure (SM45C) using a coated carbide tool (TT7100). To predict the cutting force, we analyzed the experimental results with a regression analysis. The results were as follows: We confirmed that the factors affecting the principal force and radial force were cutting speed, depth of cut, and feed rate. From the multi-regression analysis, we deduced regression equations, and their coefficients of determination were 89.6, 89.27, and 28.27 for the principal, radial, and feed forces, respectively. This means that the regression equations were significant for the principal and radial forces but not for the feed force.

• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1299-1311
• /
• 2017

Sounds based machine fault diagnosis recovers all the studies that aim to detect automatically faults or damages on machines using the sounds emitted by these machines. Conventional methods that use mathematical models have been found inaccurate because of the complexity of the industry machinery systems and the obvious existence of nonlinear factors such as noises. Therefore, any fault diagnosis issue can be treated as a pattern recognition problem. We present here an automatic fault diagnosis system of hand drills using discrete wavelet transform (DWT) and pattern recognition techniques such as principal component analysis (PCA) and artificial neural networks (ANN). The diagnosis system consists of three steps. Because of the presence of many noisy patterns in our signals, we first conduct a filtering analysis based on DWT. Second, the wavelet coefficients of the filtered signals are extracted as our features for the pattern recognition part. Third, PCA is performed over the wavelet coefficients in order to reduce the dimensionality of the feature vectors. Finally, the very first principal components are used as the inputs of an ANN based classifier to detect the wear on the drills. The results show that the proposed DWT-PCA-ANN method can be used for the sounds based automated diagnosis system.

• 지질공학
• /
• 제32권3호
• /
• pp.377-388
• /
• 2022

최근, 다수의 연구가 지수적으로 증가하는 지진 자료를 효율적이고 정확하게 처리하기 위해 기계학습을 활용하고 있다. 본 연구는 지진의 발생 시간, 위치, 규모의 정보를 확장하여 기계학습에 적용 가능한 자료를 제작한 후, 주성분 분석을 통해 추출한 자료의 주요 성분으로 자료의 차원을 축소하였다. 차원이 확장된 자료는 36,699개의 지진 사건을 포함하는 Global Centroid Moment Tensor 카탈로그로부터 얻은 지진 정보의 통계량으로 구성되었다. 표준화와 최대-최소화 스케일링을 활용하여 자료 전처리를 수행하였으며, 스케일링이 완료된 자료에 주성분 분석을 적용하여 자료의 주요 특징을 추출하였다. 스케일링은 상이한 단위로 인한 특징 값의 차이를 현저히 감소시켰으며, 그 중 표준화는 다른 전처리에 비해서 각 특징의 중앙값을 더 균등하게 변환하였다. 주성분 분석이 스케일링이 적용되지 않은 자료로부터 추출한 여섯 개의 주성분은 원본 자료의 정보를 99% 설명하였다. 표준화와 최대-최소 스케일링이 적용된 자료로부터 추출한 열여섯 개의 주성분은 원본 자료의 정보의 98%를 재구성하였다. 이는 특징 값의 분포가 균등한 자료의 정보를 보존하기 위해서는 더 많은 주성분이 필요함을 지시한다. 본 연구는 지진 데이터와 지진 거동과의 관계를 분석하는 효율적이고 정확한 기계 학습 모형을 훈련시키기 위한 데이터 처리 방법을 제안하였다.

• 한국안전학회지
• /
• 제34권5호
• /
• pp.95-102
• /
• 2019

Non-destructive methods such as rebound hardness method and ultrasonic method are widely studied for evaluating the physical properties, condition and damage of concrete, but are not suitable for detecting delamination and cracks near the surface due to various constraints of the site as well as the accuracy. Therefore, in this study, the impact resonance method was applied to detect the separation cracks occurring near the surface of the concrete slab and bridge deck. As a next step, the principal component analysis were performed by extracting various features using the FFT data. As a result of principal component analysis, it was analyzed that the reliability was high in distinguishing defects in concrete. This feature extraction and application of principal component analysis can be used as basic data for future use of machine learning technique for the better accuracy.