• Journal of the Korean Society for Precision Engineering
• /
• v.28 no.7
• /
• pp.814-820
• /
• 2011

It takes long time in numerical simulation because structural design for tire requires the nonlinear material property. Neural networks has been widely studied to engineering design to reduce numerical computation time. The numbers of hidden layer, hidden layer neuron and training data have been considered as the structural design variables of neural networks. In application of neural networks to optimize design, there are a few studies about arrangement method of input layer neurons. To investigate the effect of input layer neuron arrangement on neural networks, the variables of tire contour design and tension in bead area were assigned to inputs and output for neural networks respectively. Design variables arrangement in input layer were determined by main effect analysis. The number of hidden layer, the number of hidden layer neuron and the number of training data and so on have been considered as the structural design variables of neural networks. In application to optimization design problem of neural networks, there are few studies about arrangement method of input layer neurons. To investigate the effect of arrangement of input neurons on neural network learning tire contour design parameters and tension in bead area were assigned to neural input and output respectively. Design variables arrangement in input layer was determined by main effect analysis.

• Transactions of Materials Processing
• /
• v.29 no.2
• /
• pp.89-96
• /
• 2020

The finite element method has been widely applied in the sheet metal forming process. However, the finite element method is computationally expensive and time consuming. In order to tackle this problem, surrogate modeling methods have been proposed. An artificial neural network (ANN) is one such surrogate model and has been well studied over the past decades. However, when it comes to ANN with two or more layers, so called deep neural networks (DNN), there is distinct a lack of research. We chose to use DNNs our surrogate model to predict the behavior of sheet metal in the deep drawing process. Thickness variation is selected as an output of the DNN in order to evaluate workpiece feasibility. Input variables of the DNN are radius of die, die corner and blank holder force. Finite element analysis was conducted to obtain data for surrogate model construction and testing. Sampling points were determined by full factorial, latin hyper cube and monte carlo methods. We investigated the performance of the DNN according to its structure, number of nodes and number of layers, then it was compared with a radial basis function surrogate model using various sampling methods and numbers. The results show that our DNN could be used as an efficient surrogate model for the deep drawing process.

• Fire Science and Engineering
• /
• v.26 no.2
• /
• pp.40-52
• /
• 2012

This paper presents the uncertainty analysis results of fire modeling input parameters for motor control center in switchgear room of nuclear power plants. FDS (Fire Dynamics simulator) 5.5 was used to simulate the fire scenario and Latin Hyper Cube Monte Carlo simulations were employed to generate random samples for FDS input parameters. The uncertainty analysis results of input parameters are compared with those of the model uncertainty analysis and sensitivity analysis approaches of NUREG-1934. The study results show that the input parameter uncertainty analysis approach may lead to more conservative results than the uncertainty analysis and sensitivity analysis methods of NUREG-1934.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.211-213
• /
• 2005

데이터 웨어하우스는 축적된 대량의 데이터를 분석하여 의사결정을 지원하는 시스템이다. 의사결정을 위한 대량의 데이터 분석은 많은 비용을 요구하므로, 질의 처리 성능을 높이고 의사 결정자에게 빠른 응답을 제공하는 효율적인 데이터 큐브 생성 기법이 연구되었다. 기존 기법으로는 Multiway Array 기법과 H-Cubing 기법이 있다. Multiway Array 기법은 다차원 집계 연산에 필요한 모든 데이터를 배열로 저장하는 것으로 데이터의 양이 많아질수록 메모리 사용이 증가한다. H-Cubing 기법은 Hyper-Tree를 기반으로 튜플을 트리로 구축하므로 모든 튜플을 트리로 구축해야 하는 비용이 증가한다. 본 논문에서는 데이터 웨어하우스에서 해쉬 테이블을 이용한 효율적인 데이터 큐브 생성 기법을 제안한다. 제안 기법은 데이터 큐브 생성 시 필드 해쉬 테이블과 레코드 해쉬 테이블을 사용한다. 필드 해쉬 테이블은 저장될 레코드 순서 계산을 위하여 각 필드에 대해 레벨 값을 해쉬 테이블로 관리한다. 레코드 해쉬 테이블은 데이터 큐브 테이블에 저장될 레코드의 순서와 데이터 큐브 테이블에 저장하기 위한 임시 레코드의 위치를 관리한다. 필드 해쉬 테이블을 이용하여 다차원 데이터의 저장될 레코드 순서를 빠르게 찾아 저장함으로서 데이터 큐브의 생성속도가 향상된다. 또한 해쉬 테이블 만을 유지하면 되므로 메모리 사용량이 감소한다. 따라서 해쉬 테이블의 사용으로 데이터의 빠른 검색과 데이터 큐브 생성 요청에 빠른 응답이 가능하다.

