• 한국신재생에너지학회:학술대회논문집
• /
• 2008.10a
• /
• pp.278-281
• /
• 2008

In these days the installation of wind turbines or wind parks includes a high financial risk. So for the planning and the constructing of wind farms, long-term data of wind speed and wind direction is required. However, in most cases only few data are available at the designated places. Traditional Measure-Correlate-Predict (MCP) can extend this data by using data of nearby meteorological stations. But also Neural Networks can create such long-term predictions. The key issue of this paper is to demonstrate the possibility and the quality of predictions using Neural Networks. Thereto this paper compares the results of different MCP Models and Neural Networks for creating long-term data with various indexes.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2007.04a
• /
• pp.211-216
• /
• 2007

The objective of this paper is to present the results of an experimental and a finite-element investigation into the behavior of X80 grade pipes subjected to bending. For the pipe specimens comprising the test series, different D/t is applied to be representative of those that can be expected in the field. Results from the numerical models are checked against the observations in the testing program and the ability of numerical solutions to predict pipe moment capacity. curvatures. and buckling modes is established. A finite-element model was developed using the finite-element simulator to predict the local buckling behavior of pipes. The comparison between the numerical and the experimental results demonstrates the ability of the analytical model to predict the local buckling behavior of pipes when deformed well into the post-yield range.

• Proceedings of the Korean Printing Society Conference
• /
• 1998.10a
• /
• pp.31-35
• /
• 1998

Spectral reflrectance of the object should be measured to predict the color of object under various illuminants. The spectral reflectance can be represented in a multidemensional space. Generally we can obtain only three-channel data from input device such as CCD camera, color scanner etc. The estimation from three dimensional to multidimension can be achieved using principal components of spectral reflectance. In this paper, A method to predict the spectral reflectance of skin color taken by 3-channel input device is discribed. To confirm this method, we simulate color represent under various illuminants about yellow, white and colored women face.

• Korean Journal of Computational Design and Engineering
• /
• v.6 no.4
• /
• pp.222-228
• /
• 2001

In the milling operation, the burr can be generated on the intersection of cutting tool and workpiece. Due to burr formation, we expect lower efficiency in the operation and the cost increase. In order to understand the burr formation mechanism in the milling operation on the arbitrary feature, we developed an algorithm to analyse and predict the exit burr formation mechanism. Firstly, the recognition of arbitrary shaped workpiece was done through the CAD data. This data includes point information on the vertices of the workpiece. Secondly, tile CAM data regarding tool geometry, tool path, cutting speed, and material data are retrieved to simulate the actual cutting process. Thirdly, we predict the exit burr formation on the edge of workpiece based on the geometric analysis. Lastly, an algorithm implemented in the Windows environment to visualize the burr formation simulation. With this information, we can predict which portion of workpiece would have the exit burr in advance so that we call manage to find a way to minimize the edit burr formation in the actual cutting.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1998.10b
• /
• pp.649-655
• /
• 1998

In this study, the personal-computer program was developed to predict prestressing losses containment structures of Nuclear Power Plants by concrete creep and shrinkage. This program is constituted of three parts, which are pre-processor, calculation module and post-processor. Input data for his program are : material properties of concrete, rebar, liner and duct, test results of concrete creep and shrinkage, relative humidity, dimension of containment structures, and the number of prestressing tendon related on containment structures. To obtain better results, this program was made to reflect the prestressing losses due to influence that occurred after prestressing each tendon, thus it can predict prestressing losses and allowable prestressing forces of each tendon. As a case study, this program was applied to containment structures of Youngwang 3 & 4 NPP's and analytical result was compared with test results in In-service Inspection of containment structures. From this comparison, it was proved that this program could well predict prestressing losses by concrete creep and shrinkage.

