Efficient Correlation Method for Satellite Thermal Analysis Model Using Multiple Linear Regression and Optimization Algorithms

Jaewon Kang,김건웅,Somin Shin,김정호 한국항공우주학회 2023 International Journal of Aeronautical and Space Sc Vol.24 No.5

As the thermal analysis model of satellites is used as an important indicator for thermal design, it must accurately simulate the thermal behaviour of actual satellites for precise thermal design. To increase the accuracy of the thermal analysis model, it must be correlated using the thermal balance test data for actual satellite models. To achieve this, we herein propose an efficient correlation method for satellite thermal analysis models using multiple linear regression techniques with quadratic terms and optimization algorithms. The proposed method reduces the amount of computation by choosing dominant parameters through sensitivity analysis and creating a multiple linear regression model that can replace the thermal analysis model in the subsequent optimization process. Subsequently, optimization algorithms are applied to the multiple linear regression model to perform the correlation of the thermal analysis model. In this study, the numerical validation of the proposed method was performed using numerical data from a reference thermal analysis model to verify the reliability and accuracy of the proposed method before it was applied to the correlation of the thermal analysis model using experimental data. The thermal analysis result of the reference thermal analysis model was set as the target value to correlate, and quantitative performance evaluation was performed for various combinations of optimization algorithms and design of experiments methods by comparing the estimated analysis parameters. The results of this study demonstrate that the proposed method can efficiently produce an accurate correlation model for thermal analysis.

A Model on Credit Analysis: Combining A First Passage Model and Survival Analysis for Corporate Default

조승모 예금보험공사 2013 金融安定硏究 Vol.14 No.1

In this paper, we offer a new set of credit analysis models combining two traditional approaches of corporate default prediction: the survival analysis approach and the structural model approach. We first derive a modified version of the Black and Cox (1976) first passage model (a structural model) and estimate its model parameters based on the Miyake and Inoue (2009) method originally applied to parameter estimation for the Merton (1974) model(another structural model). And then based on the previous literature and various statistical techniques, we select some significant factors affecting corporate default and construct various survival analysis regression models with those factors including or excluding our first passage model as an independent variable. And by comparing our new combined models (the survival analysis regression models including the first passage model as an independent variable) with their corresponding plain models (the survival analysis regression models excluding the first passage model as an independent variable), we conclude that our new combined models are superior to their counterparts. And finally, by comparing those superior models, we propose a single best-fitting combined model for corporate default: an accelerated failure time model by Kalbfleisch and Prentice (1980) under the Weibull distribution with our modified Black and Cox (1976) model as an independent variable.

공간 이용자의 이동빈도를 이용한 행위자 기반 공간 네트워크 분석모델 연구

권지훈(Kweon, Jihoon) 한국디지털디자인협의회 2015 디지털디자인학연구 Vol.15 No.2

본 연구는 이용자의 이동빈도를 고려하여 행위자 기반 공간 네트워크 분석모델을 제안하기 위해 수행되었다. 현재까지 공간 구문론이나 가시성 그래프 분석 이론에 기반을 둔 공간분석 모델은 개별 공간 간의 균질적인 상호 가시성을 기술하는 이원적 링크를 고려하는 한계를 보여 왔다. 행위자 기반 모델을 사용하는 선행 분석모델의 연구들은 공간 네트워크의 개념을 고려하지 않는 또 다른 한계를 가져 결과적으로 이들이 부적합한 행위자의 이동을 허용하였고, 물리적 공간에서 이용자의 이동을 수용하지 않는 결과의 생산을 야기하였다. 본 연구에서 제시된 공간 네트워크의 분석을 위한 행위자 기반 모델은 공간 네트워크와 행위자 기반 모델을 그 구현함에 있어 개념 및 분석방법에서 병합하였다. 이 연구는 요소 공간들 간의 구성적 관계를 재현하는 공간 네크워크 모델과 그 모델 상에서 구현된 이용자의 행위자 기반 모델을 사용하였다. 결과적으로 이들 두 모델의 병합은 제안된 분석모델이 건축공간에서 이용자의 이동에 대한 시공간적 특성을 고려하게 하였다. 제시된 모델이 공간에서 다중 사용자의 이동을 고려하는 분석적 연구에 이용되기를 기대한다. This study aimed to suggest an agent-based model for spatial network analysis considering user"s movement frequency. The existing space analysis models based on Space Syntax theory and Visibility Graph Analysis theory have shown limits which considered binal links representing homogeneous inter-visibility between individual spaces. The precedent studies using agent-based models had also another limit neglecting the concept of spatial network so that they allowed inappropriate movements of agents and caused the production of analysis results not affording user"s movements in physical spaces. The proposed agent-based model for spatial network analysis combined the spatial network and agent-based model in concept and analysis method for its implementation. This study used a spatial network model representing configurational conditions between component spaces and an agent-based user model implemented on the spatial network model. As a result, the combination of these two models make the proposed analysis model consider spatio-temporal property of user"s movement in architectural space. The suggest analysis model is expected to be used for the analytic studies considering multiple users" movements in spaces.

