• Water Engineering Research
• /
• 제4권2호
• /
• pp.69-85
• /
• 2003

Previously developed empirical equations used to calculate the parameters of the transient storage model are analyzed in depth in order to evaluate their behavior in representing solute transport in the natural streams with storage zone. A comparative analysis of the existing theoretical and experimental equations used to predict parameters of the transient storage (TS) model is reported. New simplified equations for predicting 4 key parameters of the TS model using hydraulic data sets that are easily obtained in the natural streams are also developed. The weighted one-step Huber method, which is one of the nonlinear multi-regression methods, is applied to derive new parameters equation. These equations are proven to be superior in explaining mixing characteristics of natural streams with the transient storage zone more precisely than the other existing equations.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2004년도 춘계학술대회 논문집
• /
• pp.70-73
• /
• 2004

The approach in this thesis is for prediction of deformed strip profile in hot rolling mill. This approach shows how to make an expression as a mathematical form in predicting strip profile. This approach is based on the velocity field, shear stress and material flow on the strip edge along width direction and lateral displacement and stress along width are analytically calculated. Roll force is calculated in each section and then combined together to show roll force distribution along width. All the assumptions to make equation form for this approach are supported by FEM simulation result and the result of model is verified by FEM result. So, this model will supply very useful tool to the researcher and engineers which takes less time and has similar accuracy in predicting roll force profile comparing to FEM simulation. This model has to be combined with deformed roll profile model, which include thermal crown prediction and wear prediction model to predict deformed strip profile.

• Journal of Mechanical Science and Technology
• /
• 제18권1호
• /
• pp.153-166
• /
• 2004

This paper presents a comparative study of a fully coupled, upwind, compressible Navier-Stokes code with three two-equation models and the Baldwin-Lomax algebraic model in predicting transonic/supersonic flow. The k-$\varepsilon$ turbulence model of Abe performed well in predicting the pressure distributions and the velocity profiles near the flow separation over the axisymmetric bump, even though there were some discrepancies with the experimental data in the shear-stress distributions. Additionally, it is noted that this model has y$\^$*/ in damping functions instead of y$\^$＋/. The turbulence model of Abe and Wilcox showed better agreements in skin friction coefficient distribution with the experimental data than the other models did for a supersonic compression ramp problem. Wilcox's model seems to be more reliable than the other models in terms of numerical stability. The two-equation models revealed that the redevelopment of the boundary layer was somewhat slow downstream of the reattachment portion.

• 한국포장학회지
• /
• 제4권1호
• /
• pp.23-32
• /
• 1997

The change in moisture content of moisture sensitive products in moisture-semipermeable packages was investigated for the purpose of predicting the shelf life of a product-package combination. A mathematical model, and a computer program based on the physiochemical properties of the product and the moisture permeability of the package was developed. The moisture content for products in moisture-semipermeable packages was determined under various environmental conditions and the results were compared with the predicted values by means of the simulation model. These experimental studies demonstrated that the prediction of the change in moisture content of packaged products over time by the simulation model is accurate, within a practical range of temperature and relative humidity values. The developed semi-empirical model is considered to have applications in industry, since it provides product shelf life information for a range of temperature and relative humidity conditions, with a limited number of experimentally obtained data points.

• 소성∙가공
• /
• 제33권5호
• /
• pp.354-362
• /
• 2024

In this paper, the capabilities of recurrent neural networks (RNNs) to describe the viscoelastic properties of acrylonitrile-butadiene styrene (ABS) are investigated. The RNN model was trained using one-dimensional strains and corresponding stress data generated by the finite element method. The optimal model was then employed to predict the viscoelastic behavior of unseen test data. Furthermore, the viscoelastic-based RNN model was tested for extrapolation using other types of strain and corresponding stress data beyond the training set. The agreement between the predicted and actual stresses demonstrates the robust performance of the trained RNN model in predicting different types of strain inputs for larger strain tests, despite being trained only with step strain inputs. Therefore, the use of RNNs can be considered a viable alternative to conventional models for predicting viscoelastic behavior.

• 한국학교수학회논문집
• /
• 제10권1호
• /
• pp.113-127
• /
• 2007

본 고는 5년 간의 고등학교 연합학력평가자료 분석을 통하여 수리영역의 난이도 예측요인을 분석하였다. 난이도 예측을 위한 통계적 모형을 산출하기 위하여 먼저 문항분석을 통해 수리영역의 난이도 예측에 영향을 미칠 것으로 판단되는 주요 변인들을 '내용영역', '행동영역', '문항의 형식' 등의 범주에 따라 추출하였다. 추출된 독립 변인들에 대하여 단계선택방법을 사용한 다중회귀분석을 실시함으로써 정답률 예측에 유의미한 변수들을 선택하였으며, 교차 타당도를 통하여 최종적으로 선택된 예측 모형이 독립적으로 수집된 자료에 대하여 어느 정도의 설명력을 보이는지 검증하였다. 본 연구는 대학수능시험 출제나 현장에서 수리영역의 평가문항을 개발하는데 있어서 사전 정답률을 예측하는데 있어 고려해야 할 요인을 제시함으로써 보다 정확한 정답률 예측에 필요한 기초정보를 제공하는데 그 의의를 두고 있다.

