• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.20 no.7
• /
• pp.2289-2297
• /
• 1996

Flows in the combustion chamber near the spark plug are measured using LDv.A single cylinder DOHC S.I. engine of compression ratio 9.5:1 with a transparent quartz window piston is used. Combustion chamber shape is semi-wedge type. Measured data are analyzed using the ensemble averaged analysis and the cycle resolved analysis which uses FFT Filtering. Turbulent intensity and mean velocity are studied in the main flow direction and the normal to main flow direction as a function of engine speeds. The results shows that the turbulent intensity obtained by the ensemble averaged analysis is greater than that calculated by the cycle resolved analysis. Especially, the ensemble averaged analysis shows increase in turbulence at the end of compression stroke although the cycle resolved analysis shows increase only in the cycle-by-cycle variation with no noticeable increase in turbulence. The mean velocity in the main flow direction increase as engine speed increase. But the mean velocity normal to the main flow does not show such increase. Turbulent intensity in both direction increase in proportion to engine speeds. The magnitude of turbulent intensity is about 0.3 ~ 0.4 times the mean piston speeds at the end of the compression stroke.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.4
• /
• pp.912-919
• /
• 1994

The engine combustion is one of the most important processes affecting performance and emissions. One effective way to improve the engine combustion is to control the motion of the charge inside a cylinder by means of optimum induction system design, because the flame speed is mainly determined by the turbulence at compression(TDC) process in S.I. engine. It is believed that the tumble and swirl motion generated during intake stroke breaks down into small-scale turbulence in the compression stroke of the cycle. However, the exact nature of this relationship is not well known. This paper describes the tumble flow measurements inside the cylinder of a 4-valve S.I. engine using laser Doppler velocimetry(LDV) under motoring(non-firing) conditions. This is conducted on an optically assesed single cylinder research engine under motored conditions at an engine speed of 1000rpm. Three different cylinder head intake port configurations are studied to develop a better understanding the tumble flow generation, development, and breakdown mechanisms.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.217-217
• /
• 2023

상수도관망은 사용자에게 고품질의 물을 안정적으로 공급하는 것을 목적으로 하며, 이를 평가하기 위한 지표 중 하나로 압력을 활용한다. 최근 스마트 센서의 설치가 확장됨에 따라 기계학습기법을 이용한 실시간 데이터 기반의 분석이 활발하다. 따라서 어디에서 데이터를 수집하느냐에 대한 센서 위치 결정이 중요하다. 본 연구는 eXtreme Gradient Boosting(XGBoost) 모델을 활용하여 대규모 상수도관망 내 센서 위치를 최적화하는 방법론을 제안한다. XGBoost 모델은 여러 의사결정 나무(decision tree)를 활용하는 앙상블(ensemble) 모델이며, 오차에 따른 가중치를 부여하여 성능을 향상시키는 부스팅(boosting) 방식을 이용한다. 이는 분산 및 병렬 처리가 가능해 메모리리소스를 최적으로 사용하고, 학습 속도가 빠르며 결측치에 대한 전처리 과정을 모델 내에 포함하고 있다는 장점이 있다. 모델 구현을 위한 독립 변수 결정을 위해 압력 데이터의 변동성 및 평균압력 값을 고려하여 상수도관망을 대표하는 중요 절점(critical node)를 선정한다. 중요 절점의 압력 값을 예측하는 XGBoost 모델을 구축하고 모델의 성능과 요인 중요도(feature importance) 값을 고려하여 센서의 최적 위치를 선정한다. 이러한 방법론을 기반으로 상수도관망의 특성에 따른 경향성을 파악하기 위해 다양한 형태(예를 들어, 망형, 가지형)와 구성 절점의 수를 변화시키며 결과를 분석한다. 본 연구에서 구축한 XGBoost 모델은 추가적인 전처리 과정을 최소화하며 대규모 관망에 간편하게 사용할 수 있어 추후 다양한 입출력 데이터의 조합을 통해 센서 위치 외에도 상수도관망에서의 성능 최적화에 활용할 수 있을 것으로 기대한다.

• Journal of Intelligence and Information Systems
• /
• v.15 no.3
• /
• pp.1-15
• /
• 2009

In a classification problem, data imbalance occurs when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. This paper proposes a Geometric Mean-based Boosting (GM-Boost) to resolve the problem of data imbalance. Since GM-Boost introduces the notion of geometric mean, it can perform learning process considering both majority and minority sides, and reinforce the learning on misclassified data. An empirical study with bankruptcy prediction on Korea companies shows that GM-Boost has the higher classification accuracy than previous methods including Under-sampling, Over-Sampling, and AdaBoost, used in imbalanced data and robust learning performance regardless of the degree of data imbalance.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.156-156
• /
• 2016

