Determination of the best distribution and effective interval using statistical characterization of uncertain variables

Joo, Minho,Doh, Jaehyeok,Lee, Jongsoo Society for Computational Design and Engineering 2018 Journal of computational design and engineering Vol.5 No.3

In this paper, an algorithm for estimating the best distribution about data containing uncertainties is proposed. The proposed algorithm combines sequence statistical modeling (SSM) and a method for determining the minimum experimental data. SSM is a method for selecting the best distribution about data using a goodness of fit (GoF) test and a comparison of the likelihood. The method used to determine the minimum experimental data determines the minimum data needed to an estimate the best distribution. The SSM presented herein is a method for selecting a suitable data distribution when considering only a parametric distribution. Thus, in this paper, the SSM was improved in order to select the correct distribution of both parametric and non-parametric distributions simultaneously. In addition, with the existing method for determining the minimum data, the data should be added based on actual experiments when the results data show an insufficient number, and there is a limitation in that the designers cannot broadly identify the data required. To overcome this limitation, SSM and random sampling are applied to the method to determine the minimum data, and thereby, ensure that the designer knows the approximate minimum data needed. To verify the validity of the proposed algorithm, it was applied to a real world case study on determining multiple statistical parameters in the bolt fastening problem. The sequence of verification methods used is as follows: First, the best distribution of the bearing surface and thread friction coefficient estimated by the proposed algorithm and based on a normal distribution are selected as comparison targets. Second, the bearing surface and thread friction coefficient data are sampled within the 95% confidence interval of the two distributions. Third, the reliability of the sampled friction coefficient data are compared using a Monte-Carlo simulation and an equation to calculate the bolt fastening force. In this study, the effectiveness of the proposed algorithm is validated.

Determination of the best distribution and effective interval using statistical characterization of uncertain variables

Minho Joo,Jaehyeok Doh,Jongsoo Lee 한국CDE학회 2018 Journal of computational design and engineering Vol.5 No.3

In this paper, an algorithm for estimating the best distribution about data containing uncertainties is pro-posed. The proposed algorithm combines sequence statistical modeling (SSM) and a method for deter-mining the minimum experimental data. SSM is a method for selecting the best distribution about data using a goodness of fit (GoF) test and a comparison of the likelihood. The method used to determine the minimum experimental data determines the minimum data needed to an estimate the best distribu-tion. The SSM presented herein is a method for selecting a suitable data distribution when considering only a parametric distribution. Thus, in this paper, the SSM was improved in order to select the correct distribution of both parametric and non-parametric distributions simultaneously. In addition, with the existing method for determining the minimum data, the data should be added based on actual experi-ments when the results data show an insufficient number, and there is a limitation in that the designers cannot broadly identify the data required. To overcome this limitation, SSM and random sampling are applied to the method to determine the minimum data, and thereby, ensure that the designer knows the approximate minimum data needed. To verify the validity of the proposed algorithm, it was applied to a real world case study on determining multiple statistical parameters in the bolt fastening problem. The sequence of verification methods used is as follows: First, the best distribution of the bearing surface and thread friction coefficient estimated by the proposed algorithm and based on a normal distribution are selected as comparison targets. Second, the bearing surface and thread friction coefficient data are sampled within the 95% confidence interval of the two distributions. Third, the reliability of the sampled friction coefficient data are compared using a Monte-Carlo simulation and an equation to calculate the bolt fastening force. In this study, the effectiveness of the proposed algorithm is validated.

마이크로 유전알고리즘의 최적설계 응용에 관한 연구

이종수(Jongsoo Lee),김종헌(Jonghun Kim),이형주(Hyung-Joo Lee),구본흥(Bon-Heung Koo) 대한기계학회 2002 대한기계학회 춘추학술대회 Vol.2002 No.5

The paper describes the development and application of advanced evolutionary computing techniques referred to as micro genetic algorithms (μGA) in the context of engineering design optimization. The basic concept behind μGA draws from the use of small size of population irrespective of the bit string length in the representation of design variable. Such strategies also demonstrate the faster convergence capability and more savings in computational resource requirements than simple genetic algorithms (SGA). The paper first explores ten-bar truss design problems to see the optimization performance between ,μGA and SGA. Subsequently, μGA is applied to a realistic engineering design problem in the injection molding process optimization.