• The KSFM Journal of Fluid Machinery
• /
• v.20 no.1
• /
• pp.21-28
• /
• 2017

Optimization process of pre-swirl nozzle geometry was conducted to improve the discharge coefficient of pre-swirl system by using CFD. The optimization of pre-swirl nozzle shape covered the converging angle and the location of the converging nozzle. Optimization process included Optimal Latin Hyper-cube Design method to get the experimental points and the Kriging method to create the response surface which gives candidate points. The process was finished when the difference between the predicted value and CFD value of candidate point was less than 0.1 %. This paper compared the Reference model, Initial model which is the first model of optimization and Optimized model to study flow characteristics. Finally, the discharge coefficient of Optimized model is improved about 17 % to the Reference model.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.22 no.2
• /
• pp.45-54
• /
• 2018

The seismically isolated nuclear power plants shall be designed for design basis earthquake (DBE) and considered to ensure safety against beyond design basis earthquake (BDBE). In order to limit the excessive displacement of the seismic isolation system of the seismically isolated structure, the moat is installed at a certain distance from the upper mat supporting the superstructure. This certain distance is called clearance to stop (CS) and is calculated from the 90th percentile displacement of seismic isolation system subjected to BDBE. For design purposes, the CS can be obtained simply by multiplying the median displacement of the seismic isolation system against DBE by scale factor with a value of 3. The DBE and BDBE used in this study were generated by using 30 sets of artificial earthquakes corresponding to the nuclear standard design spectrum. In addition, latin hyper cube sampling was applied to generate 30 sets of artificial earthquakes corresponding to maximum - minimum spectra. For the DBE, the median displacement and the 99th percentile displacement of the seismic isolation system were calculated. For the BDBE, the suitability of the scale factor was assessed after calculating the 90th percentile displacement of the seismic isolation system.

• Korean Journal of Agricultural Science
• /
• v.47 no.3
• /
• pp.633-644
• /
• 2020

Rice paper is an element of Vietnamese cuisine that can be used to wrap vegetables and meat. Rice and starch are the main ingredients of rice paper and their mixing ratio is important for quality control. In a commercial factory, assessment of food safety and quantitative supply is a challenging issue. A rapid and non-destructive monitoring system is therefore necessary in commercial production systems to ensure the food safety of rice and starch flour for the rice paper wrap. In this study, fluorescence hyperspectral imaging technology was applied to classify grain flours. Using the 3D hyper cube of fluorescence hyperspectral imaging (fHSI, 420 - 730 nm), spectral and spatial data and chemometric methods were applied to detect and classify flours. Eight flours (rice: 4, starch: 4) were prepared and hyperspectral images were acquired in a 5 (L) × 5 (W) × 1.5 (H) cm container. Linear discriminant analysis (LDA), partial least square discriminant analysis (PLSDA), support vector machine (SVM), classification and regression tree (CART), and random forest (RF) with a few preprocessing methods (multivariate scatter correction [MSC], 1^st and 2^nd derivative and moving average) were applied to classify grain flours and the accuracy was compared using a confusion matrix (accuracy and kappa coefficient). LDA with moving average showed the highest accuracy at A = 0.9362 (K = 0.9270). 1D convolutional neural network (CNN) demonstrated a classification result of A = 0.94 and showed improved classification results between mimyeon flour (MF)1 and MF2 of 0.72 and 0.87, respectively. In this study, the potential of non-destructive detection and classification of grain flours using fHSI technology and machine learning methods was demonstrated.