• The Journal of Engineering Geology
• /
• v.9 no.3
• /
• pp.187-206
• /
• 1999

A computer program to numerically predict the permeability tensor of fractured rocks is developed using information on discontinuities which Borehole Televiewer and Borehole Image Processing System (BIPS) provide. It uses orientation and thickness of a large number of discontinuities as input data, and calculates relative values of the 9 elements consisting of the permeability tensor by the formulation based on the EPM model, which regards a fractured rock as a homogeneous, anisotropic porous medium. In order to assess feasibility of the program on field sites, the numerically calculated tensor was obtained using BIPS logs and compared to the results of pumping test conducted in the boreholes of the study area. The degree of horizontal anisotropy and the direction of maximum horizontal permeability are 2.8 and $N77^{\circ}CE$, respectively, determined from the pumping test data, while 3.0 and $N63^{\circ}CE$ from the numerical analysis by the developed program. Disagreement between two analyses, especially for the principal direction of anisotropy, seems to be caused by problems in analyzing the pumping test data, in applicability of the EPM model and the cubic law, and in simplified relationship between the crack size and aperture. Aside from these problems, consideration of hydraulic parameters characterizing roughness of cracks and infilling materials seems to be required to improve feasibility of the proposed program. Three-dimensional assessment of its feasibility on field sites can be accomplished by conducting a series of cross-hole packer tests consisting of an injecting well and a monitoring well at close distance.

• International journal of advanced smart convergence
• /
• v.2 no.2
• /
• pp.7-9
• /
• 2013

Current physics-based simulation is an important tool in the fluid animation. However some problems require a new change to current research trends which depend only on the simulation. The ultimate goal of this project is to obtain information of flow example, analyze an example through machine learning and the novel fluid animation reconfigure without physical simulation.

• The Journal of Engineering Geology
• /
• v.10 no.1
• /
• pp.13-25
• /
• 2000

A three-dimensional discrete fracture network model based on probabilistic characteristics of fracture geometry and transmissivity was designed to calculate the conductivity tensor and to estimate theanisotropy of conductivity. The conductivities, $K_p$, obtained from the numerical simulation of single-holepacker test corresponded well to those from the field tests. From this, it can be concluded that thefracture network model designed in this study can represent hydraulic characteristics of in-situ fractured rock mass. Block-scale conductivities, $K_b$, estimated from the modelling of steady-state flow through the REV-scale block were ranged between the arithmetic mean and harmonic mean of theconductivity estimates from packer tests. The conductivity along north-south direction was 1.4 timesgreater than that along the east-west direction. It was concluded that the anisotropy of conductivitywas insignificant. It was also found that there was a little correlation between $K_b$ and $K_p$. This would be to that the conductivities from the packer test simulation was strongly dependent on thetransmissivity and the number of fractures within the packer test intervals.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.3
• /
• pp.165-171
• /
• 2019

The study of the computer architecture has shown that major improvements in price-to-energy performance stems from domain-specific hardware development. This paper analyzes the tensor processing unit (TPU) ASIC which can accelerate the reasoning of the artificial neural network (NN). The core device of the TPU is a MAC matrix multiplier capable of high-speed operation and software-managed on-chip memory. The execution model of the TPU can meet the reaction time requirements of the artificial neural network better than the existing CPU and the GPU execution models, with the small area and the low power consumption even though it has many MAC and large memory. Utilizing the TPU for the tensor flow benchmark framework, it can achieve higher performance and better power efficiency than the CPU or CPU. In this paper, we analyze TPU, simulate the Python modeled OpenTPU, and synthesize the matrix multiplication unit, which is the key hardware.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.1
• /
• pp.87-95
• /
• 2001

A numerical study was made of heat transfer in locally-forced turbulent separated and reattaching flow over a backward-facing step. The local forcing was given to the flow by means of sinusoidally oscillating jet from a separation line. A Rhee and Sung version of the unsteady $\kappa$-$\varepsilon$-f(sub)u model and the diffusivity tensor heat transfer model were employed. The Reynolds number was fixed at Re(sub)H=33,000 and the forcing frequency was varied in the range 0$\leq$fH/U(sub)$\infty$$\leq$2. The condition of constant heat flux was imposed at the bottom wall. The predicted results were compared and validated with the experimental data of Chun and Sung and Vogel and Eaton. The enhancement of heat transfer in turbulent separated and reattaching flow by local forcing was evaluated and analyzed.

