• Structural Engineering and Mechanics
• /
• v.85 no.5
• /
• pp.635-644
• /
• 2023

The design of honeycomb sandwich structures is often challenging because these structures can be tailored from a variety of possible cores and face sheets configurations, therefore, the design of sandwich structures is characterized as a time-consuming and complex task. A data-driven computational approach that integrates the analytical method and Artificial Neural Network (ANN) is developed by the authors to rapidly predict the design of sandwich structures for a targeted maximum structural deflection. The elaborated ANN reverse design approach is applied to obtain the thickness of the sandwich core, the thickness of the laminated face sheets, and safety factors for composite sandwich structure. The required data for building ANN model were obtained using the governing equations of sandwich components in conjunction with the Monte Carlo Method. Then, the functional relationship between the input and output features was created using the neural network Backpropagation (BP) algorithm. The input variables were the dimensions of the sandwich structure, the applied load, the core density, and the maximum deflection, which was the reverse input given by the designer. The outstanding performance of reverse ANN model revealed through a low value of mean square error (MSE) together with the coefficient of determination (R2) close to the unity. Furthermore, the output of the model was in good agreement with the analytical solution with a maximum error 4.7%. The combination of reverse concept and ANN may provide a potentially novel approach in designing of sandwich structures. The main added value of this study is the elaboration of a reverse ANN model, which provides a low computational technique as well as savestime in the design or redesign of sandwich structures compared to analytical and finite element approaches.

• Geomechanics and Engineering
• /
• v.33 no.1
• /
• pp.11-18
• /
• 2023

This study explores a new energy partitioning approach to determine the fracture toughness of 3-D printed pristine/recycled high density polyethylene (HDPE) blends employing the essential work of fracture (EWF) concept. The traditional EWF approach conducts a uniaxial tensile test with double-edge notched tensile (DENT) specimens and measures the total energy defined by the area under a load-displacement curve until failure. The approach assumes that the entire total energy contributes to the fracture process only. This assumption is generally true for extruded polymers that fracture occurs in a material body. In contrast to the traditional extrusion manufacturing process, the current 3-D printing technique employs fused deposition modeling (FDM) that produces layer-by-layer structured specimens. This type of specimen tends to include separation energy even after the complete failure of specimens when the fracture test is conducted. The separation is not relevant to the fracture process, and the raw experimental data are likely to possess random variation or noise during fracture testing. Therefore, the current EWF approach may not be suitable for the fracture characterization of 3-D printed specimens. This paper proposed a new energy partitioning approach to exclude the irrelevant energy of the specimens caused by their intrinsic structural issues. The approach determined the energy partitioning location based on experimental data and observations. Results prove that the new approach provided more consistent results with a higher coefficient of correlation.

• Nuclear Engineering and Technology
• /
• v.54 no.11
• /
• pp.4134-4145
• /
• 2022

The collimator is one of the high-heat-flux components used to avoid a series of vacuum and thermal problems. In this paper, the heat load distribution throughout the collimator is first calculated through experimental data, and a transient thermodynamic simulation analysis of the original model is carried out. The error of the pipe outlet temperature between the simulated and experimental values is 1.632%, indicating that the simulation result is reliable. Second, the model is optimized to improve the heat transfer performance of the collimator, including the contact mode between the pipe and the flange, the pipe material and the addition of a twisted tape in the pipe. It is concluded that the convective heat transfer coefficient of the optimized model is increased by 15.381% and the maximum wall temperature is reduced by 16.415%; thus, the heat transfer capacity of the optimized model is effectively improved. Third, to adapt the long-pulse steady-state operation of the experimental advanced superconducting Tokamak (EAST) in the future, steady-state simulations of the original and optimized collimators are carried out. The results show that the maximum temperature of the optimized model is reduced by 37.864% compared with that of the original model. The optimized model was changed as little as possible to obtain a better heat exchange structure on the premise of ensuring the consumption of the same mass flow rate of water so that the collimator can adapt to operational environments with higher heat fluxes and long pulses in the future. These research methods also provide a reference for the future design of components under high-energy and long-pulse operational conditions.