핵연료집합체 3 차원 축소모델 개발 및 모드해석

무하마드 수반(Muhammad Subhan),남궁인(Ihn Namgung) 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.4

APR1400 원전의 지진해석은 집중질량 빔요소로 구성된 2 차원 모델로서 수직 방향과 수평 방향 모델로 분리되어 있다. 원자로 코어의 모든 핵연료는 모두 합쳐져서 몇 개의 빔요소와 집중질량으로 모델이 이루어져 있다. AP1000의 경우에는 3 차원 집중질량과 빔요소를 사용하여 모델을 개발하였으며, 3 차원 지진해석을 수행하였다. 본연구의 목적은 노심의 핵연료집합체를 통합하지 않고 개별 핵연료집합체에 대한 축소 동해석 모델을 개발하여 노심 동해석을 가능하게 하는 것이다. 이를 위해 본 논문에서는 최신 컴퓨터 기술 및 모델링 기술을 접목해서 핵연료집합체의 3 차원 동해석 방법론을 개발하는 것이다. 3 차원 핵연료 집합체의 축소모델을 개발하기 위해 핵연료봉은 질량으로 나타냈고, CEA 가이드 튜브와 스페이서 그리드는 빔요소를 사용하였고, 하단지지체와 상단지지체는 쉘요소로 취급하여 전체적으로 축소된 모델을 개발하였다. 이렇게 개발된 축소모델은 2,740 개의 요소와 5,148 개의 요소를 가진다. 이 모델에 대해 모드해석을 하여 시험결과와 비교를 통해 1 차모드를 수렴하도록 하였다. 모드해석을 통해 2 차, 3 차 및 4 차 모드도 시험 결과와 유사한 경향을 보임을 확인하였다. 따라서 본 축소모델을 사용하여 지진해석 및 시간이력해석이 가능할 것으로 판단된다. 본 연구에서 모드해석 결과 기존의 통합 모델에서 관찰할 수 없던 트위스트 모드를 보였다. 이 결과는 핵연료사이의 접촉가능성을 보여주는 것으로서 핵연료 사이의 동적 거동에 대한 추가적인 연구가 필요하다. Currently, the seismic analysis model of APR 1400 consists of beam and lumped mass in 2D separated model in horizontal direction and vertical direction. Also in the core, fuel assemblies are aggregated and simplified into few beam and lumped masses. WEC also used beam and lumped mass elements for the AP1000 seismic analysis model, but used 3D beam and mass to represent whole core where fuel assemblies are aggregated as well. The main purpose of the lumped mass and beam modeling of fuel assembly is to simplify model and make core dynamic analysis possible without aggregating all fuels in core into few beam elements and lumped masses. This research investigates 3D dynamic analysis modeling methodology of nuclear fuel assembly reflecting performance improvement of computer technology and analysis software functionality. A 3D modeling approach of FA was proposed in which fuel rods mass was lumped, CEA guide tubes and spacer grid are represented by beam elements and lower end-fitting and upper end-fitting are represented by shell elements. The FA dynamic analysis model so developed consists of 2,740 elements and 5,148 nodes. The model was optimized for the 1st mode from the test results of FA applying modal analysis. The model was successfully represents 1st mode modal values. Also the modal analysis results exhibits similarities for 2nd, 3rd and 4th modes, hence the model can be used for further dynamic analysis of fuel assembly such as seismic analysis or time-history analysis. The corresponding mode shapes revealed additional mode of twist motion that was not shown by current FA dynamic model of APR1400 or AP1000. This research revealed there are extra modes of twist motion that could induce contact between FAs and would require further investigation of dynamic interaction between FAs.