NOx 및 SOOT 저감분석을 위한 역전파 신경망을 이용한 디젤 엔진 배기시스템의 근사 모델링

이승주(Seung Joo Lee),박재인(Jae In Park),이수홍(Soo Hong Lee),이준규(Joon Kyu Lee),이종수(Jongsoo Lee) 한국자동차공학회 2011 한국자동차공학회 부문종합 학술대회 Vol.2011 No.5

Recently, the research for reducing harmful automotive emissions have been recognized seriously. Particularly, the improvement for reducing harmful automotive emissions depends on diesel engine combustion characteristics such as EGR rate, Swirl ratio, Start of Injection and etc. In this paper, the examination and simulation data are extracted from the diesel engine combustion characteristics. Based on these results, the amount of emissions such as NOx, SOOT, and etc are anticipated by performing approximations on back-progagation neural networks which have benefit nonlinear characteristic prediction. Consequently, the possibility of predicted approximation based on the characteristic condition of diesel engine combustion is examined.

물리적 불확실성을 내재한 입력변수의 확률 통계 기반 유효 범위 결정 방법 및 신뢰성 평가

주민호(Minho Joo),도재혁(Jaehyeok Doh),최수교(Sukyo Choi),이종수(Jongsoo Lee) 대한기계학회 2017 大韓機械學會論文集A Vol.41 No.11

동일한 시험조건에서 반복시험으로부터 얻어진 실험 데이터는 이론적으로 동일한 값을 가져야 한다. 그러나 실제 데이터 결과는 다양한 환경 요소들에 의해 발생하는 오차와 불확실성을 가지게 되어 시험 값이 변동량을 가진다. 이는 정확한 실험 데이터를 얻는데 제한사항이 된다. 본 연구에서는 확률통계 방법을 이용하여 불확실성을 가진 입력변수의 유효범위를 결정하는 알고리즘을 제안하였다. 또한 실제 현장에서 사용되는 볼트 체결 마찰계수 데이터를 이용하여 제안된 알고리즘을 적용하여 불확실성을 내재한 입력변수의 유효범위를 산출하고 이에 대한 신뢰성 평가를 하였다. Data points obtained by conducting repetitive experiments under identical environmental conditions are, theoretically, required to correspond. However, experimental data often display variations due to generated errors or noise resulting from various factors and inherent uncertainties. In this study, an algorithm aiming to determine valid bounds of input variables, representing uncertainties, was developed using probabilistic and statistical methods. Furthermore, a reliability assessment was performed to verify and validate applications of this algorithm using boltfastening friction coefficient data in a sample application.

기계학습기반 분류 알고리즘을 통한 금속 탐지 드론 개발

주민호(Minho Joo),이종수(Jongsoo Lee),윤재현(Jaehyun Lee) 대한기계학회 2018 대한기계학회 춘추학술대회 Vol.2018 No.4

개선된 순차적 통계모델링 기법을 이용한 차량용 고무 물성의 확률 분포 추정 연구

주민호(Minho Joo),도재혁(Jaehyeok Doh),장용석(Yongsok Jang),장홍석(Hongsok Jang),박종찬(Jongchan Park),이종수(Jongsoo Lee) 대한기계학회 2017 대한기계학회 춘추학술대회 Vol.2017 No.5

터보코딩 및 고차변조를 적용하는 3GPP GERAN 진화 시스템

오형주(Hyeong-Joo Oh),최병조(Byoungjo Choi),황승훈(Seung-Hoon Hwang),최종수(Jongsoo Choi) 한국통신학회 2008 韓國通信學會論文誌 Vol.33 No.6A

