• 열처리공학회지
• /
• 제36권6호
• /
• pp.396-401
• /
• 2023

Aus-tempering heat treatment is suitable for thin and small-sized in precision parts. However, the heat treatment process relies on the experience and skill of the operator, making it challenging to produce precision parts due to the cold forging process. The aims of this study is to explore suitable machine learning models using data from the aus-tempering heat treatment process and analyze the factors that significantly impact the mechanic properties (e.g. hardness). As a result, the study analyzed, from a machine learning perspective, how hardness prediction varies based on the quenching temperature, carbon (C), and copper (Cu) contents.

• 산업융합연구
• /
• 제20권1호
• /
• pp.23-33
• /
• 2022

본 연구는 제조기업의 디지털 전환(DX)을 위한 실행모형을 개발하기 위해 진행하였다. 이를 위하여 경영혁신과 디지털전환 과정에 대한 선행연구를 검토하였다. 인터넷 비즈니스의 혁신과정에서 적용된 NEBIC이론과 혁신이론을 기반으로 DX과정모형을 도출하였다. 여기에 최고경영층(TMT)의 추진의지 요인을 포함한 연구모형을 구축하고 이를 검증하기 위해 험자료를 통해 확인하였다. 디지털 전환을 추진하고 있는 중견 제조기업의 구성원을 대상으로 수집된 자료를 기반으로 연구가설을 검증하였다. 회귀분석을 통해 연구모형(기술지식 → 기회인지 → 기대성과 → 실행의지)의 각 단계의 영향관계를 확인하였다. DX과정에 최고경영층의 DX추진의지에 대한 구성원의 인식의 매개효과를 파악하기 위해 위계적 회귀분석을 실시였다. Sobel test 확인결과, 각 단계별 관계에 경영층의 DX추진에 대한 인식이 부분적으로 매개하는 사실을 확인하였다. 본 연구는 중견 제조업의 디지털 전환에 적용 가능한 모형을 제시했다는 점에서 의미가 있다. 또한 혁신연구 및 NEBIC 확장에 대한 경험적 근거를 제공한 것에 가치를 지닌다. 동적특성을 지닌 과정모형에 대한 실증자료의 한계를 극복하기 위해서 종단연구가 필요하며 연구결과의 일반화를 위해 제조업 이외 다양한 분야에서 확장된 실증연구가 요구된다.

• KSBB Journal
• /
• 제8권2호
• /
• pp.178-184
• /
• 1993

Extractive fermentations with the extract of Jerusalem artichoke in an aqueous-two-phase-system of polyethyleneglycol(PEG) and dextran(Dx) were investigated to obtain the effects of composition of PEG and Dx on both fermentation ,characteristics and partition ratio of alcohol. The specific growth rate of K. Fraglis CBS 1555 increased with a decrease of concentration of PEG and Dx. It augmented along with concentration of initial sugar up to 80g/l but decreased thereafter. The specific production rate of alcohol showed a rising tendency up to 100g/lof initial sugar, whereafter represented a decreasing trend. The partition ratio of alcohol between two phases augmented according to decrease of Dx comic. and increase of PEG cone. regardless of initial sugar concentrations. The ratio, however, decreased with Increment of initial sugar concentration at constant composition of PEG and Dx. The partition coefficient of alcohol had an ascending effect to the increase of PEG cone, but it had little effect on the changes of concentrations of Dx and initial sugar. The present study suggests that the optimum composition of PEG and Dx in the aqueous-two-phase-system by extractive fermentation were around 6.5%(w/v) of PEG and 3%(w/v) of Dx in considerations of emulsion state, sedimentation and separation of two phases, alcohol partition ratio, and specific growth rate.

• 소성∙가공
• /
• 제32권6호
• /
• pp.329-334
• /
• 2023

The process of heat treatment in cold forging is an essential role in enhancing mechanical properties. However, it relies heavily on the experience and skill of individuals. The aim of this study is to predict hardness using machine learning to optimize production efficiency in cold forging manufacturing. Random Forest (RF), Gradient Boosting Regressor (GBR), Extra Trees (ET), and ADAboosting (ADA) models were utilized. In the result, the RF, GBR, and ET models show the excellent performance. However, it was observed that GBR and ET models leaned significantly towards the influence of temperature, unlike the RF model. We suggest that RF model demonstrates greater reliability in predicting hardness due to its ability to consider various variables that occur during the cold forging process.

