• Journal of Advanced Marine Engineering and Technology
• /
• 제31권6호
• /
• pp.681-686
• /
• 2007

The flow analysis was made by applying the turbulent models in the scroll volume of centrifugal compressor. The $k-{\varepsilon}.\;k-{\omega}$, Spalart-Allmaras and reynolds stress models are used in which the hybrid grid is applied for the simulation. The velocity vector the Pressure contour. the change of residual along the iteration number. and the dynamic head are simulated by solving the Navier-Stokes equations for the comparison of four example cases.

• 천문학회보
• /
• 제36권2호
• /
• pp.58.2-58.2
• /
• 2011

We present smoothed particle hydrodynamic models of the interactions in the compact galaxy group, Stephan's Quintet. Adding thermohydrodynamic effects to the earlier collisionless N-body simulations of Renaud et al. (2010), we further investigate the dynamical interaction history and evolution of the intergalactic gas of Stephan's Quintet. Specifically, we model the formation of the hot X-ray gas, the group-wide shock, and emission line gas as the result of NGC 7318b colliding with the group as well as reproduce the tidal structures in the group. We compare our model results to multi-wavelength observations.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2000년도 춘계공동학술대회 논문집
• /
• pp.289-289
• /
• 2000

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2007년도 하계종합학술대회 논문집
• /
• pp.307-308
• /
• 2007

This paper proposes a discriminative training algorithm for estimating hidden Markov model (HMM) parameters. The proposed algorithm estimates the Parameters by minimizing the p-norm of log-likelihood difference (PLD) between the utterance probability given the correct transcription and the most competitive transcription.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 1993년도 제5회 한글 및 한국어정보처리 학술대회
• /
• pp.149-157
• /
• 1993

음성인식을 위해 다양한 방법들이 제안되어 있다. 본 연구에서는 음소단위 각각의 벡터 양자화된 코드북의 색인을 학습하는 HMM을 이용하여 한국어 숫자음을 대상으로 인식 실험을 수행하였다. 실험결과, 기존의 단어단위 HMM과 음소단위로 이루어진 유한상태기계(FSM)구조의 인식기에 비해 높은 인식율을 보였다.

• 지능정보연구
• /
• 제1권1호
• /
• pp.23-41
• /
• 1995

Business researchers have traditionally used statistical techniques for classification. In late 1980's, inductive learning started to be used for business classification. Recently, neural network began to be a, pp.ied for business classification. This study reviews the business classification studies, identifies a neural network a, pp.oach as the most powerful classification tool, and discusses the problems and issues in neural network a, pp.ications.

• 한국에너지공학회:학술대회논문집
• /
• 한국에너지공학회 1994년도 추계학술발표회 초록집
• /
• pp.146-149
• /
• 1994

A nonlinear dynamic optimization has been performed for reactor power control system of CANDU 6 nuclear power plant considering xenon, fuel and moderator temperature feedback effects. Integral-of-Time-multiplied Absolute-Error (ITAE) criterion has been used as a performance index of the system behavior. Optimum controller gain are found by searching algorithm of Sequential Quadratic Programming (SQP). System models are referenced from most recent literatures. Signal flow network construction and optimization have been done by using commercial computer software package.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2021년도 학술발표회
• /
• pp.131-131
• /
• 2021

Deep learning models, especially those based on long short-term memory (LSTM), have presented their superiority in addressing time series data issues recently. This study aims to comprehensively evaluate the performance of deep learning models that belong to the supervised learning category in streamflow prediction. Therefore, six deep learning models-standard LSTM, standard gated recurrent unit (GRU), stacked LSTM, bidirectional LSTM (BiLSTM), feed-forward neural network (FFNN), and convolutional neural network (CNN) models-were of interest in this study. The Red River system, one of the largest river basins in Vietnam, was adopted as a case study. In addition, deep learning models were designed to forecast flowrate for one- and two-day ahead at Son Tay hydrological station on the Red River using a series of observed flowrate data at seven hydrological stations on three major river branches of the Red River system-Thao River, Da River, and Lo River-as the input data for training, validation, and testing. The comparison results have indicated that the four LSTM-based models exhibit significantly better performance and maintain stability than the FFNN and CNN models. Moreover, LSTM-based models may reach impressive predictions even in the presence of upstream reservoirs and dams. In the case of the stacked LSTM and BiLSTM models, the complexity of these models is not accompanied by performance improvement because their respective performance is not higher than the two standard models (LSTM and GRU). As a result, we realized that in the context of hydrological forecasting problems, simple architectural models such as LSTM and GRU (with one hidden layer) are sufficient to produce highly reliable forecasts while minimizing computation time because of the sequential data nature.

• The Journal of Advanced Prosthodontics
• /
• 제5권3호
• /
• pp.326-332
• /
• 2013

PURPOSE. The purpose of this study was to compare stress distributions of implant-supported crown placed in fibula bone model with those in intact mandible model using three-dimensional finite element analysis. MATERIALS AND METHODS. Two three-dimensional finite element models were created to analyze biomechanical behaviors of implant-supported crowns placed in intact mandible and fibula model. The finite element models were generated from patient's computed tomography data. The model for grafted fibula was composed of fibula block, dental implant system, and implant-supported crown. In the mandible model, same components with identical geometries with the fibula model were used except that the mandible replaced the fibula. Vertical and oblique loadings were applied on the crowns. The highest von Mises stresses were investigated and stress distributions of the two models were analyzed. RESULTS. Overall stress distributions in the two models were similar. The highest von Mises stress values were higher in the mandible model than in the fibula model. In the individual prosthodontic components there was no prominent difference between models. The stress concentrations occurred in cortical bones in both models and the effect of bicortical anchorage could be found in the fibula model. CONCLUSION. Using finite element analysis it was shown that the implant-supported crown placed in free fibula graft might function successfully in terms of biomechanical behavior.

• Steel and Composite Structures
• /
• 제21권4호
• /
• pp.703-753
• /
• 2016

The ratcheting behavior was studied experimentally for Z2CND18.12N elbow piping under cyclic bending and steady internal pressure. Dozens of cyclic plasticity models for structural ratcheting responses simulations were used in the paper. The four models, namely, Bilinear (BKH), Multilinear (MKIN/KINH), Chaboche (CH3), were already available in the ANSYS finite element package. Advanced cyclic plasticity models, such as, modified Chaboche (CH4), Ohno-Wang, modified Ohno-Wang, Abdel Karim-Ohno and modified Abdel Karim-Ohno, were implemented into ANSYS for simulating the experimental responses. Results from the experimental and simulation studies were presented in order to demonstrate the state of structural ratcheting response simulation by these models. None of the models evaluated perform satisfactorily in simulating circumferential strain ratcheting response. Further, improvement in cyclic plasticity modeling and incorporation of material and structural features, like time-dependent, temperature-dependent, non-proportional, dynamic strain aging, residual stresses and anisotropy of materials in the analysis would be essential for advancement of low-cycle fatigue simulations of structures.