본 논문에서는 터보코딩 및 고차변조를 적용하는 3GPP GERAN 진화 시스템 물리계층에 비트 신뢰도 기반의 심볼 매핑 방법(Symbol Mapping based on Priority: SMP)을 최초로 적용하여 그 성능을 살펴보고자 한다. SMP는 터보코딩에서 정보비트(systematic bits)의 중요성을 이용하여 정보비트를 신뢰도 높은 비트에 매핑시키는 기술로서 본 기술을 GERAN 진화 시스템에 적용시켰을 때, 링크 레벨 시뮬레이션을 통해 16QAM(DAS-8)과 32QAM(DAS-11)의 변조/코딩 방식에 대해서 성능이득이 있음을 확인할 수 있다. In this paper, we investigate the performance of SMP-assisted 3GPP GERAN evolution system employing high order modulation and turbo coding. When applying the SMP which maps systematic bits into highly reliable bit positions, it is confirmed that there is the performance gain for the modulation and coding schemes of 16QAM(DAS-8) as well as 32QAM(DAS-11) by link level simulation.

지뢰탐지를 위한 기계학습 모델을 이용한 금속 종류 분류 방법

주민호(Minho Joo),이종수(Jongsoo Lee) 대한기계학회 2017 대한기계학회 춘추학술대회 Vol.2017 No.11

A number of landmines buried around the world are substantial, and the damage is also occurring every year. In Korea, there is more than one occurrence of damage caused by landmines per year. To detect and remove these mines, the human uses metal detectors directly or unmanned robots in areas where suspected of mined land. However, not only landmines but also various metals can be buried along the mine, so it is difficult to determine whether a detected metal is a landmine or normal metal. In this study, the algorithm to determine whether a detected metal is landmine or normal metal was proposed using support vector machine (SVM) which is most commonly used for classification of machine learning models. Also, the effectiveness of this algorithm was verified through real experiments. For this purpose, electric current measurements sensor was mounted on a commercial metal detector for print out a changed current to numerical data when metal was detected. The comparison target of the experiment to verify the algorithm is a generic beverage can and a 9V square battery. The noise from the data produced by the sensors filtered out using filters. Through this, the SVM was learned in advance through the each 40 training data of the beverage can and battery, and the algorithm determined it whether the detected metal is the beverage can or battery when the metal detector detected the metal. If the detected data of a mine apply to the SVM, the proposed algorithm can determine that the detected metal is landmine or not.

전층각막이식술 후 발생한 고안압증 환자에서 이식 실패로 이행하는 위험인자

주종수,이유경,주천기,Jongsoo Joo,You Kyung Lee,Choun-Ki Joo 대한안과학회 2012 대한안과학회지 Vol.53 No.3

Purpose: To assess the risk factors proceeding to graft failure in post-keratoplasty ocular hypertension patients. Methods: In 35 eyes diagnosed with post-keratoplasty ocular hypertension (graft failure: 13 eyes; graft survival: 22 eyes), relationships between graft status at the observation time and pre-keratoplasty diagnosis, lens status, history of graft failure, donor size, difference between donor and recipient graft size, donor corneal endothelial cell count, post-keratoplasty intraocular pressure (after 1 week and maintenance intraocular pressure after surgery), and number of antiglaucomatic agents were investigated. The relative risks of each factor to induce graft failure were also evaluated. Results: Previous graft failure history, pre-existing pseudophakic bullous keratopathy and aphakia/pseudophakia showed statistically significant high probabilities of proceeding to graft failure (p < 0.05). In particular, the intraocular pressure 1 week after the graft was statistically higher (p < 0.05) in the graft failure group (24.31 ± 8.82 mm Hg) than in the graft survival group (16.81 ± 6.69 mm Hg). Conclusions: Strict management of intraocular pressure in the early phase of penetrating keratoplasty could contribute to reducing graft failure in post-keratoplasty ocular hypertension patients. J Korean Ophthalmol Soc 2012;53(3):385-389