• 한국의료질향상학회지
• /
• 제9권1호
• /
• pp.18-32
• /
• 2002

Background : We developed a model for predicting premature discharge and identifying related factors. Methods : Prediction model was developed by data mining techniques. Basic data were collected from the total discharge data base of a university hospital in Chungnam Province during the period from July 1, 1999 to June 30, 2000. Results : 1. Among 22,873 patients, the number of patients discharged with usual discharge orders were 21,695 or 94.8%. The number of the prematurely discharged patients were 1,178 or 5.2%. 2. The primary reason for unusual discharge was transfer to other hospital. Move to a local hospital closer to their home and burdensome medical expenses were main reasons. 3. Predictability of each model was tested using the top 10 percent of patients with the highest probabilities of premature discharge. The neural network model was chosen as the most appropriate model for predicting prematurely discharged patients. 4. Ten percent of the total number of patients had been selected randomly to test the effectiveness of the neural network model. We have chosen the threshold of the neural network model as 0.7. The number of patients who were expected to discharge prematurely was 312. Among them, 241 had been discharged prematurely (77.2%). Conclusion : Of the several data mining techniques used, the neural network model was the most effective, It can be used to identify and manage the patients who are expected to discharge prematurely.

• Advances in concrete construction
• /
• 제14권1호
• /
• pp.71-77
• /
• 2022

Creep phenomenon affects the stability and integrity of concrete structures. An inaccurate prediction of these strains may lead to the appearance of cracks and excessive deflections which may cause in some cases the demolition of structures. In fact, the measured values of these uncontrolled strains appear often to be clearly different and larger than the expected ones. Therefore, an accurate prediction of concrete deformations is a necessity. As a matter of fact, the codified descriptions of this phenomenon are unreliable and don't consider the effect of admixtures. The physical nature of creep is not well understood and almost all creep models are mainly of empirical nature. To overcome this issue, a study of the correlation between different parameters affecting concrete creep is performed and a new model for predicting creep of concrete is elaborated. This new model considers the effect of admixtures, specifically the silica fume, in predicting concrete creep and allows an accurate prediction of this phenomenon. The proposed model is based on the observation of physical behavior of creep phenomenon. It targets at expressing creep compliance in terms of structural and environmental parameters. In fact, the experimental observations show that creep curves follow two kinetic regimes leading to a model called Phenomenological Creep Model. By adequate regressions and substitutions, and according to this model, we can express creep compliance in terms of structural, environmental parameters and admixture types and percentage. The proposed new Phenomenological Creep Model Silica Fume (PCM19SF) calculates accurately creep of concrete by considering the effect of silica fume.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제9권5호
• /
• pp.161-168
• /
• 2020

스포츠 경기 결과예측은 전반적인 경기의 흐름과 승패에 영향을 미치는 변인들의 분석을 통해 팀의 전략 수립을 가능하게 해준다. 이와 같은 스포츠 경기결과 예측에 대한 연구는 주로 통계학적 기법과 기계학습 기법을 활용하여 진행되어 왔다. 승부예측 모델은 무엇보다 예측 성능이 가장 중요시된다. 그러나 최적의 성능을 보이는 예측 모델은 학습에 사용되는 데이터에 따라 다르게 나타나는 경향을 보였다. 본 논문에서는 이러한 문제를 해결하기 위해 데이터가 달라지더라도 해당 데이터에 대한 예측 시 가장 좋은 성능을 보이는 모델의 선택이 가능한 기존의 축구경기결과 예측에서 좋은 성능을 보여온 통계학적 모델과 기계학습 모델을 결합한 새로운 앙상블 모델을 제안한다. 본 논문에서 제안하는 앙상블 모델은 각 단일모델들의 경기 예측결과와 실제 경기결과를 병합한 데이터로부터 최종예측모델을 학습하여 경기 승부예측을 수행한다. 제안 모델에 대한 실험 결과, 기존 단일모델들에 비해 높은 성능을 보였다.

• Computers and Concrete
• /
• 제12권5호
• /
• pp.697-715
• /
• 2013

Numerous studies have been conducted to understand the shear behavior of reinforced concrete (RC) beams since it is a complex phenomenon. The diagonal cracking strength of a RC beam is critical since it is essential for determining the minimum amount of stirrups and the contribution of concrete to the shear strength of the beam. Most of the existing equations predicting the diagonal cracking strength of RC beams are based on experimental data. A powerful computational tool for analyzing experimental data is an artificial neural network (ANN). Its advantage over conventional methods for empirical modeling is that it does not require any functional form and it can be easily updated whenever additional data is available. An ANN model was developed for predicting the diagonal cracking strength of RC slender beams without stirrups. It is shown that the performance of the ANN model over the experimental data considered in this study is better than the performances of six design code equations and twelve equations proposed by various researchers. In addition, a parametric study was conducted to study the effects of various parameters on the diagonal cracking strength of RC slender beams without stirrups upon verifying the model.