• Steel and Composite Structures
• /
• v.51 no.4
• /
• pp.441-456
• /
• 2024

This paper aims to develop Machine Learning (ML) algorithms to predict the buckling resistance of cold-formed steel (CFS) channels with restrained flanges, widely used in typical CFS sheathed wall panels, and provide practical design tools for engineers. The effects of cross-sectional restraints were first evaluated on the elastic buckling behaviour of CFS channels subjected to pure axial compressive load or bending moment. Feedforward multi-layer Artificial Neural Networks (ANNs) were then trained on different datasets comprising CFS channels with various dimensions and properties, plate thicknesses, and restraining conditions on one or two flanges, while the elastic distortional buckling resistance of the elements were determined according to the Finite Strip Method (FSM). To develop less biased networks and ensure that every observation from the original dataset has the chance of appearing in the training and test set, a K-fold cross-validation technique was implemented. In addition, the hyperparameters of the ANNs were tuned using a grid search technique to provide ANNs with optimum performances. The results demonstrated that the trained ANNs were able to predict the elastic distortional buckling resistance of CFS flange-restrained elements with an average accuracy of 99% in terms of coefficient of determination. The developed models were then used to propose a simple ANN-based design formula for the prediction of the elastic distortional buckling stress of CFS flange-restrained elements. Finally, the proposed formula was further evaluated on a separate set of unseen data to ensure its accuracy for practical applications.

• International Journal of Highway Engineering
• /
• v.11 no.3
• /
• pp.97-109
• /
• 2009

One of the main causes of asphalt rutting is high temperature of the pavement. Nevertheless, there has been few research on lowering the pavement temperature for reducing rutting. This study investigated the performance characteristics of moisture-retaining porous asphalt pavement, which is known to have a temperature reducing effect. The purpose of this study is to quantify the temperature reducing effect of moisture-retaining porous asphalt pavement and its effect of reducing rutting through Accelerated Pavement Testing(APT). Additionally, the possibility of reducing the thickness of the pavement in comparison to general dense grade pavement by analyzing structural layer coefficient of moisture retaining pavement. A total of three test sections consisting of two moisture-retaining porous asphalt pavement sections and one general dense-grade porous asphalt pavement section were constructed for this study. Heating and spraying of water were carried out in a regular cycle. The loading condition was 8.2 ton of wheel load, the tire pressure of $7.03kgf/cm^2$, and the contact area of $610cm^2$. The result of this experiment revealed that the temperature reducing effect of the pavement was about $6.6{\sim}7.9^{\circ}C$(average of $7.4^{\circ}C$) for the middle layer and $7.9{\sim}9.8^{\circ}C$(average of $8.8^{\circ}C$) for surface course, resulting in a rutting reduction of 26% at the pavement surface. Additionally, the structural layer coefficient of moisture retaining pavement measured from a laboratory test was 0.173, about 1.2 times that of general dense grade pavement. The general dense-grade porous asphalt pavement test section exhibited rutting at all layers of surface course, middle layer, and base layer, while the test sections of moisture-retaining porous asphalt pavement manifested rutting mostly at surface course only.

• Journal of Environmental Science International
• /
• v.21 no.8
• /
• pp.977-988
• /
• 2012

This study analyzed the characteristics of stormwater runoff by rainfall type in orchard areas and transportation areas for 2 years(2010~2011year). Effluents were monitored to calculate the Event Mean Concentrations(EMCs) and runoff loads of each pollutant. The pollutant EMCs by volume of stormwater runoff showed the ranges of BOD 0.9~13.6 mg/L, COD 13.7~45.2 mg/L, SS 4.1~236.4 mg/L, T-N 2.123~21.111 mg/L, T-P 0.495~2.214 mg/L in the orchard areas, and was calculated as BOD 2.3~22.5mg/L, COD 4.4~91.1 mg/L, SS 4.3~138.3 mg/L, T-N 0.700~13.500 mg/L, T-P 0.082~1.345 mg/L in the transportation areas. The correlation coefficient of determination in the orchard area was investigated in the order of Total Rainfall(0.81) > Total Runoff(0.76) > Rainfall Intensity(0.56) > Rainfall Duration(0.46) > Antecedent Dry Days(0.27). Also, in the case of the transportation area was investigated in the order of Total Rainfall (0.55) > Total Runoff(0.54) > Rainfall Intensity(0.53) > Rainfall Duration(0.24) > Antecedent Dry Days(0.14). As the result, comparing valuables relating to runoff of non-pollutant source between orchard areas and transportation areas, orchard area($R^2{\geq}0.5$ : X3, X4, X5) was investigated to have more influence of diverse independent valuables compared to the transportation area($R^2{\geq}0.5$ : X3, X4) and the difference of discharge influence factor by the land characteristics appeared apparently.

