• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2007.11a
• /
• pp.345-349
• /
• 2007

In this study a fuzzy trend monitoring method for detecting the engine mechanical faults was proposed through analyzing performance trends of measurement data. The trend monitoring is an engine conditioning method which can find engine faults by monitoring important measuring parameters such as fuel flow, exhaust gas temperatures, rotational speeds, vibration. etc. Using engine condition data set as a input which generated by linear regression analysis of real engine instrument data, an application of fuzzy logic in diagnostics estimate a cause of fault in each components.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.13 no.4 s.56
• /
• pp.148-158
• /
• 2009

Firstly the problems of existing foreign code concerning wheel load distribution factor for skew box girder bridges have been examined, and the main parameters which have effects on wheel load distribution factors are evaluated in this study. Further finite element analyses on various skew steel box girder bridges are carried out. Based on the analysis results, formulas to determine wheel load distribution factors are proposed using multiple regression analysis. It is found when using the proposed formulas in this study weak points of existing specifications could be improved and also time spent at structural analysis should be saved a lot, so that the validity and practicality could be verified.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.7
• /
• pp.299-306
• /
• 2022

In recent, sensors embedded in robots, equipment, and circuits have become common, and research for diagnosing device failures by learning measured sensor data is being actively conducted. This failure diagnosis study is divided into a classification model for predicting failure situations or types and a regression model for numerically predicting failure conditions. In the case of a classification model, it simply checks the presence or absence of a failure or defect (Class), whereas a regression model has a higher learning difficulty because it has to predict one value among countless numbers. So, the reason that regression modeling is more difficult is that there are many irregular situations in which it is difficult to determine one output from a similar input when predicting by matching input and output. Therefore, in this paper, we focus on input and output data with periodicity, analyze the input/output relationship, and secure regularity between input and output data by performing sliding window-based input data patterning. In order to apply the proposed method, in this study, current and temperature data with periodicity were collected from MMC(Modular Multilevel Converter) circuit system and learning was carried out using ANN. As a result of the experiment, it was confirmed that when a window of 2% or more of one cycle was applied, performance of 97% or more of fit could be secured.

• The Korean Journal of Applied Statistics
• /
• v.10 no.2
• /
• pp.227-239
• /
• 1997

PLSR is a powerful multivariate statistical tool that has been successfully applied to the quantitative analyses of data in spectroscopy, chemistry, and industrial process control. Data in spectorscopy is represented by spectrum matrix measured in many wavelengths. Problems of many kinds of noise in data and itercorrelation between wavelengths are quite common in such data. PLSR utilizes whole data set measured in many wavelengths to the analysis, and handles such problems through data compression method. We investigated the PLSR theory, and applied this method to the data for spectroscopic diagnosis of Total Hemoglobin in whole blood.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.97-108
• /
• 2018

It has been routine practice in regression analysis to check the validity of the assumed model by the use of regression diagnostics tools. Outliers and influential observations often distort the regression output in an undesired manner. Jang and Anderson-Cook (Quality and Reliability Engineering International, 30, 1409-1425, 2014) proposed a graphical method (called a firework plot) so that there could be an exploratory visualization of the trace of the impact of the possible outliers and influential observations on individual regression coefficients and the overall residual sum of the squares measure. This paper further extends a graphical approach to a multi-response surface methodology problem.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.1373-1376
• /
• 2012

복무적합도 검사는 정신질환이나 사고가능성이 있는 병사를 감별하고, 입대 후 적응문제로 조기 전역할 수 있는 집단을 예측하는 신인성검사 중 하나로, 현재 군에서 징병 및 입영단계에 실시하는 인성검사이다. 이는 전체 검사대상자를 상대로 정신과적 문제 식별을 위한 개별면담이 불가능하기 때문에 위 검사를 통해 대상자를 효율적으로 선별하기 위함이다. 본 연구는 데이터 마이닝을 통해 복무적합도 검사의 판정을 예측 할 수 있을지 확인하고자 하였다. 이를 위해 데이터 마이닝의 기법 중 회귀분석의 로지스틱 회귀분석 기법이 복무적합도검사 판정에 우수한 성능을 보임을 확인하였고, 로지스틱 회귀분석의 추정된 회귀계수를 이용하여 만든 반응확률에 대한 예측 모형식은 높은 정분류율을 보였고 평가 결과 통계적으로 의미가 있음을 증명하였다. 따라서 본 연구 결과를 활용하면 소수의 문항으로 복무적합도 검사 이전의 선별용 검사 개발이나 자가 진단용 검사 개발로 활용이 가능 할 것으로 기대한다.

• Journal of IKEEE
• /
• v.6 no.1 s.10
• /
• pp.102-110
• /
• 2002

An effective scheme to detect and diagnose multiple failures in a dynamic system is described for the case where the measurement noise is correlated sequentially in time. It is based on the modified interacting multiple model (MIMM) estimation algorithm in which a generalized decorrelation process is developed by employing the autoregressive (AR) model for the colored noise and applying measurement difference method.

• Communications for Statistical Applications and Methods
• /
• v.15 no.1
• /
• pp.27-41
• /
• 2008

A graphical method of checking the adequacy of a generalized linear model is proposed. The graph helps to assess the assumption that the link function of mean can be expressed as a linear combination of explanatory variables in the generalized linear model. For the graph the boosting technique is applied to estimate nonparametrically the relationship between the link function of the mean and the explanatory variables, though any other nonparametric regression methods can be applied. Through simulation studies with normal and binary data, the effectiveness of the graph is demonstrated. And we list some limitations and technical details of the graph.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.13 no.1 s.53
• /
• pp.106-114
• /
• 2009

The steel design based on the width of transverse crack which is the major factor to affect a long-term performance of continuously reinforced concrete pavement was developed. For this study, twenty-one cities of Texas were selected and the temperature data was collected at those locations during the past ten years. From the data, zero-stress temperatures were calculated by the PavePro program and the widths of transverse crack were analyzed by the CRCP program. The variables used to this numerical analysis were slab thickness, coefficient of thermal expansion of concrete, steel ratio, and design temperature. The total of 448 factorial runs were made and the regression analysis was performed using the results. Steel ratios from the regression equations were backcalculated and a steel design table was proposed.

• The Korean Journal of Health Service Management
• /
• v.13 no.4
• /
• pp.133-144
• /
• 2019

Objectives: To analyze the radiologic results of patients treated with bone defects treated with beta-tricalcium phostphate (β-TCP) in orthopedic surgery. Methods: Medical records of 49 patients, consisting of 24 (49.0%) men and 25 (51.0%) women were retrospectively reviewed. Graft incorporation was evaluated based on Irwin's radiologic staging. Results: The explanatory power of the total regression equation in Irwin's stage at 3 months postoperatively according to three explanatory variables, namely graft tightness, age at diagnosis, and lesion volume, was 65.6%; the explanatory power of Irwin's stage at 6 months postoperatively according to two variables, namely lesion volume and graft tightness, was 32.9%; and the explanatory power of Irwin's stage at 9 months postoperatively for two variables, namely longitudinal lesion length and graft tightness in the total regression, was 30.8%. Conclusions: Graft tightness, lesion volume, age at diagnosis, and longitudinal lesion length are the common factors affecting graft incorporation.