• Journal of Information Processing Systems
• /
• v.8 no.2
• /
• pp.241-262
• /
• 2012

An understanding of quality attributes is relevant for the software organization to deliver high software reliability. An empirical assessment of metrics to predict the quality attributes is essential in order to gain insight about the quality of software in the early phases of software development and to ensure corrective actions. In this paper, we predict a model to estimate fault proneness using Object Oriented CK metrics and QMOOD metrics. We apply one statistical method and six machine learning methods to predict the models. The proposed models are validated using dataset collected from Open Source software. The results are analyzed using Area Under the Curve (AUC) obtained from Receiver Operating Characteristics (ROC) analysis. The results show that the model predicted using the random forest and bagging methods outperformed all the other models. Hence, based on these results it is reasonable to claim that quality models have a significant relevance with Object Oriented metrics and that machine learning methods have a comparable performance with statistical methods.

• Proceedings of the International Microelectronics And Packaging Society Conference
• /
• 2002.05a
• /
• pp.117-120
• /
• 2002

In this paper, several methods to predict the solder joint shape are studied. Although there are various methods to predict the solder joint shape, such as truncated sphere method, force-bal tranced analytical solution, and energy-based methods like surface evolver developed by Ken Brakke, we calculate solder joint shape of $\mu$BGA by two solder joint shape prediction methods(truncated sphere method and surface evolver) and then compare results of each method. The results in dicate that two methods can accurately predict the solder joint shape in an accurate range. After that, we calculate reliability solder joint shape under thermal cycle test by FEA program ANSYS. As a result, it could be found that optimal solder joint shape calculated by solder joint prediction method has best reliability in thermal cycle test.

• Proceedings of the KSR Conference
• /
• 2002.10a
• /
• pp.54-59
• /
• 2002

In general, the aluminum extrusions are used to the light construction of the high speed rail vehicle structures. However, the research works ok the crashworthy design of the high speed rail vehicle structures are not published sufficiently because the crash test of high speed rail vehicle structures costs high and is complicated. So, a method that can predict crash characteristics of a large size structure like a high speed tail vehicle should be suggested. In this study, the scale model studies are performed to predict the impact energy absorption characteristics of full scale model. In the first place, we verified the theory of scale law using FE-simulation from the crashworthiness point of view. Secondly, we performed the crush test using scale model, made of aluminum sub structure. As a result, we could predict the crash characteristics using scale model by 10∼20% error.

• Journal of the Korean Geotechnical Society
• /
• v.16 no.5
• /
• pp.117-127
• /
• 2000

터널 시공과 굴착과정에서 파쇄대, 절리, 연약대, 균열 등 암반에서의 불연속면은 중요한 역할을 한다. 본 연구에서는 지반 고유의 특징인 불확실성에 의한 터널 설계와 시공 과정에서 겪는 많은 시행오차를 최소화하기 위해서 국내의 터널 붕락 현장의 지반조사 자료를 분석하여 터널 붕락 유형 및 규모를 제시할수 있는 Geo-predict 시스템을 개발하였다. Geo-predict 시스템은 총 104개 터널 붕괴/붕락자료(국외84개, 국내20개)를 분석한 자료를 테이터베이스로 인공신경망 학습을 토해서 터널 붕괴 형태와 규모를 추론하는 시스템이다. 본 논문에서는 Geo-predict의 개발과정 및 구성.기능을 소개하였으며 104개 터널 현장 자료를 지반조건별로 분석하고 이를 데이터베이스화하여 인공신경함을 이용한 추론 시스탬을 구축하고, 2개 고속전철 터널현장과 1개 지하철 시공현장에 적용성 평가를 실시하여, 터널의 붕락 가능 및 붕락 규모를 추론하였다.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2030-2032
• /
• 2001

In recent years, many attempts have been made to predict the behavior of bonds, currencies, stock, or other economic markets. Most previous experiments used multilayer perceptrons(MLP) for stock market forecasting. The Kospi 200 Index is modeled using different neural networks and fuzzy system predictions. In this paper, a multilayer perceptron architecture, a dynamic polynomial neural network(DPNN) and a fuzzy system are used to predict the Kospi 200 index. The results of prediction is compared with the root mean squared error(RMSE) and the scatter plot. Results show that both networks can be trained to predict the index. And the fuzzy system is performing slightly better than DPNN and MLP.