• Steel and Composite Structures
• /
• v.45 no.6
• /
• pp.877-894
• /
• 2022

Geopolymer concrete (GPC) has emerged as a feasible choice for construction materials as a result of the environmental issues associated with the production of cement. The findings of this study contribute to the development of machine learning methods for estimating the properties of eco-friendly concrete to help reduce CO₂ emissions in the construction industry. The compressive strength (f_c) of GPC is predicted using artificial intelligence approaches in the present study when ground granulated blast-furnace slag (GGBS) is substituted with natural zeolite (NZ), silica fume (SF), and varying NaOH concentrations. For this purpose, two machine learning methods multi-layer perceptron (MLP) and radial basis function (RBF) were considered and hybridized with arithmetic optimization algorithm (AOA), and grey wolf optimization algorithm (GWO). According to the results, all methods performed very well in predicting the f_c of GPC. The proposed AOA - MLP might be identified as the outperformed framework, although other methodologies (AOA - RBF, GWO - RBF, and GWO - MLP) were also reliable in the fc of GPC forecasting process.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.2 no.1
• /
• pp.1-9
• /
• 1999

To find pertinent soil type and maintenance method for artificial planting ground, the effects of soil compositions{sandy loam(S), vermiculite(V), sandy loam+vermiculite+sand(SVS), sandy loam+ carbonized rice husk+sand(SCS), sandy loam+humus sawdust+sand(SHS)}, and fertilizers (organic, chemical) on plant(kentuckyblue grass) growth were measured and compared from the field experiment. The results are summarized as follows 1. the highest germination rate is found from "vermiculite(V)" and the lowest from "sandy loam(S)" among tested 5 soil compositions. 2. "sandy loam+vermiculite+sand(SVS)" composition shows the highest plant height growth effect (5cm growth during tested 3 months) comparing to other 4 compositions. 3. "sandy loam+vermiculite+sand(SYS)" composition shows the highest ground covering rate after first two months, but it concede its order to "sandy loam+humus sawdust+sand(SHS)" composition after next one month growing. 4. the effects of fertilizers are follows 1) Among the blocks where no fertilizer was tried, the predominant height growth was obvious in "sandy loam+carbonized rice husk+sand(SCS)" and "sandy loam+humus sawdust+sand(SHS)" composition. 2) Among the blocks where chemical fertilizer was tried, relatively positive results were found from "vermiculite(V)" and "sandy loam+vermiculite+sand(SYS)" blocks on germination and growth rate. But on the ground coverage ratio, the effect of "sandy loam+carbonized rice husk+sand(SCS)" composite precede that of those 2 composites. 3) Among the blocks where organic fertilizer was tried, "sandy loam+humus sawdust+sand(SHS)" and "vermiculite(V)" blocks show relatively high ground coverage rate, growth rate than others. 4) When compositional differences were not considered, the block where organic fertilizer was tried shows most positive effects on all 3 measurements-germination ratio, height growth and ground covering.

• The Journal of Engineering Geology
• /
• v.27 no.2
• /
• pp.153-164
• /
• 2017

The geological factors for causing ground subsidence are very diverse. It can be affected by any geological or extrinsic influences, and even within the same geological factor, the soil depression impact factor can be determined by different physical properties. As a result of reviewing a large number of papers and case histories, it can be seen that there are seven categories of ground subsidence factors. The depth and thickness of the overburden can affect the subsidence depending on the existence of the cavity, whereas the depth and orientation of the boundary between soil and rock are dominant factors in the ground composed of soil and rock. In case of soil layers, more various influencing factors exist such as type of soil, shear strength, relative density and degree of compaction, dry unit weight, water content, and liquid limit. The type of rock, distance from the main fracture and RQD can be influential factors in the bedrock. When approaching from the hydrogeological point of view, the rainfall intensity, the distance and the depth from the main channel, the coefficient of permeability and fluctuation of ground water level can influence to ground subsidence. It is also possible that the ground subsidence can be affected by external factors such as the depth of excavation and distance from the earth retaining wall, groundwater treatment methods at excavation work, and existence of artificial facilities such as sewer pipes. It is estimated that to evaluate the ground subsidence factor during the construction of underground structures in urban areas will be essential. It is expected that ground subsidence factors examined in this study will contribute for the reliable evaluation of the ground subsidence risk.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.5 no.2
• /
• pp.19-27
• /
• 2002

This paper deals with the experimental investigation on the hydrodynamic behavior of the submerged artificial seabed system in regular waves. This system can function as a basis of seaweed forest which will cultivate coastal fishing ground and enhance coastal productivity. The experiment was conducted with the submerged rectangular plates of different length and depth in 2-D wave flume of KRISO/KORDI. The wave exciting forces, mooring line tension and 2-D motion response are measured and analyzed to figure out the design strategy.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.27 no.2
• /
• pp.130-139
• /
• 1999

The survey site, the Hyundai Apartment Complex in the Bundang Model Complex, includes three housing layout types; super high-rise, high-rise and low-rise apartment buildings. The site includes artificial ground over underground parking lots. The overall objective of this study is to evaluate social and physical factors of housing environments in each types of layout. The data has been complied from residents of apartment through questionnaire. The questionnaire include elements of neighborhood, outdoor space, parking zones, and the overall complex design in each layout types. The predictors of outdoor space satisfaction in apartment housing complex are found as follows: "abundance of trees in quantity", "the role as front yards", "harmony of buildings with landscape", "the more distance between buildings" and "maintenance quality of site". Layout of super high-rise apartment site is most satisfied. Introduction of car-free deck space is favored by resident because of safty and quiet resting area. However the low quality of green and lack of shades on the artificial land are identified as problems.on the artificial land are identified as problems.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2013.05a
• /
• pp.321-322
• /
• 2013

