• Smart Structures and Systems
• /
• v.29 no.2
• /
• pp.337-350
• /
• 2022

This paper focuses on predicting the post-heating mechanical properties of cementitious composites reinforced with multi-scale additives using the Artificial Neural Network (ANN) approach. A total of four different feed-forward ANN models are developed using 261 data sets collected from 18 published sources. The models are optimized using 12 input parameters selected based on a comprehensive literature review to predict the residual compressive strength, the residual flexural strengths, elastic modulus, and fracture energy of heat-damaged cementitious specimens. Furthermore, the ANN is employed to predict the impact of several variables including; the content of polypropylene (PP) microfibers and carbon nanotubes (CNTs) used in the concrete, mortar, or paste mix design, length of PP fibers, the average diameter of CNTs, and the average length of CNTs. The influence of the studied parameters is investigated at different heating levels ranged from 25℃ to 800℃. The results demonstrate that the developed ANN models have a strong potential for predicting the mechanical properties of the heated cementitious composites based on the mixing ingredients in addition to the heating conditions.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.82-95
• /
• 2024

As automated essay scoring (AES) has progressed from handcrafted techniques to deep learning, holistic scoring capabilities have merged. However, specific trait assessment remains a challenge because of the limited depth of earlier methods in modeling dual assessments for holistic and multi-trait tasks. To overcome this challenge, we explore providing comprehensive feedback while modeling the interconnections between holistic and trait representations. We introduce the DualBERT-Trans-CNN model, which combines transformer-based representations with a novel dual-scale bidirectional encoder representations from transformers (BERT) encoding approach at the document-level. By explicitly leveraging multi-trait representations in a multi-task learning (MTL) framework, our DualBERT-Trans-CNN emphasizes the interrelation between holistic and trait-based score predictions, aiming for improved accuracy. For validation, we conducted extensive tests on the ASAP++ and TOEFL11 datasets. Against models of the same MTL setting, ours showed a 2.0% increase in its holistic score. Additionally, compared with single-task learning (STL) models, ours demonstrated a 3.6% enhancement in average multi-trait performance on the ASAP++ dataset.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1995.04a
• /
• pp.325-330
• /
• 1995

Three types of model concrete with different mix ratios for 1/10 scale reinforced concrete model were made and tested to find the best solution for the simulation of mechanical characteristics of prototype concrete. Scaled reinforcing materials in diameter(D1.8) having similar con-figuration and mechanical properties with commercial deformed bars(D19) were prepared for 1/10 scale model tests of reinforced concrete structures. Two types of model test using D1.8 model reinforcing bars and model concrete, monotonic simple beam test and cyclic cantilever beam test, were performed to ensure and check the similitude of bond behavior between 1/10 scale model and prototype. The test results showed that the flexural behavior of 1/10 scale models can be simulated with accuracy enough for practical use in monotonic and cyclic loading test.

• Tribology and Lubricants
• /
• v.28 no.6
• /
• pp.272-277
• /
• 2012

Finite element analysis was carried out using wafer-scale and particle-scale models to understand the mechanism of the fast removal rate(edge effect) at wafer edges in the chemical-mechanical polishing process. This is the first to report that a particle-scale model can explain the edge effect well in terms of stress distribution and magnitude. The results also revealed that the mechanism could not be fully understood by using the wafer-scale model, which has been used in many previous studies. The wafer-scale model neither gives the stress magnitude that is sufficient to remove material nor indicates the coincidence between the stress distribution and the removal rate along a wafer surface.

• Coupled systems mechanics
• /
• v.8 no.2
• /
• pp.99-109
• /
• 2019

This work investigates the accuracy and performance of a $FE^2$ multi-scale implementation used to predict the behavior of composite materials. The equations are formulated assuming the small deformations solid mechanics approach in non-linear material models with hardening plasticity. The uniform strain boundary conditions are applied for the macro-to-micro transitions. A parallel algorithm was implemented in order to solve large engineering problems. The scheme proposed takes advantage of the domain decomposition method at the macro-scale and the coupling between each subdomain with a micro-scale model. The precision of the method is validated with a composite material problem and scalability tests are performed for showing the efficiency.