• 산업경영시스템학회지
• /
• 제36권4호
• /
• pp.71-76
• /
• 2013

For a die casting mold, generally, the casting layout design should be considered based on the relation among injection system, casting condition, gate system, and cooling system. Also, the extent or the location of product defects was differentiated according to the various relations of the above conditions. In this research, in order to optimize the casting layout design of an automotive Oil Pan_DX2E, Computer Aided Engineering (CAE) simulation was performed with two layout designs by using the simulation software (AnyCasting). The simulation results were analyzed and compared carefully in order to apply them into the production die-casting mold. During the filling process with two models, internal porosities caused by air entrapments were predicted and also compared with the modification of the gate system and overflow. With the solidification analysis, internal porosities occurring during the solidification process were predicted and also compared with the modified gate system.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제9권8호
• /
• pp.229-234
• /
• 2020

본 논문은 임베디드 소프트웨어에 대한 Concolic 테스팅을 효과적이고 효율적으로 지원하기 위해 임베디드 타겟(target) 시스템과 호스트(host) 시스템의 분산적이고 동시적으로 테스트 생성을 위한 작업을 수행하는 Concolic 테스팅 도구의 설계와 구현 사례를 소개한다. 소개하는 테스트 케이스 생성 도구는 Concolic 테스팅 과정 중 (1) 임베디드에 종속적인 특성을 갖는 테스트검증 대상 프로그램의 실행 부분은 임베디드 타겟 시스템에서 수행하고, (2) 시스템에 비종속적인 실행 부분인 탐색 전략, 제약식 해법기 실행 과정은 계산성능이 좋은 호스트 시스템에 분산하고, 독립적인 단계를 동시적으로 실행하도록 기존 Concolic 도구를 개선하였다. Arm Cortex A54 아키텍쳐의 임베디드 타겟 시스템과 x86-64 아키텍쳐의 호스트 시스템을 대상으로 본 기법을 구현하여 오픈소스 C 프로그램의 Grep, Busybox Awk, Busybox Ed를 대상으로 실험한 결과, 기존 도구 보다 1.59~2.64배 테스트케이스 생성속도가 향상됨을 확인할 수 있었다.

• KSBB Journal
• /
• 제6권2호
• /
• pp.175-180
• /
• 1991

The quantitative effects of molecular weight and concentrations of two phase-forming polymers-polyethylene glycol and crude dextran on the two phase extractive ethanol fermentation were investigated using a Box-Wilson central composite protocol. The regression model obtained was used in order to determine optimum compositions of aqueous two phase system. In the aqueous two phase extractive ethanol fermentation of Kluyueromyces fragilis CBS 1555 with Jerusalem artichoke juice, it was found from the regression model that the variables influenlcing on ethanol fermentation were PEG concentration, time, Dx concentration, and PEG molecular weight strongly in order. The interaction of PEG concentration and PEG molecular weight was also found, and the effect of PEG concentration decreased with increase in molecular weight of PEG. The ethanol concentration incresed with increase in molecular weight of PEG, and with decrease in concentration of PEG. In conolusion, maximum concentration of ethanol produced was obtained at the following compositions; PEG MW 20000, Dx concentration ranged from 4% to 5%, and PEG concentration ranged from 3% to 7%.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2023년도 춘계학술대회
• /
• pp.204-205
• /
• 2023

본 연구는 준지도학습(SSL)을 기반한 소형 어선의 충돌 경보 송출 예측 모델에 관한 연구이다. 지도학습(SL) 방법은 레이블링된 다수의 데이터가 필요하지만 레이블링 과정에서 많은 자원과 시간이 소요된다. 본 연구는 '지능형 해상교통정보 서비스'와 연계한 데이터 파이프 라인을 통해 수집된 서비스 데이터와 실해역 시험에서 수집한 데이터를 사용하였다. 실제 사용자 만족도 기반으로 레이블이 결정된 실해역 시험 데이터만 아니라 레이블이 결정되지 않은 서비스 데이터를 함께 학습시킨 결과, 모델 정확도가 향상되었다.

• 대한수학회보
• /
• 제55권5호
• /
• pp.1463-1481
• /
• 2018

Let $X=\{X_t\}_{t{\geq}0}$ be a $L{\acute{e}}vy$ process in ${\mathbb{R}}^d$ and ${\Omega}$ be an open subset of ${\mathbb{R}}^d$ with finite Lebesgue measure. The quantity $H_{\Omega}(t)={\int_{\Omega}}{\mathbb{P}}^x(X_t{\in}{\Omega})$ dx is called the heat content. In this article we consider its generalized version $H^{\mu}_g(t)={\int_{\mathbb{R}^d}}{\mathbb{E}^xg(X_t){\mu}(dx)$, where g is a bounded function and ${\mu}$ a finite Borel measure. We study its asymptotic behaviour at zero for various classes of $L{\acute{e}}vy$ processes.

• 충청수학회지
• /
• 제23권1호
• /
• pp.37-48
• /
• 2010

In this paper we use a generalized Brownian motion process to defined an analytic operator-valued generalized Feynman integral. We then obtain explicit formulas for the analytic operatorvalued generalized Feynman integrals for functionals of the form $$F(x)=f\({\int}^T_0{\alpha}_1(t)dx(t),{\cdots},{\int}_0^T{\alpha}_n(t)dx(t)\)$$, where x is a continuous function on [0, T] and {${\alpha}_1,{\cdots},{\alpha}_n$} is an orthonormal set of functions from ($L^2_{a,b}[0,T]$, ${\parallel}{\cdot}{\parallel}_{a,b}$).