2016년 2월 5일, 짐바브웨는 극심한 가뭄으로 인해 인구의 4분의 1이상이 식량난을 겪고 있다며 '국가 재난 사태'를 선포하였다. 한때 아프리카 곡창지대로 불리던 짐바브웨가 극심한 가뭄을 겪게 된 데에는 2015/16년 슈퍼엘니뇨의 영향이 크게 한 몫을 하였는데, 이는 남반구의 여름인 11월부터 이듬해 3월까지인 짐바브웨의 우기가 2015/16년 슈퍼엘니뇨 강도가 절정에 달했던 시기(10월에서 2월)와 겹쳐져 짐바브웨의 강수량이 슈퍼 엘니뇨의 영향을 받게 되었기 때문이다. 게다가 4월부터는 엘니뇨의 영향을 받은 우기가 끝나고 건기가 시작되기 때문에 앞으로 가뭄이 얼마나 더 악화될지 우려되는 상황이다. 짐바브웨의 기후를 살펴보면, 증발량이 강수량보다 많은 건조기후 중에서도 비교적 그 정도가 약한 기후인 반건조 지대에 속한다. 하지만 연강수량 변동에 따라서, 비가 내리는 해에는 토양 수분이 과잉되고 비가 적게 내리는 해에는 심한 물 부족 현상이 일어나게 되기 때문에, 건기가 시작되는 4월부터 짐바브웨 강수 예측은 가뭄이 얼마나 지속될지를 파악하는 데에 아주 중요한 요소가 될 수 있다. 따라서 본 연구에서는 강수 예측 결과를 중심으로 2016년 짐바브웨의 가뭄이 얼마나 지속되고, 또 가뭄의 강도는 어떻게 될지 알아보는 것에 목적을 두고, GCM을 이용하여 2016년 3월에서 10월까지 장기예측을 수행하였다. 경계 자료로는 ECMWF (European Centre for Medium Range Weather Forecasts)에서 제공하는 Sea Ice자료와, NOAA OI (National Oceanic and Atmospheric Administration Optimum Interpolation) Weekly SST자료를 사용하였고 엘니뇨의 영향을 고려하기 위해 IRI (International Research Institute)의 ENSO forecast를 참고하여 SST아노말리에 월별 가중치를 적용하였다. 초기 입력 자료로는 1월 21-30일 10일간의 ECMWF의 재분석 자료를 이용하여 총 10개 멤버의 앙상블 예측을 수행하였고, 8개월(3-10월) 기간에 대해 약 한 달간의 spin-up time을 주었다. 예측 자료를 모델 climatology와 비교하여 월 평균 강수 전망을 분석하였고, 기온과 해면기압의 월 평균자료도 추가 분석하였다. 또한 짐바브웨 지역의 강수 관측 자료와 모델 예측 자료를 이용하여 특정 도시들의 1년 누적강수를 예측 및 분석하였고, 최종적으로 이 결과를 통해 짐바브웨의 가뭄지속가능성을 살펴보았다.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.5
• /
• pp.713-721
• /
• 2001

Characteristics of the conical vortices on the roof corner of a rectangular prism have been investigated by using a PIV(Particle Image Velocimetry) technique. The Reynolds number based on the free stream velocity and the height of the model was 5.3$\times$10$^3$. The mean, instantaneous velocity vector fields, vorticity fields, and turbulent kinetic energy distribution were measured for two different angles of attack, 30$^{\circ}$and 45$^{\circ}$. The PIV measurements clearly observed not only the conical main vortex and the secondary vortex but also the tertiary vortex which is firstly reported in this paper. Asymmetric formation of the corner vortex for the case of 30$^{\circ}$angle of attack produces relatively the high magnitude of vorticity and turbulent kinetic energy around the bigger vortex which generates the peak suction pressure on the roof. Fairly symmetric features of the roof vortex are observed in the case of 45$^{\circ}$angle of attack, however, the dynamic characteristics are proved to be asymmetric due to the rectangular shape of the roof.

• Journal of the Korean earth science society
• /
• v.28 no.2
• /
• pp.214-226
• /
• 2007

Because precipitation is influenced by various atmospheric variables, it is highly nonlinear. Although precipitation predicted by a dynamic model can be corrected by using a nonlinear Artificial Neural Network, this approach has limits such as choices of the initial weight, local minima and the number of neurons, etc. In the present paper, we correct simulated precipitation by using a multiple linear regression (MLR) method, which is simple and widely used. First of all, Ensemble hindcast is conducted by the PNU/CME Coupled General Circulation Model (CGCM) (Park and Ahn, 2004) for the period from April to August in 1979-2005. MLR is applied to precipitation simulated by PNU/CME CGCM for the months of June (lead 2), July (lead 3), August (lead 4) and seasonal mean JJA (from June to August) of the Northeast Asian region including the Korean Peninsula $(110^{\circ}-145^{\circ}E,\;25-55^{\circ}N)$. We build the MLR model using a linear relationship between observed precipitation and the hindcasted results from the PNU/CME CGCM. The predictor variables selected from CGCM are precipitation, 500 hPa vertical velocity, 200 hPa divergence, surface air temperature and others. After performing a leave-oneout cross validation, the results are compared with the PNU/CME CGCM's. The results including Heidke skill scores demonstrate that the MLR corrected results have better forecasts than the direct CGCM result for rainfall.

• Tunnel and Underground Space
• /
• v.32 no.6
• /
• pp.502-517
• /
• 2022

As the use of TBM increases, research has recently increased to to analyze TBM data with machine learning techniques to predict the exchange cycle of disc cutters, and predict the advance rate of TBM. In this study, a regression prediction of disc cutte wear of slurry shield TBM site was made by combining machine learning based on the machine data and the geotechnical data obtained during the excavation. The data were divided into 7:3 for training and testing the prediction of disc cutter wear, and the hyper-parameters are optimized by cross-validated grid-search over a parameter grid. As a result, gradient boosting based on the ensemble model showed good performance with a determination coefficient of 0.852 and a root-mean-square-error of 3.111 and especially excellent results in fit times along with learning performance. Based on the results, it is judged that the suitability of the prediction model using data including mechanical data and geotechnical information is high. In addition, research is needed to increase the diversity of ground conditions and the amount of disc cutter data.