Recently, the exhaustion of resource and environmental damage is serious due to the global warming because of the CO₂ exhaust and each type the natural aggregate picking described below. meanwhile, The rest is the actual condition gone to the dumping ground that there is nearly no use which the waste glass can recycle and it is recycled. This research applied the waste glass as the cement substitute material the inorganic binder and coares aggregate substitute material. It utilizes the substitute material of the cement according to it and natural aggregate and tries to develop the environment-friendly artificial stone. The inorganic binder used the blast furnace slag, red mud, and fly ash. The straight type steel fiber, PVA fiber, PA fiber, and cellulosic fiber were used with a kind of fiber. As to the experimental item according to it, the compressive strength is the flexural strength and compressive strength.

• Geomechanics and Engineering
• /
• v.17 no.5
• /
• pp.421-427
• /
• 2019

Coastal erosion is becoming a significant problem in Greece, Bangladesh, and globally. For the prevention and minimization of damage from coastal erosion, combinations of various structures have been used conventionally. However, most of these methods are expensive. Therefore, creating artificial beachrock using local ureolytic bacteria and the MICP (Microbially Induced Carbonate Precipitation) method can be an alternative for coastal erosion protection, as it is a sustainable and eco-friendly biological ground improvement technique. Most research on MICP has been confined to land ureolytic bacteria and limited attention has been paid to coastal ureolytic bacteria for the measurement of urease activity. Subsequently, their various environmental effects have not been investigated. Therefore, for the successful application of MICP to coastal erosion protection, the type of bacteria, bacterial cell concentration, reaction temperature, cell culture duration, carbonate precipitation trend, pH of the media that controls the activity of the urease enzyme, etc., are evaluated. In this study, the effects of temperature, pH, and culture duration, as well as the trend in carbonate precipitation of coastal ureolytic bacteria isolated from two coastal regions in Greece and Bangladesh, were evaluated. The results showed that urease activity of coastal ureolytic bacteria species relies on some environmental parameters that are very important for successful sand solidification. In future, we aim to apply these findings towards the creation of artificial beachrock in combination with a geotextile tube for coastal erosion protection in Mediterranean countries, Bangladesh, and globally, for bio-mediated soil improvement.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.30 no.2
• /
• pp.34-43
• /
• 2022

In Efforts to prevent CFIT accidents so far, have been emphasizing various education measures to minimize the occurrence of human errors, as well as enforcement measures. However, current engineering measures remain in a system (TAWS) that gives warnings before colliding with ground or obstacles, and even actual automatic avoidance maneuvers are not implemented, which has limitations that cannot prevent accidents caused by human error. Currently, various attempts are being made to apply machine learning-based artificial intelligence agent technologies to the aviation safety field. In this paper, we propose a deep reinforcement learning-based artificial intelligence agent that can recognize CFIT situations and control aircraft to avoid them in the simulation environment. It also describes the composition of the learning environment, process, and results, and finally the experimental results using the learned agent. In the future, if the results of this study are expanded to learn the horizontal and vertical terrain radar detection information and camera image information of radar in addition to the terrain database, it is expected that it will become an agent capable of performing more robust CFIT autonomous avoidance.

• Nuclear Engineering and Technology
• /
• v.55 no.1
• /
• pp.190-200
• /
• 2023

As the field of application of in-situ gamma spectroscopy is diversified, proficiency is required for consistent and accurate analysis. In this study, a program was developed to virtually create gamma energy spectra of artificial nuclides, which are difficult to obtain through actual measurements, for training. The virtual spectrum was created by synthesizing the spectra of the background radiation obtained through actual measurement and the theoretical spectra of the artificial radionuclides obtained by a Monte Carlo simulation. Since the theoretical spectrum can only be obtained for a given geometrical structure, representative major geometries for in-situ measurement (ground surface, concrete wall, radioactive waste drum) and the detectors (HPGe, NaI(Tl), LaBr₃(Ce)) were predetermined. Generated virtual spectra were verified in terms of validity and harmonization by gamma spectrometry and energy calibration. As a result, it was confirmed that the energy calibration results including the peaks of the measured spectrum and the peaks of the theoretical spectrum showed differences of less than 1 keV from the actual energies, and that the calculated radioactivity showed a difference within 20% from the actual inputted radioactivity. The verified data were assembled into a database and a program that can generate a virtual spectrum of desired condition was developed.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.673-687
• /
• 2023

Convolutional neural network-based deep learning technology is the most commonly used in image identification, but it requires large-scale data for training. Therefore, application in specific fields in which data acquisition is limited, such as in the military, may be challenging. In particular, the identification of ground weapon systems is a very important mission, and high identification accuracy is required. Accordingly, various studies have been conducted to achieve high performance using small-scale data. Among them, the ensemble method, which achieves excellent performance through the prediction average of the pre-trained models, is the most representative method; however, it requires considerable time and effort to find the optimal combination of ensemble models. In addition, there is a performance limitation in the prediction results obtained by using an ensemble method. Furthermore, it is difficult to obtain the ensemble effect using models with imbalanced classification accuracies. In this paper, we propose a transfer learning-based feature fusion technique for heterogeneous models that extracts and fuses features of pre-trained heterogeneous models and finally, fine-tunes hyperparameters of the fully connected layer to improve the classification accuracy. The experimental results of this study indicate that it is possible to overcome the limitations of the existing ensemble methods by improving the classification accuracy through feature fusion between heterogeneous models based on transfer learning.