• Journal of the Korean Geosynthetics Society
• /
• v.12 no.4
• /
• pp.11-19
• /
• 2013

Recently despite the development of analysis program and construction technologies, collapse at the many earth retaining wall construction site of the structure due to the economic and human damage has occurred. The results of geothechnical investigation studies field, it was found to differ from the results of the original design. There may be errors parameters calculated from the results of ground investigation in such a case. And it can be estimated that it is irrational to behavior analysis of the earth retaining wall were analyzed by utilizing the parameters. And in this study, parameters that affect the earth retaining wall the correlations were analyzed using elasto-plastic method. Analysis method was changed various parameters (cohesion, subgrade reaction coefficient, load condition) applied to the elasto-plastic method. And due to a change in the behavior of earth retaining wall materials were analyzed. As a result, the cohesion greatly affects the behavior of earth retaining wall materials in various parameters. For this reason, the results of the geothechnical investigation, confirmation of the actual ground is very important in the design of the earth retaining wall. And, calculating accurate and reasonable of the cohesion of the various parameters is very important.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.28 no.2
• /
• pp.153-160
• /
• 2015

The objective of this study is to construct the TPU aerodynamic database with wind tunnel test data of overall wind loads and responses on tall buildings. In this study, wind tunnel tests were conducted to investigate characteristics of wind forces and the effect of wind load combination by cross-correlation analysis among along-wind overturning moment, across-wind overturning moment and torsional moment on a tall building with various side ratios(D/B=0.33, 0.50, 0.77, 0.83, 0.91, 1.0, 1.1, 1.2, 1.3, 2.0 and 3.0) for different terrain roughnesses. The results of wind tunnel tests were compared with those of past literatures. As a result, there was no significant effects of changing of terrain roughnesses on moment coefficients and power spectral densities of across-wind overturning moment coefficients and torsional moment coefficients with various side ratios. Further, these results were good agreement with those of past literatures. From cross-correlation analysis, the across-wind overturning moment coefficients were highly correlated with the torsional moment coefficients. The results of this study will be helpful for practical designers in preliminary design stage.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.8 no.1
• /
• pp.134-142
• /
• 2020

In this study, in order to evaluate the self-healing performance of cement composites the self-healing test method and the analysis method were suggested by applying constant water head permeability test, chloride migration test and repeated bending test. The method of making a cracked specimen and controlling crack width are also proposed. Constant head water permeability test can evaluate the healing performance by using the decreasing rate of water flow passing through the crack zone of a specimen. Furthermore, the equivalent crack width can be used to intuitively investigate the healing effect with healing period. The chloride migration test can evaluate the healing rate by the decreasing rate of the diffusion coefficient obtained by ASTM C 1202. Mechanical healing performance can be evaluated using ISR and IDR estimated from load vs. CMOD relationship graph obtained through the repeated bending test. Finally, the applicability of proposed self-healing evaluation methods was examined by testing mortar specimens with or without self-healing agents.

• Journal of the Korean Geotechnical Society
• /
• v.15 no.6
• /
• pp.167-185
• /
• 1999

A series of model tests was performed to find the characteristics of lateral behavior of single rigid pile. This paper shows the results of model tests on the lateral behavior of single rigid driven pile in non-homogeneous(two layered) Nak-Dong River sands. The purpose of this paper is to investigate the effect of the ratio of lower layer thickness to embedded pile length, the coefficient ratio of the subgrade reaction and the pile construction conditions(driven & embedded piles) on the characteristics of lateral behavior of single pile. The results of model tests show that the lateral behavior in non-homogeneous soil depends upon drop energy considerably, that is, in the case of H/L=0.75, as the drop energy increases three times the decrease percentage increases about 2.12 times. In the driven pile with non-homogeneous soil of $E_{h1}/E_{h2}=5.56$, the effect of upper layer with large stiffness on the decrease of lateral deflection is remarkably smaller than embedded pile. In non-homogeneous soil, the maximum bending moment of driven pile is in the range of 100 132% in comparison with embedded pile. The reason is that the stiffness of soil around pile increases with drop vibration and so the pile behavior is similar to the flexible pile behavior by means of the increase of relative stiffness of pile, In this paper, the experimental equations for lateral load and H/L on $y_D/y_E \; & \; MBM_D/MBM_E$ are suggested from model tests.