• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.3
• /
• pp.175-185
• /
• 2020

The six-DOF aerodynamic coefficients of a missile entail inherent physical gradient constraints originated from the geometric characteristics of a cylindrical fuselage. To effectively adopt the freely available gradient information in aerodynamic coefficients modeling, this research employed gradient-enhanced (GE) Gaussian process. To investigate the accuracy of aerodynamic coefficients predicted with gradients information, we compared two Gaussian-process-based models: ordinary and GE Gaussian process models with and without gradient information, respectively. As a result, we found that GE Gaussian process models were able to comply with imposed gradient information and more accurate than ordinary Gaussian process models. However, we also found that GE Gaussian process modeling cannot handle gradient information continuously and ends up with more samples due to additional gradient information.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.10 no.2
• /
• pp.93-99
• /
• 1998

Parabolic approximation and sponge layer are applied as open boundary condition for elliptic finite difference wave model. Generalized conjugate gradient method is used as a solution procedure. Using parabolic approximation a large part of spurious reflection is removed at the spherical shoal experiment and sponge layer boundary condition needs more than 2 wave lengths of sponge layer to give similar results. Simulating the propagation of waves on a rectangular harbor, it is identified that iterative scheme can be applied easily for the non-rectangular computational region.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.392-395
• /
• 2017

본 연구에서는 일 증발접시 증발량 산정을 위한 딥러닝 (deep learning) 모형의 적용성을 평가하였다. 본 연구에서 적용된 딥러닝 모형은 deep belief network (DBN) 기반 deep neural network (DNN) (DBN-DNN) 모형이다. 모형 적용성 평가를 위하여 부산 관측소에서 측정된 기상자료를 활용하였으며, 증발량과의 상관성이 높은 기상변수들 (일사량, 일조시간, 평균지상온도, 최대기온)의 조합을 고려하여 입력변수집합 (Set 1, Set 2, Set 3)별 모형을 구축하였다. DBN-DNN 모형의 성능은 통계학적 모형성능 평가지표 (coefficient of efficiency, CE; coefficient of determination, $r^2$; root mean square error, RMSE; mean absolute error, MAE)를 이용하여 평가되었으며, 기존의 두가지 형태의 ANN (artificial neural network), 즉 모형학습 시 SGD (stochastic gradient descent) 및 GD (gradient descent)를 각각 적용한 ANN-SGD 및 ANN-GD 모형과 비교하였다. 효과적인 모형학습을 위하여 각 모형의 초매개변수들은 GA (genetic algorithm)를 이용하여 최적화하였다. 그 결과, Set 1에 대하여 ANN-GD1 모형, Set 2에 대하여 DBN-DNN2 모형, Set 3에 대하여 DBN-DNN3 모형이 가장 우수한 모형 성능을 나타내는 것으로 분석되었다. 비록 비교 모형들 사이의 모형성능이 큰 차이를 보이지는 않았으나, 모든 입력집합에 대하여 DBN-DNN3, DBN-DNN2, ANN-SGD3 순으로 모형 효율성이 우수한 것으로 나타났다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.4
• /
• pp.48-56
• /
• 2014

Even though vehicle types, gradient, pavement conditions and types of pavement should be considered for estimating fuel consumption, existing models were developed as a function of vehicle types and vehicle speed. Therefore in this study, the model of fuel consumption was developed using field test data in order that effect analysis on the passenger vehicle fuel consumption by road gradient. At first, fuel consumption was measured in second-based, using GPS device and fuel consumption measurement device for development of fuel consumption model considered road gradient. The road gradient was classified as flatland, up-hill and down-hill. Development of model was using by regression model which vehicle speed(km/h) and fuel consumption(${\ell}/km$). The on-road test proved that fuel consumption of passenger vehicle is affected by road gradient.

• Journal of Korean Society of Transportation
• /
• v.24 no.4 s.90
• /
• pp.149-159
• /
• 2006

Traditional trip tables are estimated through large-scale surveys such as household survey, roadside interviews, and license Plate matching. These methods are, however, expensive and time consuming. This paper presents two origin-destination (OD) trip matrix estimation methods from link traffic counts in stochastic assignment, which contains perceived errors of drivers for alternatives. The methods are formulated based on the relation between link flows and OD demands in logit formula. The first method can be expressed to minimize the difference between observed link flows and estimated flows, derived from traffic assignment and be solved by gradient method. The second method can be formulated based on dynamic process, which nay describe the daily movement patterns of drivers and be solved by a recursive equation. A numerical example is used for assessing the methods, and shows the performances and properties of the models.

