• 응용통계연구
• /
• 제29권3호
• /
• pp.525-537
• /
• 2016

본 연구에서는 시계열 자료의 비정상성과 비선형성과 같은 복잡성을 효과적으로 포용할 수 있는 경험적모드분해법(empirical mode decomposition; EMD)을 토대로 시계열 자료의 분석 및 예측을 위한 혼합(hybrid) 모형을 연구한다. EMD에 의하여 생성되는 내재모드함수(intrinsic mode function; IMF)는 해석 및 예측의 편리성을 개선하기 위하여 누적에너지의 개념을 사용하여 그룹화하였으며, 그룹화된 IMF 및 residue의 성분들은 그 성질에 따라서 ARIMA 모형 및 지수평활법과 결합된 혼합 모형으로 예측된다. 제안된 방법은 일별 코스피 지수의 예측을 위해서 적용하였다. 다양한 형태의 혼합 모형을 사용하여 코스피 지수를 예측하였으며 전통적인 예측 방법과 비교하였다. 분석 결과, 그룹화된 성분들은 코스피 지수의 움직임을 단기적, 중기적, 장기적으로 해석하는데 편리함을 주었으며, 그룹화된 IMF 및 residue를 각각 ARIMA 모형과 지수평활법으로 조합한 혼합 모형이 우수한 예측력을 보여주었다.

• Geomechanics and Engineering
• /
• 제27권6호
• /
• pp.551-560
• /
• 2021

Due to the low strength and high compressibility characteristics, the loess deposits tunnels are prone to large deformations and collapse. An accurate stability evaluation for loess deposits is of considerable significance in deformation control and safety work during tunnel construction. 37 groups of representative data based on real loess deposits cases were adopted to establish the stability evaluation model for the tunnel project in Yan'an, China. Physical and mechanical indices, including water content, cohesion, internal friction angle, elastic modulus, and poisson ratio are selected as index system on the stability level of loess. The data set is randomly divided into 80% as the training set and 20% as the test set. Firstly, principal component analysis (PCA) is used to convert the five index system to three linearly independent principal components X1, X2 and X3. Then, the principal components were used as input vectors for probabilistic neural network (PNN) to map the nonlinear relationship between the index system and stability level of loess. Furthermore, Leave-One-Out cross validation was applied for the training set to find the suitable smoothing factor. At last, the established model with the target smoothing factor 0.04 was applied for the test set, and a 100% prediction accuracy rate was obtained. This intelligent classification method for loess deposits can be easily conducted, which has wide potential applications in evaluating loess deposits.

• Asian Pacific Journal of Cancer Prevention
• /
• 제13권5호
• /
• pp.1851-1856
• /
• 2012

Immigration to a nation with a stronger anti-smoking environment has been hypothesized to make smoking less common. However, little is known about how environments influence risk of smoking across the lifecourse. Research suggested a linear decline in smoking over the lifecourse but these associations, in fact, might not be linear. This study assessed the possible nonlinear associations between age and smoking and examined how these associations differed by environment through comparing Koreans in Seoul, South Korea and Korean Americans in California, United States. Data were drawn from population based telephone surveys of Korean adults in Seoul (N=500) and California (N=2,830) from 2001-2002. Locally weighted scatterplot smoothing (lowess) was used to approximate the association between age and smoking with multivariable spline logistic regressions, including adjustment for confounds used to draw population inferences. Smoking differed across the lifecourse between Korean and Korean American men. The association between age and smoking peaked around 35 years among Korean and Korean American men. From 18 to 35 the probability of smoking was 57% higher (95%CI, 40 to 71) among Korean men versus 8% (95%CI, 3 to 19) higher among Korean American men. A similar difference in age after 35, from 40 to 57 years of age, was associated with a 2% (95%CI, 0 to 10) and 20% (95%CI, 16 to 25) lower probability of smoking among Korean and Korean American men. A nonlinear pattern was also observed among Korean American women. Social role transitions provide plausible explanations for the decline in smoking after 35. Investigators should be mindful of nonlinearities in age when attempting to understand tobacco use.

• 한국소음진동공학회논문집
• /
• 제22권9호
• /
• pp.896-902
• /
• 2012

For most applications of the vibration energy harvesting technology as in wireless sensor networks for smart buildings and plants, the evaluation of DC output performance of vibration energy harvesters is typically required. However, there is no dedicated algorithm for the evaluation. The lack of a dedicated algorithm results from difficulties in the direct incorporation of nonlinear rectifying and regulating circuitry into finite element models of piezoelectric vibration energy harvesters. In this study, we develop a dedicated algorithm and present software based on it for the evaluation of not only AC but also DC electrical quantities. Here, an equivalent electrical circuit model is employed. The COMSOL multiphysics simulation tool is adopted for extracting equivalent electrical circuit parameters of a piezoelectric vibration energy harvester and MATLAB is used to make a graphical user interface. The AC voltage and power outputs calculated by the proposed algorithm under various conditions are compared with those by a traditional finite element analysis. The DC output voltage and power through a rectifier are obtained for varying values of smoothing capacitance and external resistance.