경영분석지표와 의사결정나무기법을 이용한 유상증자 예측모형 개발

김명균(Myeong-Kyun Kim),조윤호(Yoonho Cho) 한국지능정보시스템학회 2012 지능정보연구 Vol.18 No.4

This study focuses on predicting which firms will increase capital by issuing new stocks in the near future. Many stakeholders, including banks, credit rating agencies and investors, performs a variety of analyses for firms’ growth, profitability, stability, activity, productivity, etc., and regularly report the firms’ financial analysis indices. In the paper, we develop predictive models for rights issues using these financial analysis indices and data mining techniques. This study approaches to building the predictive models from the perspective of two different analyses. The first is the analysis period. We divide the analysis period into before and after the IMF financial crisis, and examine whether there is the difference between the two periods. The second is the prediction time. In order to predict when firms increase capital by issuing new stocks, the prediction time is categorized as one year, two years and three years later. Therefore Total six prediction models are developed and analyzed. In this paper, we employ the decision tree technique to build the prediction models for rights issues. The decision tree is the most widely used prediction method which builds decision trees to label or categorize cases into a set of known classes. In contrast to neural networks, logistic regression and SVM, decision tree techniques are well suited for high-dimensional applications and have strong explanation capabilities. There are well-known decision tree induction algorithms such as CHAID, CART, QUEST, C5.0, etc. Among them, we use C5.0 algorithm which is the most recently developed algorithm and yields performance better than other algorithms. We obtained data for the rights issue and financial analysis from TS2000 of Korea Listed Companies Association. A record of financial analysis data is consisted of 89 variables which include 9 growth indices, 30 profitability indices, 23 stability indices, 6 activity indices and 8 productivity indices. For the model building and test, we used 10,925 financial analysis data of total 658 listed firms. PASW Modeler 13 was used to build C5.0 decision trees for the six prediction models. Total 84 variables among financial analysis data are selected as the input variables of each model, and the rights issue status (issued or not issued) is defined as the output variable. To develop prediction models using C5.0 node (Node Options: Output type = Rule set, Use boosting = false, Cross-validate = false, Mode = Simple, Favor = Generality), we used 60% of data for model building and 40% of data for model test. The results of experimental analysis show that the prediction accuracies of data after the IMF financial crisis (59.04% to 60.43%) are about 10 percent higher than ones before IMF financial crisis (68.78% to 71.41%). These results indicate that since the IMF financial crisis, the reliability of financial analysis indices has increased and the firm intention of rights issue has been more obvious. The experiment results also show that the stability-related indices have a major impact on conducting rights issue in the case of short-term prediction. On the other hand, the long-term prediction of conducting rights issue is affected by financial analysis indices on profitability, stability, activity and productivity. All the prediction models include the industry code as one of significant variables. This means that companies in different types of industries show their different types of patterns for rights issue. We conclude that it is desirable for stakeholders to take into account stability-related indices and more various financial analysis indices for short-term prediction and long-term prediction, respectively. The current study has several limitations. First, we need to compare the differences in accuracy by using different data mining techniques such as neural networks, logistic regression and SVM. Second, we are required to develop and to evaluate new prediction model