• Journal of Korean Society of Transportation
• /
• v.18 no.2
• /
• pp.41-52
• /
• 2000

In the developing country, the transportation situation is changed very quickly and the transportation environment is not stable. So the transportation planning should be frequently made in considering the limited cost and time. And the traditional large-scale survey(household survey, roadside interview, etc.) has many Problem like the difficulty for doing it and getting mood results. Therefore the study about the method of evaluation on the traffic count based O/D matrix is Processing actively recently. Though the many study for the network in the realistic size are enacted, the study for comparing with the advantage and disadvantage of each method are few. Therefore this study mainly deals with the static method among the existing models of evaluation on the traffic count based O/D matrix(in terms of the transportation plan). Bi-level(GU) and gradient method are selected as main alternative model and analyzed their capability and validity. For testing the reliability of the models, Bi-level(GLS) and gradient method are adapted to toy network. Then we analyze the result of testing, and study the way for large network.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.11
• /
• pp.53-63
• /
• 2023

This study developed a cognitive impairment predictive model as one of the screening tests for preventing dementia in the elderly by using Automated Machine Learning(AutoML). We used 'Wearable lifelog data for high-risk dementia patients' of National Information Society Agency, then conducted using PyCaret 3.0.0 in the Google Colaboratory environment. This study analysis steps are as follows; first, selecting five models demonstrating excellent classification performance for the model development and lifelog data analysis. Next, using ensemble learning to integrate these models and assess their performance. It was found that Voting Classifier, Gradient Boosting Classifier, Extreme Gradient Boosting, Light Gradient Boosting Machine, Extra Trees Classifier, and Random Forest Classifier model showed high predictive performance in that order. This study findings, furthermore, emphasized on the the crucial importance of 'Average respiration per minute during sleep' and 'Average heart rate per minute during sleep' as the most critical feature variables for accurate predictions. Finally, these study results suggest that consideration of the possibility of using machine learning and lifelog as a means to more effectively manage and prevent cognitive impairment in the elderly.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.11
• /
• pp.1399-1409
• /
• 1999

It is known that previous models are unsatisfactory in predicting adverse pressure gradient turbulent flows. In the present paper, a revised low Reynolds number $k-{\varepsilon}$ model is proposed. In this model, a newly developed term is added lo the dissipation rate equation. In order to reflect appropriate effects for an adverse pressure gradient. The added tenn is derived by considering the distribution of mean velocity and turbulent properties in the turbulent flow with, adverse pressure gradient. The new $k-{\varepsilon}$ model was applied to calculations of flat plate flow with adverse pressure gradient, conical diffuser flow and backward facing step flow. It was found that the three numerical results showed better agreement than other models compared with DNS results and experimental ones.

• Journal of Korean Society of Transportation
• /
• v.27 no.4
• /
• pp.145-154
• /
• 2009

The main goal of this paper is to develop a methodology for increasing the reliability of truck OD matrices in Seoul Metropolitan Area. We propose a Hybrid Method made up of five processes by using Non-Traffic Assignment and Gradient Method. A Hybrid Method and Gradient Method have applied for comparison and estimation in Seoul Metropolitan Area. Mean Error and Root Mean Square Error of a Hybrid Method present lower than Gradient Method. The findings of this paper show that the new truck OD matrices created by a Hybrid Method are more reliable than the existing truck OD matrices in the Seoul Metropolitan Area.

• Journal of Korean Society of Transportation
• /
• v.23 no.3 s.81
• /
• pp.61-71
• /
• 2005

We evaluated the availability of Origin-Destination Matrix from traffic counts Using conjugate gradient method to large scale networks by applying it to the networks in 246 zones. As a result of the analysis of the consistency of the model on Nationwide Networks, the upper and lower levels in model had the systematic relationship internally. From the analysis of the estimable power or the model according to the number of traffic counting links, the error in traffic volume had the estimable power in the range of permissible error. In addition, the estimable power of estimation of an Origin-Destination Matrix was more satisfactory than that of existing methods. We conclude that conjugate gradient method cab be applied to nationwide networks if we can make sure that the algorithm of the developed model is reliable by doing various kinds of experiment.