• 소음진동
• /
• 제5권1호
• /
• pp.75-83
• /
• 1995

The fluctuation of the engine torque appears to be the major source of the torsional vibration of the automotive driveline. The reduction of this torsional vibration has become a significant problem, due to an increase in the fluctuation of the torque of recent light weighted and high powered engines, along with the requirements of higher performance. The torsional vibration of the automotive driveline can be reduced by smoothing the fluctuation by adjusting the torsional characteristics of the clutch-disc. This paper presents an experimental and theoretical research on the clutch-disc torsional characteristics for the reduction of the torsional vibration at driving. The effects of clutch-damper on diminishing the torsional vibration were investigated experimentally. A dynamic model for the automotive driveline was developed, and the engine torque of the model were evaluated with experimental data. By executing a simulation using the model, it has become possible to obtain the clutch-disc torsional characteristics and the clutch-disc torsional characteristics for reducing the torsional vibration has been suggested. The results are as follows: (1) By exceuting simulations using nonlinear model of four degrees of freedom, a design technique to determine the clutch-disc torsional characteristics for reducing the torsional vibration at driving was developed. (2) The influence of barious torsional characteristics of the clutch has been studied in examining design parameters, which indicates that the domain to minimize the torsional vibration at driving depends on the characteristics of the clutch-damper, i. e., spring constant and hysteresis.

• 한국정보처리학회논문지
• /
• 제2권6호
• /
• pp.937-946
• /
• 1995

본 논문에서는 공간의존행렬과 신경망을 이용하여 문서영상에서 인식대상 문자가 포함되어 있는 블록들을 좀더 세분해 낼 수 있는 효과적인 방법을 제시 하였다. 제안 된 방법은 명암도 문서블록에서 공간의존행렬을 구하고 7가지 질감 특징을 추출한 뒤 신경망을 이용해 문서블록을 9가지 유형으로 분류할 수 있도록 하였다. 특히 기존에는 비문자영역으로 분류되던 수식, 도표, 순서도 등 주로 문자가 포함되어 있는 블록들을 세분해 낼 수 있도록 하였다. 또한 신경망 학습알고리즘인 BP 를 사용함으로써 기존의 선형분류시에 요구되던 유형별 임계값과 선형면결정지수를 찾는 어려움을 해소하였다. 명암도영상을 이진화하기 전에 먼저 Sobel연산을 적용함으로써 문서 뒷면에 의한 배경 잡음의 영향을 줄일 수 있도록 하였고, 교차 문지르기 후 분할함으로써 블록이 작은 조각으로 나누어지는 것을 방지하도록 하였다. 실험결과 제안한 방법에서는 문자가 포 함되어 있는 블록은 큰 문자, 중간문자, 작은 문자블록 및 수식, 순서도, 도표블록의 6가지로, 그리고 비문자블록은 인물사진, 그래프 등 3가지 유형으로 상세하게 분류 할수 堞있었으며 전체적인 분류성능도 우수함을 확인할 수 있었다.

• 한국수자원학회논문집
• /
• 제37권1호
• /
• pp.21-29
• /
• 2004

국내에서 예상되는 물부족 현상을 극복하기 위해서는 수문 현상의 이해를 통한 수자원의 안정된 확보, 관리, 개발 등 수자원 관련 기술격의 발전이 필수적이라 하겠다. 물순환계통의 올바른 이해와 적합한 모형의 개발 및 검증을 위해서는 강우 및 토양수분의 대규모 원격측정이 필수적일 뿐 아니라 관측 격자 내에서 일어나는 변화도에 대한 이해가 필요하다. 가까운 장래에 예상되는 전구 관측 토양수분자료의 격자크기인 10km는 중ㆍ소규모 지역의 수문ㆍ기상모델 적용에 한계를 가진다. 목적에 따라 각 모델들이 필요로 하는 입력 자료의 격자크기가 다른 반면 각 모델에 대한 적합한 크기의 격자를 가진 다양한 입력 자료의 부재는 토양수분자료에 대한 적합한 downscaling 기법을 필요로 한다. 사용 가능한 보조 자료와 토양수분의 선형상관관계는 상당히 낮으므로 이들 상호관계를 선형관계의 합으로 나타내는데 한계를 가진다. 그러므로 본 연구에서는 physically-based 분리기법과 자료들 간의 비선형 상관관계를 나타내는데 적합한 신경망 기법을 이용한 downscaling 기법을 개발하였다. 개발된 downscaling 기법은 Washita'92 실험으로부터 획득된 토양수분 및 보조 자료를 사용하여 4km자료를 0.2km자료로 downscaling 하였으며 출력자료는 기존의 전형적 기법에 의하여 smoothing된 자료보다 개선된 결과를 보여주었다.