기술 교과 교육에서 문제 중심 프로그램의 교재 내용과 체제의 내적 분석을 위한 모형 개발 및 타당화

김동하,최유현 한국기술교육학회 2009 한국기술교육학회지 Vol.9 No.1

The purpose of this study was to develop and validate model for internal analysis for workbook of problem-based program and its system in technology education. Based upon relevant literature review, I came up with theoretical model for internal analysis for Workbook of problem-based program and internal analysis of system in technology education. The feasibility of the model was validated by professionals’ decision. And using the analysis model to analyze workbooks of problem-based program in technology education and its system, I was able to secure the validity of the model. The results of this study were as follows: To develop the model for internal analysis for workbook of problem-based program and its system in technology education, I categorized 63 sub items of the internal analysis model. I also divided the analysis items into 2 analysis categories. Based upon the study result, I would like to suggest following ideas for follow-up studies. First, I developed analysis model focusing on internal analysis for workbook of problem-based program and its system in technology education. Therefore, it is necessary to conduct a study on analyzing a model for workbooks in technology education and external system of workbooks. Second, this study was focused on selection and categorization of analysis items rather than analysis methodology or system. Thus, there is limit for suggested analysis categories and items to become materialized instructions of methodology and system. Therefore, it is necessary to conduct a study for methodological instructions in the future. 이 연구의 목적은 기술 교과 교육에서 문제 중심 프로그램의 교재 내용과 체제의 내적 분석을 위한 모형을 개발하고, 이를 타당화 하는데 있다. 이러한 연구의 목적을 달성하기 위하여 문헌 고찰에 근거하여 기술 교과 교육에서 문제 중심 프로그램의 교재 내용과 체제의 내적 분석을 위한 이론적 모형을 구안하였다. 이론적 모형은 전문가로 구성된 집단의 전문적 판단에 근거하여 그 타당도를 검증하였으며, 도출된 분석 모형을 사용하여 기술 교과 교육에서 실제 활용되고 있는 문제 중심 프로그램의 교재 내용과 체제를 분석해 봄으로써, 모형의 타당도를 확보하였다. 기술 교과 교육에서 문제 중심 프로그램의 교재 내용과 체제의 내적 분석을 위한 모형을 개발하기 위하여 문헌 연구를 통해 분석 모형의 하위 요소 63개가 추출하였고, 이를 성격에 따라 11개의 분석 항목으로 유목화 하였으며, 항목은 2개의 분석 영역으로 분류하였다. 기술 교과 교육에서 사용되어지는 교재, 교과서에 대한 외적 조직 체제를 분석하는 모형에 대한 연구가 필요하다. 또한, 제시된 분석의 영역, 항목, 요소 들이 어떠한 방법과 절차를 통하여 분석해야 할지에 대한 구체적인 지침이 되기에는 한계가 있기에 보다 구체적으로 ‘어떻게 분석해야 하는가’에 대한 방법론적 지침을 제공할 수 있는 연구가 앞으로 필요하다.

측정오차를 포함한 자료의 모형기반 판별분석

송주원 한국자료분석학회 2012 Journal of the Korean Data Analysis Society Vol.14 No.6

Discriminant analysis is a statistical technique that finds a discriminant function based on various characteristics and assigns each observation to an appropriate group. Model-based cluster analysis is a clustering technique based on finite mixture models and can be applied to discriminant analysis. Fraley and Raftery (2002) assumes that each group can be represented by a mixture distribution of several distributions and suggest a model-based mixture discriminant analysis (MclustDA) using model-based cluster analysis. When some observations are measured with errors, observations with and without measurement errors may follow different distributions with different parameter values, and each group may consist of a mixture of more than one distributions. In this study, we conduct a simulation to compare misclassification rates among the standard discriminant analysis, mixture discriminant analysis assuming more than one distributions in each group, and discriminant analysis based on model-based cluster analysis. Misclassification rates tends to increase when the percentage of observations measured with errors increases. Mixture discriminant analysis and discriminant analysis based on cluster analysis show lower misclassification rates than the standard discriminant analysis. 판별분석은 측정된 변수들의 특성에 근거하여 집단을 판별하는 방법을 찾아내고 새로운 개체들을 적절한 집단에 할당하는 것을 목적으로 하는 통계적 기법이다. 모형기반 군집분석은 유한개의 혼합모형에 근거하여 군집분석을 실시하는 기법으로서 판별분석으로도 적용이 가능하다. Fraley, Raftery(2002)는 각 군집에서의 자료가 한 개의 분포를 따른다고 가정하는 대신 각 군집이 여러 개의 분포의 혼합분포로 표현될 수 있다고 가정하고 모형기반 군집분석을 이용한 모형기반 혼합판별분석(MclustDA)을 제안하였다. 일부 관측 개체가 부정확하게 측정된 경우 오차를 포함한 자료는 정확하게 측정된 자료와 다른 모수를 가진 분포를 따를 수 있으므로 각 집단은 한 개의 분포 대신 한 개 이상의 분포로 구성된다고 볼 수 있다. 본 연구에서는 일부 개체가 오차를 포함한 자료에서 각 집단에 대하여 한 개 이상의 모형의 혼합분포를 가정하는 혼합판별분석이나 모형기반 군집분석을 이용한 판별분석을 실시한다면 기대 오분류율을 낮출 수 있는지 모의실험을 통해 비교하였다. 오차를 가지고 측정된 자료의 비율이 증가할수록 부정확한 자료로 인하여 오분류율이 증가할 수 있으며 혼합판별분석이나 모형기반 군집분석을 이용한 판별분석이 각 집단이 한 개의 분포로 이루어졌다고 가정하는 판별분석보다 오분류율이 작게 나타났다.

