• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.103-116
• /
• 2021

Analysis of runoff is substantial for effective water management in the watershed. Runoff occurs by reaction of a watershed to the rainfall and has non-linearity and uncertainty due to the complex relation of weather and watershed factors. ANN (Artificial Neural Network), which learns from the data, is one of the machine learning technique known as a proper model to interpret non-linear data. The performance of ANN is affected by the ANN's structure, the number of hidden layer nodes, learning rate, and activation function. Especially, the activation function has a role to deliver the information entered and decides the way of making output. Therefore, It is important to apply appropriate activation functions according to the problem to solve. In this paper, ANN models were constructed to estimate runoff with different activation functions and each model was compared and evaluated. Sigmoid, Hyperbolic tangent, ReLU (Rectified Linear Unit), ELU (Exponential Linear Unit) functions were applied to the hidden layer, and Identity, ReLU, Softplus functions applied to the output layer. The statistical parameters including coefficient of determination, NSE (Nash and Sutcliffe Efficiency), NSEln (modified NSE), and PBIAS (Percent BIAS) were utilized to evaluate the ANN models. From the result, applications of Hyperbolic tangent function and ELU function to the hidden layer and Identity function to the output layer show competent performance rather than other functions which demonstrated the function selection in the ANN structure can affect the performance of ANN.

• Steel and Composite Structures
• /
• v.47 no.2
• /
• pp.167-184
• /
• 2023

In this study, a new recentering friction device (RFD) to retrofit steel moment frame structures is introduced. The device provides both self-centering and energy dissipation capabilities for the retrofitted structure. A hybrid performance-based seismic design procedure considering multiple limit states is proposed for designing the device and the retrofitted structure. The design of the RFD is achieved by modifying the conventional performance-based seismic design (PBSD) procedure using computational intelligence techniques, namely, genetic algorithm (GA) and artificial neural network (ANN). Numerous nonlinear time-history response analyses (NLTHAs) are conducted on multi-degree of freedom (MDOF) and single-degree of freedom (SDOF) systems to train and validate the ANN to achieve high prediction accuracy. The proposed procedure and the new RFD are assessed using 2D and 3D models globally and locally. Globally, the effectiveness of the proposed device is assessed by conducting NLTHAs to check the maximum inter-story drift ratio (MIDR). Seismic fragilities of the retrofitted models are investigated by constructing fragility curves of the models for different limit states. After that, seismic life cycle cost (LCC) is estimated for the models with and without the retrofit. Locally, the stress concentration at the contact point of the RFD and the existing steel frame is checked being within acceptable limits using finite element modeling (FEM). The RFD showed its effectiveness in minimizing MIDR and eliminating residual drift for low to mid-rise steel frames models tested. GA and ANN proved to be crucial integrated parts in the modified PBSD to achieve the required seismic performance at different limit states with reasonable computational cost. ANN showed a very high prediction accuracy for transformation between MDOF and SDOF systems. Also, the proposed retrofit showed its efficiency in enhancing the seismic fragility and reducing the LCC significantly compared to the un-retrofitted models.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.378-384
• /
• 2023

In this study, we aim to derive the descriptor vector conditions that can simultaneously achieve the efficiency and accuracy of artificial Neural Network Potentials (NNP). The material system selected is silicon, a highly applicable material in various industries. Atomic structure-dependent energy data for training artificial neural networks were generated through density functional theory calculations. Behler-Parrinello type atomic-centered symmetric functions were employed as descriptors, and various length vector NNPs were generated. These NNPs were applied to reproduce the structure and mechanical properties of silicon materials in molecular dynamics simulations. In our findings, the minimum vector length for achieving both learning and computational efficiency while maintaining property reproducibility is approximately 50. It was also observed that, for the same conditions, incorporating more angle-dependent symmetric functions into the descriptor vector, could enhance the accuracy of NNP. Our results can provide guidelines for optimizing the conditions of descriptor vectors to achieve both efficiency and accuracy of NNP, simultaneously.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.119-126
• /
• 2024

In building information model (BIM), it is difficult to train an artificial intelligence (AI) model due to the lack of sufficient data about individual projects in an architecture firm. In this paper, we present a methodology to correctly train an AI neural network model based on a large language model (LLM) to predict the steel structure product weight ratios in BIM. The proposed method, with the aid of the LLM, can overcome the inherent problem of limited data availability in BIM and handle a combination of natural language and numerical data. The experimental results showed that the proposed method demonstrated significantly higher accuracy than methods based on a smaller language model. The potential for effectively applying large language models in BIM is confirmed, leading to expectations of preventing building accidents and efficiently managing construction costs.

• Applied Microscopy
• /
• v.38 no.2
• /
• pp.73-79
• /
• 2008

In Biology, Studies Using Electron Microscopy for making Cell Structure to 3D reconstruction very fast development. Recently, by using Cryo fixation, we can see cell 3D structure without structural change, instead of using chemical fixation which can change cell structure. Before using this technology, we could understand cell structures only in 2D images. But now, through cryo-ET, 3D reconstruction of cell structure without artificial structure changes can be possible and this technology will give us many advantages in Drug delivery and Nanothechnology.

