• Electrical & Electronic Materials
• /
• v.11 no.11
• /
• pp.1-8
• /
• 1998

This paper presents an arificial neuro-fuzzy technique based prtial discharge (PD) pattern classifier to power system application. This may require a complicated analysis method employ -ing an experts system due to very complex progressing discharge form under exter-nal stress. After referring briefly to the developments of artificical neural network based PD measurements, the paper outlines how the introduction of new emerging technology has resulted in the design of a number of PD diagnostic systems for practical applicaton of residual lifetime prediction. The appropriate PD data base structure and selection of learning data size of PD pattern based on fractal dimentsional and 3-D PD-normalization, extraction of relevant characteristic fea-ture of PD recognition are discussed. Some practical aspects encountered with unknown stress in the neuro-fuzzy techniques based real time PD recognition are also addressed.

• Journal of Korean Association for Spatial Structures
• /
• v.7 no.2 s.24
• /
• pp.115-124
• /
• 2007

The initial shape is arrived at by a self-formation process, which accomplishes a form in the natural world, or is determined analytically by considering the equilibrium of initial stress only. Therefore, the self-formation process, which accomplishes a form in the natural world is grasped and the types of modeling techniques available to find the shapes of soft structures are well investigated and classified. To establish a form-finding modeling techniques, the models of string, soap film, fabric, rubber, plaster, and etc. are used. These modeling techniques can be used as a method of understanding the characteristics of structures when the material of model shows similar characteristics. Generally, the model test confirms the structure based on numerical analysis, at the same time it is important preceding process to develop such a program. With the above process, the relationship between model test and numerical analysis becomes a feedback process. Therefore, in this study, two examples which have been accomplished from such a technique are investigated and considered according to modeling process.

• Journal of Korean Society of Steel Construction
• /
• v.10 no.1 s.34
• /
• pp.63-72
• /
• 1998

The stiffness design method is presented as a drift control model of steel structures and applied to design of space trusses subjected to stress and displacement constraints. The stiffness design method is developed by integrating the resizing techniques for an effective drift control algorithm with the strength design process according to the commonly used design specifications such as allowable stress design. In the resizing technique the amount of material to be modified depends on the member displacement participation factors and is determined by an optimization technique. Using the stiffness design method, a structural design model for steel structures is proposed and applied to two verifying examples. As demonstrated in the examples, the displacement of the structures can be effectively controlled without expensive computational cost.

• Steel and Composite Structures
• /
• v.18 no.4
• /
• pp.909-924
• /
• 2015

This paper investigates the vibration phenomenon of a nanobeam subjected to a time-dependent heat flux. Material properties of the nanobeam are assumed to be graded in the thickness direction according to a novel exponential distribution law in terms of the volume fractions of the metal and ceramic constituents. The upper surface of the functionally graded (FG) nanobeam is pure ceramic whereas the lower surface is pure metal. A nonlocal generalized thermoelasticity theory with dual-phase-lag (DPL) model is used to solve this problem. The theories of coupled thermoelasticity, generalized thermoelasticity with one relaxation time, and without energy dissipation can extracted as limited and special cases of the present model. An analytical technique based on Laplace transform is used to calculate the variation of deflection and temperature. The inverse of Laplace transforms are computed numerically using Fourier expansion techniques. The effects of the phase-lags (PLs), nonlocal parameter and the angular frequency of oscillation of the heat flux on the lateral vibration, the temperature, and the axial displacement of the nanobeam are studied.

• Korean Management Science Review
• /
• v.10 no.1
• /
• pp.193-205
• /
• 1993

This paper presents a mathematical model for a class of vertical integration decisions. The problem structure of interest consists of raw material vendors, components suppliers, components processing plants, final product (assembly) plants and external components buyers. Economic feasibility of operating components plants instead of keeping outside suppliers is our major concern. The model also determines assignment of product lines and production volumes to each open plant considering the cost impacts of economies of scale and plant complexity. The problem formulation leads to a concave, mixed integer mathematical program. Given the state of the art of nonlinear programming techniques, it is often not possible to find global optima for reasonably sized such problems. We developed an optimization solution algorithm within the framework of Benders decomposition for the case of a piecewise linear concave cost function. It is shown that our algorithm generates optimal solutions efficiently.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.09a
• /
• pp.194-205
• /
• 2009

In geotechnical engineering, the mechanical characteristics of soil, the main material of geotechnical engineering, is highly related to the confining stress. Reduced-scale physical modeling is often conducted to evaluate the performance or to verify the behavior of the geotechnical systems. However, reduced-scale physical modeling cannot replicate the behavior of the full-scale prototype because the reduced-scale causes difference of self weight stress level. Geotechnical centrifuges are commonly used for physical model tests to compensate the model for the stress level. Physical modeling techniques using centrifuge are widely adopted in most of geotechnical engineering fields these days due to its various advantages. In this paper, fundamentals of geotechnical centrifuge modeling and its application area are explained. State-of-the-art geotechnical centrifuge equipment is also described as an example of KOCED geotechnical centrifuge facility at KAIST.