Analysis of Purchase History Data Based on a New Latent Class Model for RFM Analysis

Qian Zhang,Haruka Yamashita,Kenta Mikawa,Masayuki Goto 대한산업공학회 2020 Industrial Engineeering & Management Systems Vol.19 No.2

Recently, it has become easier to make use of various kinds of information on customers (e.g. customers’ purchase history), due to the development of information technology. Especially in the marketing field, in fact, many compa-nies try to employ customer segmentation for the services customization which leads to increase customer loyalty and to keep high customer retention. One of the well-known approaches for the customer analysis based on purchase history data is the RFM analysis. The RFM analysis is usually used to segment customers into several groups by using three variables; how long it has been since their last purchase, how many times they purchased, and how much they spent. However, the conventional method of the RFM analysis did not assume a generative model. Therefore, when applying to an actual data set and scoring each index of R, F, M scores, several problems occur. The main problem is that an analyst should arbitrarily decide the threshold for the scores of RFM. On the other hand, in the field of ma-chine learning, the probabilistic latent semantic analysis is widely used for soft clustering. The latent class model ena-bles us to cluster customers into latent classes and to calculate the assignment probabilities of each customer to each latent class. In this paper, we propose a new latent class model for the RFM analysis based on the purchase history data. The proposed model enables to decide the scoring of RFM and segment customers automatically, and the soft clustering approach helps the interpretation of the result. Furthermore, the proposed model takes account of the generation model of RFM scores. From the result of actual data analysis, it became clear that it is possible to extract latent classes that express the statistical characteristics of data well. Given a generative model estimated from the given data, it is also possible to predict future purchase behaviors of customers or to generate virtual data for simulation analysis and make decisions based on the result. We verify the effectiveness of our model by analyzing a real purchase history data of a Japanese major retail company.

한의학 연구동향 분석시스템 구현을 위한 모형개발

예상준(Sang-Jun Yea),장현철(Hyun Chul Jang),김진현(JinHyun Kim),김철(Chul Kim),김상균(Sang-Kyun Kim),송미영(Mi-Young Song) 한국콘텐츠학회 2009 한국콘텐츠학회논문지 Vol.9 No.12

연구동향을 분석하기 위해서 일반적으로 연구자와 기획자는 특허맵/논문맵 분석을 이용하고 있으나, 한의학계의 특수성으로 기존 시스템을 이용할 수 없는 상황이다. 그래서 선행연구와 선행시스템을 파악하여 적용 가능한 분석모델을 도출하여 한의학 관련 연구동향을 체계적으로 분석할 수 있는 기본분석, 상세분석, 복합분석의 14개 모델을 설계하였다. 약 16,000건의 한의학 논문이 수록되어 있는 오아시스 데이터베이스에 ‘경락’의 키워드를 사용하여 103건의 논문을 검색하였다. 추출된 논문에 제안된 분석모델을 적용하여 한의학의 경락 연구는 성숙기에 접어들고 있으며 생리학 등 타 분야와 밀접한 관련이 있는 결과를 얻었으며, 선행 시스템의 분석모델과의 비교를 통해서 제안된 분석모델을 검증하였다. 개발된 분석모델은 한의학 논문맵 분석 시스템을 위한 서비스 시나리오 개발 및 시스템 구현에 이용될 수 있을 것으로 기대된다. The researchers and planners are using patent/paper map system to analyze the research trend, but we can't use the existing analysis system because of the specialty of korean traditional medicine. Thus we deduced the analysis models from preceding research and system and designed 14 analysis models which are composed of basic, detail and complex models. We verified the analysis models using papers which has 'meridian' keyword among 16,000 papers stored in OASIS. From the analysis result, we know that the meridian study has just entered in the maturity and is closely related with other parts of korean traditional medicine as physiology etc. And we verified the proposed analysis model from the comparison with the analysis models of preceding systems. The analysis model will be used for the development of Korean traditional medicine paper map analysis service scenario and system.