• KIISE Transactions on Computing Practices
• /
• v.23 no.7
• /
• pp.440-445
• /
• 2017

Due to the advance of e-commerce systems, the number of people using online shopping and products has significantly increased. Therefore, the need for an accurate recommendation system is becoming increasingly more important. Recurrent neural network is a deep-learning algorithm that utilizes sequential information in training. In this paper, an evaluation is performed on the application of recurrent neural networks to recommendation systems. We evaluated three recurrent algorithms (RNN, LSTM and GRU) and three optimal algorithms(Adagrad, RMSProp and Adam) which are commonly used. In the experiments, we used the TensorFlow open source library produced by Google and e-commerce session data from RecSys Challenge 2015. The results using the optimal hyperparameters found in this study are compared with those of RecSys Challenge 2015 participants.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.11
• /
• pp.9-16
• /
• 2020

The TA-50 aircraft is conducting simulated training on various situations, including air-to-air and air-to-ground fire training, in preparation for air warfare. It is also used for pilot training before actual deployment. However, the TA-50 does not have the ability to operate smart weapon forces, limiting training. Therefore, the purpose of this study is to implement the TA-50 aircraft to enable virtual training of one of the smart weapons, the Point Direct Attack Munition (JDAM). First, JDAM functions implemented in FA-50 aircraft, a model similar to TA-50 aircraft, were analyzed. In addition, since functions implemented in FA-50 aircraft cannot be directly utilized by source code, algorithms were extracted using machine learning techniques(TensorFlow). The implementation of this function is expected to enable realistic training without actually having to be armed. Finally, based on the results of this study, we would like to propose ways to supplement the limitations of the research so that it can be implemented in the same way as it is.

• Journal of Convergence for Information Technology
• /
• v.11 no.11
• /
• pp.38-43
• /
• 2021

As the number of augmented reality games using augmented reality technology increases, the demands of users are also increasing. Game technologies used in augmented reality games are mainly games using MARKER, MARKERLESS, GPS, etc. Games using this technology can augment the background and other objects. To solve this problem, we want to help develop augmented reality games by analyzing objects in the background, which is an important element of augmented reality. To analyze the background in the augmented reality game, the background object was analyzed by applying a deep learning model using TensorFlow Lite in the UNITY engine. Using this result, we obtained the result that augmented objects can be placed in the game according to the types of objects analyzed in the background. By utilizing this research, it will be possible to develop advanced augmented reality games by augmenting objects that fit the background.

• Korean Journal of Agricultural Science
• /
• v.46 no.2
• /
• pp.381-393
• /
• 2019

A smart farm is a system that combines information and communication technology (ICT), internet of things (IoT), and agricultural technology that enable a farm to operate with minimal labor and to automatically control of a greenhouse environment. Machine learning based on recently data-driven techniques has emerged with big data technologies and high-performance computing to create opportunities to quantify data intensive processes in agricultural operational environments. This paper presents research on the application of machine learning technology to diagnose the growth status of crops and predicting the harvest time of strawberries in a greenhouse according to image processing techniques. To classify the growth stages of the strawberries, we used object inference and detection with machine learning model based on deep learning neural networks and TensorFlow. The classification accuracy was compared based on the training data volume and training epoch. As a result, it was able to classify with an accuracy of over 90% with 200 training images and 8,000 training steps. The detection and classification of the strawberry maturities could be identified with an accuracy of over 90% at the mature and over mature stages of the strawberries. Concurrently, the experimental results are promising, and they show that this approach can be applied to develop a machine learning model for predicting the strawberry harvesting time and can be used to provide key decision support information to both farmers and policy makers about optimal harvest times and harvest planning.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.34 no.12
• /
• pp.85-94
• /
• 2018

The purpose of this paper was to propose a model that recognizes potential users' emotional response toward design by classifying Electroencephalography(EEG). Studies in neuroscience and psychology have made an effort to recognize subjects' emotional response by analyzing EEG data. And this approach has been adopted in design since it is critical to monitor users' subjective response in the preface of design. Moreover, the building design process cannot be reversed after construction, recognizing clients' affection toward design alternatives plays important role. An experiment was conducted to record subjects' EEG data while they view their most/least liked images of small-house designs selected by them among the eight given images. After the recording, a subjective questionnaire, PANAS, was distributed to the subjects in order to describe their own affection score in quantitative way. Google TensorFlow was used to build and train the model. Dataset for model training and testing consist of feature columns for recorded EEG data and labels for the questionnaire results. After training and testing, the measured accuracy of the model was 0.975 which was higher than the other machine learning based classification methods. The proposed model may suggest one quantitative way of evaluating design alternatives. In addition, this method may support designer while designing the facilities for people like disabled or children who are not able to express their own feelings toward alternatives.