• Particle and aerosol research
• /
• v.5 no.2
• /
• pp.45-52
• /
• 2009

Black carbon, which is a by-product of combustion of fossil fuel and biomass burning, is the component that imposes the largest uncertainty on quantifying aerosol climate effect. The direct, indirect and semi-direct climate effects of black carbon depend on its state of the mixing with other water-soluble aerosol components. The process that transforms hydrophobic externally mixed black carbon particles into hygroscopic internally mixed ones is called "aging". In most climate models, simple parameterizations for the aging time scale are used instead of solving detailed dynamics equations on the aging process due to the computation cost. In this study, a new parameterization for the black carbon aging time scale due to condensation and coagulation is presented as a function of the concentration of hygroscopic atmospheric components and the black carbon particle size. It is shown that the black carbon aging time scale due to condensation of sulfuric acid vapors varies to a large extent depending on the sulfuric acid concentration and the black carbon particle size. This result indicates that the constant aging time scale values suggested in the literature cannot be directly applied to a global scale modeling. The aging time scale due to coagulation with internally mixed aerosol particles shows an even stronger dependency on particle size, which implies that the use of a particle-size-independent aging time scale may lead to a large error when the aging is dominated by coagulation.

• Journal of Mechanical Science and Technology
• /
• v.18 no.8
• /
• pp.1479-1489
• /
• 2004

The effect of fluctuation of Equivalence Ratio (ER) in a turbulent reactive field has been studied in order to check the global combustion characteristics induced by the local fluctuation. When the flow is premixed on a large scale, closer examination on a small scale reveals that local fluctuations of ER exist in an imperfectly premixed mixture, and that these fluctuations must be considered to correctly estimate the mean reaction rate. The fluctuation effect is analyzed with DNS by considering the joint PDF of reactive scalar and ER, followed by modeling study where an extension of stochastic mixing models accounting for the ER fluctuation is reviewed and tested. It was found that models prediction capability as well as its potential is in favor to this case accounting the local ER fluctuation. However, the effect of local fluctuation did not show any notable changes on the mean global characteristics of combustion when statistical independence between the reactive scalar and ER field is imposed, though it greatly influenced the joint PDF distribution. The importance of taking into account the statistical dependency between ER and combustible at the initial phase is demonstrated by testing the modeled reaction rate.

• Structural Engineering and Mechanics
• /
• v.17 no.3_4
• /
• pp.493-526
• /
• 2004

Semiactive control systems have received considerable attention for protecting structures against natural hazards such as strong earthquakes and high winds, because they not only offer the reliability of passive control systems but also maintain the versatility and adaptability of fully active control systems. Among the many semiactive control devices, magnetorheological (MR) fluid dampers comprise one particularly promising class. In the field of civil engineering, much research and development on MR fluid damper-based control systems has been conducted since this unique semiactive device was first introduced to civil engineering applications in mid 1990s. In 2001, MR fluid dampers were applied to the full-scale in-service civil engineering structures for the first time. This state-of-the-art paper includes a detailed literature review of dynamic models of MR fluid dampers for describing their complex dynamic behavior and control algorithms considering the characteristics of MR fluid dampers. This extensive review provides references to semiactive control systems using MR fluid dampers. The MR fluid damper-based semiactive control systems are shown to have the potential for mitigating the responses of full-scale civil engineering structures under natural hazards.

• Wind and Structures
• /
• v.24 no.2
• /
• pp.119-144
• /
• 2017

The traditional method for the simulation of high Reynolds number (Re) effects on wind loads on cooling tower models in wind tunnels focuses only on the mean wind pressure distribution. Based on observed effects of some key factors on static/dynamic flow characteristics around cooling towers, the study reported in this paper describes a comprehensive simulation method using both mean and fluctuating wind pressure distributions at high Re as simulation targets, which is indispensable for obtaining the complete full-scale wind effects in wind tunnels. After being presented in this paper using a case study, the proposed method is examined by comparing the full covariance matrices and the cross-spectral densities of the simulated cases with those of the full-scale case. Besides, the cooling tower's dynamic structural responses obtained using the simulated wind pressure fields are compared with those obtained by using the full-scale one. Through these works, the applicability and superiority of the proposed method is validated.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.42 no.2
• /
• pp.35-48
• /
• 2017

As the technology convergence paradigm emerges, the need for "CIA techniques" to analyze the mutual effects of technology is increasing. However, since the CIA input parameter estimation is difficult, the present study suggests a "CIA input parameter setting model" to alleviate the difficulty of CIA input parameter estimation. This paper is focused on the difference of measurement difficulty by each scale which expert's estimation behavior was defined as measurement activity quantifying the judgment of future technology. Therefore, this model is designed to estimate the input variable as a sequence or isometric scale that is relatively easy to measure, and then converts it into a probability value. The input parameter setting model of the CIA technique consists of three sub-models : 'probability value derivation model', 'influence estimation model', and 'impact value calculation model', in order to develop a series of models the Thurstone V model, Regression Analysis, etc has been used.