• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2004년도 제5회 압연심포지엄 신 시장 개척을 위한 압연기술
• /
• pp.336-345
• /
• 2004

Investigated via a series of finite-element(FE) process simulation is the effect of diverse process variables on some selected non-dimensional parameters characterizing the thermo-mechanical behavior of the roll and strip in hot strip rolling. Then, on the basis of these parameters, on-line models are derived for the precise prediction of the temperature changes occurring in the bite zones as well as in the inter-stand zones in a finishing mill. The prediction accuracy of the proposed models is examined through comparison with predictions from a FE process model.

• 대한조선학회 특별논문집
• /
• 대한조선학회 2007년도 특별논문집
• /
• pp.22-25
• /
• 2007

Angular distortion in welding process is decided by the various variables. We can make the prediction of welding distortion in similar reality by the analysis of data through many specimen tests. However it is difficult that prediction of welding distortion applies to the ship building. We can establish that angular distortion varies directly as the specimen size. And, it makes clear distortion's difference between constraint and unconstraint, according to the change of a plate thickness.

• 한국기계가공학회지
• /
• 제18권10호
• /
• pp.42-47
• /
• 2019

Helical machining is an efficient method for machining holes using an endmill. In this study, a surface roughness prediction model was constructed for improving the productivity of hole machining. Experiments were conducted to form holes by the helical machining of AL6061-T4 aluminum sheets and correlation analysis was performed to examine the relationships between the variables based on the measured results. Meanwhile, a regression analysis technique was used to construct and evaluate the prediction model. Through these analyses, the parameter which has the greatest influence on the surface roughness when the hole is formed by the helical machining is the feed, followed by the number of revolutions of the endmill. Moreover, for the axial feed of the endmill, it was concluded that the influence of the surface roughness is small compared to the other two parameters but it is a factor worth considering to improve the accuracy when constructing the predictive model.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2022년도 춘계학술대회
• /
• pp.391-393
• /
• 2022

최근에는 딥러닝과 빅데이터를 기반으로 한 수요예측 기술이 전자상거래, 물류, 유통 분야의 스마트화를 가속화하고 있다. 특히, 글로벌 운송 네트워크와 현대적인 지능형 물류의 중심인 항만은 4차 산업혁명으로 인한 세계 경제 및 항만 환경의 변화에 발 빠르게 대응하고 있습니다. 항만물동량 예측은 신항만 건설, 항만확장, 터미널 운영 등 다양한 분야에서 중요한 영향을 담당하고 있다. 따라서 본 연구의 목적은 항만 물동량 예측에 자주 쓰이는 시계열 분석과 타 산업에서 좋은 결과를 도출해내고 있는 딥러닝 분석을 비교하여 부산항의 미래 컨테이너 예측에 적합한 예측모델을 제시하는 것이다. 부산항 컨테이너 물동량을 이용하여 학습시키고 그 이후 물동량 예측을 진행하였다. 또한, 상관관계 분석을 통해 물동량 변화와 관련된 외부변수를 선정하여 다변량 딥러닝 예측모델에 적용하였다. 그 결과 부산항 컨테이너 물동량만 이용한 단일변수 예측모델에서 LSTM의 오차가 가장 낮았고, 외부변수를 이용한 다변수 예측모델에서도 LSTM의 성능이 가장 우수하였다.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2014년도 춘계 학술논문 발표대회
• /
• pp.162-163
• /
• 2014

Operation and maintenance stage consists the largest portion of project life cycle cost. Appropriate management and analysis of such stages have massive effect on the total project cost. The effective prediction of optimized repair period is one of main factors in ㅌ management. However, it has been analyzed that the prediction of appropriate repair period revealed limitations in reliability. Therefore, this study suggests a methodology of repair period prediction by dividing finished projects into similar groups with same properties to be compared with the target project using quantitative variables.

• 한국생산제조학회지
• /
• 제19권6호
• /
• pp.804-811
• /
• 2010

The fatigue crack propagation is stochastic in nature, because the variables affecting the fatigue behavior are random and have uncertainty. Therefore, the fatigue life prediction is critical for the design and the maintenance of many structural components. In this study, fatigue experiments are conducted on the specimens of magnesium alloy AZ31 under various conditions such as thickness of specimen, the load ratio and the loading condition. The probability distribution fit to the fatigue failure life are investigated through a probability plot paper by these conditions. The probabilities of failure at various conditions are also estimated. The fatigue design life is predicted by using the Weibull distribution.

• Journal of Welding and Joining
• /
• 제31권3호
• /
• pp.44-48
• /
• 2013

Artificial neural network (ANN) model is applied to predict arc welding process window for automotive steel plate. Target weldment was various automotive steel plate combination with lap fillet joint. The accuracy of prediction was evaluated through comparison experimental result to ANN simulation. The effect of ANN variables on the accuracy is investigated such as number of hidden layers, perceptrons and transfer function type. A static back propagation model is established and tested. The result shows comparatively accurate predictability of the suggested ANN model. However, it restricts to use nonlinear transfer function instead of linear type and suggests only one single hidden layer rather than multiple ones to get better accuracy. In addition to this, obvious fact is affirmed again that the more perceptrons guarantee the better accuracy under the precondition that there are enough experimental database to train the neural network.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2021년도 가을 학술논문 발표대회
• /
• pp.86-87
• /
• 2021

As non-road mobile pollutants such as construction equipment are emerging as the main cause of air pollutants emission, construction equipment regulations are gradually strengthening. Research was conducted by correcting the emission coefficient to calculate and predict air pollutant emissions of construction equipment, but it did not reflect site variables such as field and equipment conditions that affect actual emissions. This study derived an Artificial Neural Network emission prediction model based on the actual emission data of excavators and trucks measured at the site and proposed a platform to predict the emission of air pollutants at the site according to the working size and conditions. Through this, it is possible to establish an eco-friendly process plan using a model from the construction plan.

• Communications for Statistical Applications and Methods
• /
• 제16권1호
• /
• pp.185-193
• /
• 2009

Machine learning methods such as support vector machines and random forests yield nonparametric prediction functions of the form y = $f(x_1,{\ldots},x_p)$. As a sequel to the previous article (Huh and Lee, 2008) for visualizing nonparametric functions, I propose more sensible graphs for visualizing y = $f(x_1,{\ldots},x_p)$ herein which has two clear advantages over the previous simple graphs. New graphs will show a small number of prototype curves of $f(x_1,{\ldots},x_{j-1},x_j,x_{j+1}{\ldots},x_p)$, revealing statistically plausible portion over the interval of $x_j$ which changes with ($x_1,{\ldots},x_{j-1},x_{j+1},{\ldots},x_p$). To complement the visual display, matching importance measures for each of p predictor variables are produced. The proposed graphs and importance measures are validated in simulated settings and demonstrated for an environmental study.

• 한국구조물진단유지관리공학회 논문집
• /
• 제5권2호
• /
• pp.137-146
• /
• 2001

The non-destructive tests are widely used to predict the strength of existing structures. The purpose of the present study is to propose the prediction equations for strength evaluation of concrete structures. The present study focuses on the rebound method and ultrasonic pulse velocity method for quartzite aggregate concrete. The major test variables include the water-cement ratio and curing methods. The water-cement ratio are 0.4, 0.5, 0.6, 0.7, respectively and the curing method covers ail-dry condition and standard curing condition. The prediction equations for strength of concrete are proposed from the present test data.