SWOT Analysis and the Development of an Evaluation Model for Electronic Trade Settlement (TSU/BPO)

주혜영(Hye Young Joo),유병부(Byoung Boo You) 중앙대학교 한국전자무역연구소 2018 전자무역연구 Vol.16 No.3

연구목적: 본 연구는 TSU/BPO 방식의 활용가능성을 높이기 위한 전략적 방안을 모색하는 것을 주요한 목적으로 한다. 이를 위해 SWOT 분석을 실시하여 TSU/BPO의 주요당사자별로 내부환경 및 외부환경의 강점과 약점, 기회와 위협요인을 도출하였다. 또한 이를 기초로 실증 분석을 위한 연구모델을 구축하였다. 논문구성/논리: 제1장 서론에서는 연구의 목적과 배경을 기술하였고 제2장에서는 TSU/BPO 제도의 특징을 살펴보았다. 제3장에서는 SWOT 분석을 실시하였고 이에 기초하여 실증분석을 위한 연구모델을 개발하였다. 제4장의 결론에서는 주요한 연구내용의 요약과 함께 시사점이 논의되었다. 결과: SWOT 분석을 통해 TSU/BPO 결제방식을 입체적으로 조망하였고 SWOT 분석결과와 기술수용모델을 결합하여 실증적 평가모델을 구축하였다. 구체적으로, TSU/BPO의 유용성, TSU/BPO의 편의성, TSU/BPO 제도의 안정성 및 TSU/BPO로의 전환비용에 따라 TSU/BPO의 활용가능성이 영향을 받는 것으로 설계되었다. 독창성/가치: TSU/BPO와 관련된 선행연구들이 거의 대부분 문헌연구를 통해 TSU/BPO에 대한 이해를 도모하고 있는데 비해 본 연구는 SWOT 분석을 통해 TSU/BPO제도를 입체적으로 조망하고 있으며 추후 TSU/BPO 활용에 관해 실증적 평가가 가능한 연구모델을 제공하고 있다는 점에서 연구의 독창성 뿐 아니라 연구의 가치가 인정된다. Purpose: The main purpose of this study is to find strategic ways to increase the utilization of the TSU/BPO method. To do this, SWOT analysis was conducted to derive the strengths, weaknesses, opportunities, and threats of internal and external environments for each major TSU/BPO party. Based on this, we constructed a research model for empirical analysis. Composition/Logic: This paper starts with background information and presents the goals of the study in Section I 1. In Section II, we discuss the characteristics of the TSU/BPO system. In Section III, SWOT analysis is conducted and a research model for empirical analysis is developed based on the SWOT analysis. In the conclusions of Section IV, implications are discussed along with a summary of the main research contents. Findings: In this study, SWOT analysis is used to view the TSU/BPO settlement method from various angles, and an empirical evaluation model is constructed by combining the SWOT analysis results with the technology acceptance model. Specifically, the model shows that the utilization of TSU/BPO is affected by its usefulness, convenience, institutional stability, and conversion costs. Originality/Value: Although previous research related to the TSU/BPO research area has mostly aimed at understanding TSU/BPO through literature reviews, this study takes an original view of the TSU/BPO system through SWOT analysis. The value of this study, as well as the originality of the research, is recognized in that it provides an empirical research model that can evaluate TSU/BPO possibilities for practical use.