• Nuclear Engineering and Technology
• /
• v.54 no.9
• /
• pp.3390-3397
• /
• 2022

Among the different techniques available, nuclear methods, including gamma-gamma logging tools, are of special importance. Though the real environment which surrounds the drilled borehole is a complex fractured medium which the fluid can flow through the porosities, simulation studies generally use the traditional model of a homogeneous mixture of formation and the liquid. Considering a previously published study, which shows that modeling of fluid flow in fractured reservoirs and simulating the formation as an inhomogeneous fractured medium leads to different results compared with those of homogeneous mixture, here we study the effect of the presence of drilling fluid (mudcake) on the response of the detectors in both the models. To study this effect, a typical gamma-gamma logging tool was modeled by using the MCNPX Monte Carlo code. The results show that the responses of the detectors in the mixture model in the presence of various thicknesses of mudcake are sensitive to the density of the formation material. However, this effect is not notable in the inhomogeneous fractured medium. These results emphasize the importance of the model employed for simulation of the medium in gamma-gamma well-logging.

• The Journal of Advanced Prosthodontics
• /
• v.9 no.4
• /
• pp.287-293
• /
• 2017

PURPOSE. To compare the dimensional accuracy of three impression techniques- a separating foil impression, a custom tray impression, and a stock tray impression. MATERIALS AND METHODS. A machined mandibular complete-arch metal model with special modifications served as a master cast. Three different impression techniques (n = 6 in each group) were performed with addition-cured silicon materials: i) putty-wash technique with a prefabricated metal tray (MET) using putty and regular body, ii) single-phase impression with custom tray (CUS) using regular body material, and iii) two-stage technique with stock metal tray (SEP) using putty with a separating foil and regular body material. All impressions were poured with epoxy resin. Six different distances (four intra-abutment and two inter-abutment distances) were gauged on the metal master model and on the casts with a microscope in combination with calibrated measuring software. The differences of the evaluated distances between the reference and the three test groups were calculated and expressed as mean (${\pm}SD$). Additionally, the 95% confidence intervals were calculated and significant differences between the experimental groups were assumed when confidence intervals did not overlap. RESULTS. Dimensional changes compared to reference values varied between -74.01 and $32.57{\mu}m$ (MET), -78.86 and 30.84 (CUS), and between -92.20 and 30.98 (SEP). For the intra-abutment distances, no significant differences among the experimental groups were detected. CUS showed a significantly higher dimensional accuracy for the inter-abutment distances with -0.02 and -0.08 percentage deviation compared to MET and SEP. CONCLUSION. The separation foil technique is a simple alternative to the custom tray technique for single tooth restorations, while limitations may exist for extended restorations with multiple abutment teeth.

• Journal of information and communication convergence engineering
• /
• v.21 no.3
• /
• pp.208-215
• /
• 2023

Deep learning techniques provide powerful solutions to several pattern-recognition problems, including Raman spectral classification. However, these networks require large amounts of labeled data to perform well. Labeled data, which are typically obtained in a laboratory, can potentially be alleviated by data augmentation. This study investigated various data augmentation techniques and applied multiple deep learning methods to Raman spectral classification. Raman spectra yield fingerprint-like information about chemical compositions, but are prone to noise when the particles of the material are small. Five augmentation models were investigated to build robust deep learning classifiers: weighted sums of spectral signals, imitated chemical backgrounds, extended multiplicative signal augmentation, and generated Gaussian and Poisson-distributed noise. We compared the performance of nine state-of-the-art convolutional neural networks with all the augmentation techniques. The LeNet5 models with background noise augmentation yielded the highest accuracy when tested on real-world Raman spectral classification at 88.33% accuracy. A class activation map of the model was generated to provide a qualitative observation of the results.

• Journal of Computational Design and Engineering
• /
• v.3 no.1
• /
• pp.63-70
• /
• 2016

Deep drawing is a forming process in which a blank of sheet metal is radially drawn into a forming die by the mechanical action of a punch and converted to required shape. Deep drawing involves complex material flow conditions and force distributions. Radial drawing stresses and tangential compressive stresses are induced in flange region due to the material retention property. These compressive stresses result in wrinkling phenomenon in flange region. Normally blank holder is applied for restricting wrinkles. Tensile stresses in radial direction initiate thinning in the wall region of cup. The thinning results into cracking or fracture. The finite element method is widely applied worldwide to simulate the deep drawing process. For real-life simulations of deep drawing process an accurate numerical model, as well as an accurate description of material behavior and contact conditions, is necessary. The finite element method is a powerful tool to predict material thinning deformations before prototypes are made. The proposed innovative methodology combines two techniques for prediction and optimization of thinning in automotive sealing cover. Taguchi design of experiments and analysis of variance has been applied to analyze the influencing process parameters on Thinning. Mathematical relations have been developed to correlate input process parameters and Thinning. Optimization problem has been formulated for thinning and Genetic Algorithm has been applied for optimization. Experimental validation of results proves the applicability of newly proposed approach. The optimized component when manufactured is observed to be safe, no thinning or fracture is observed.