• Korean Journal of Environmental Biology
• /
• v.30 no.1
• /
• pp.71-76
• /
• 2012

Spills of $M/V$ Hebei Spirit on $7^{th}$ December 2007 caused a seriously damage to the ecosystem of Korean coast. Of these, microbial communities (i.e., attached benthic micro-algae) were reported to be sentive to the environmental change so it can be used for ecological risk assessment. Our experiment was designed to examine the ecological risk assessments for oil pollutant using benthic attached algal community on the artificial substrates of acrylic plates. Field monitoring in the culture system was conducted in Jangmok Bay. The abundances of attached micro-algae on artificial substrates gradually increased with increasing of sampling times. Among them, diatoms were the most important colonizer of coastal water, with the genera $Cylindrotheca$ and $Navicular$ most abundant. In particular, developed the culture system has correctly measured qualitative and quantitative abundance of attached micro-algae because same acrylic plates as artificial substrates were used. Thus, this culture system may be directly applied to the ecological risk experiments of microbial community structure from oil pollutants.

• Journal of Environmental Science International
• /
• v.22 no.3
• /
• pp.309-318
• /
• 2013

This study was conducted for determining the influence of artificial structures on benthic macroinvertebrate communities in stream. Sampling was taken at upper(pool), down(riffle) and control(riffle) from two check dams, two weirs, one agricultural reservoir, and one multipurpose dam in northern part of Gyeongsangbuk-do. The benthic macroinvertebrate communities of these structures were surveyed during 2009 to 2011. The simple habitat of benthic macroinvertebrates occurred at the upper sites due to pooling effects from artificial structures. Specifically, Check dam1, Jusanji, Imha dam showed very low biological attribute values compared to the down and control sites, which have greater difference in substrate characteristics. However, in the upper sites of Check dam2, Weir1 and Weir2, the difference of values of biological attributes was relatively smaller. Also, proportion of functional feeding groups and functional habit groups were relatively simpler at upper stream and the degree of community differences was greater between upper and down, control sites. Spearman's correlation between biological attributes and substrate characteristics, water quality parameters had significant correlations; particularly, the substrate characteristics were more significantly related. In conclusion, the pool caused by artificial structures had negative effects on benthic macroinvertebrate communities thus leading to simplified stream habitats at upper stream ecosystems.

• Korean Journal of Microbiology
• /
• v.35 no.1
• /
• pp.28-34
• /
• 1999

Fragmentation vectors are used to analyze function and genomic structure of a gene of interest by creating deletion derivatives of large fragments of genomic DNA cloned as yeast artificial chromosomes (YACs). Herein, we developed a new hicistronic fragmentation vector that contains internal ribosomal entry sile (IRES) of encephalomyocarditis vin~s (EMCV) and $\beta$-galactosidase as a reporter gene. This vector system provides a novcl loo1 to analyze expression patterns of a gene of interest due to simultaneous expression of a target gene as well as $\beta$-galactosidase driven from a single message. In addition, the bicistronic fragmentation vector contains four rare-cutting restriction enzyme sites in the polycloning sites which can be used to conveniently insert any kinds of genes and therefore facilitates targeting DNA scgments into YAC by means of homologous recombination. This approach establishes a paradigm for manipulation of mammalian DNA segments and characterization of expression and regulatory regions of mammalian gene cloned as YAC.

• Journal of Information Technology Services
• /
• v.16 no.4
• /
• pp.137-148
• /
• 2017

Researches on the Artificial Intelligence has been explosively activated in various fields since the advent of AlphaGo. Particularly, researchers on the application of multi-layer neural network such as deep learning, and various machine learning algorithms are being focused actively. In this paper, we described a development of an artificial intelligence Janggi game based on reinforcement learning algorithm and MCTS (Monte Carlo Tree Search) algorithm with accumulated game data. The previous artificial intelligence games are mostly developed based on mini-max algorithm, which depends only on the results of the tree search algorithms. They cannot use of the real data from the games experts, nor cannot enhance the performance by learning. In this paper, we suggest our approach to overcome those limitations as follows. First, we collects Janggi expert's game data, which can reflect abundant real game results. Second, we create a graph structure by using the game data, which can remove redundant movement. And third, we apply the reinforcement learning algorithm and MCTS algorithm to select the best next move. In addition, the learned graph is stored by object serialization method to provide continuity of the game. The experiment of this study is done with two different types as follows. First, our system is confronted with other AI based system that is currently being served on the internet. Second, our system confronted with some Janggi experts who have winning records of more than 50%. Experimental results show that the rate of our system is significantly higher.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.7
• /
• pp.768-773
• /
• 2012

A branch-line hybrid using microstrip artificial transmission lines(ATLs) with slotted-triangular patches is proposed. The proposed artificial transmission line is compact in structure as well as easy to adjust the characteristic impedance and electrical length of equivalent transmission line by changing the slot's parameters; hence, it is useful for miniaturizing conventional transmission lines. The designed branch-line hybrid, because of the use of the right angled isosceles triangular shaped artificial transmission lines as building blocks, has no useless empty space, and hence optimally miniaturized. A fabricated 3 dB branch-line hybrid shows the coupling variation of ${\pm}0.5$ dB and the phase difference between two output ports of $91^{\circ}{\pm}4^{\circ}$ within 15 % bandwidth at 2.45 GHz center frequency. The size of proposed branch-line hybrid is only 38% of the conventional branch-line hybrid.