• Journal of the Society of Naval Architects of Korea
• /
• v.35 no.3
• /
• pp.79-87
• /
• 1998

The purpose of this paper is to develope a neurofuzzy system for a ship design which can determine efficiently design values e.g. principal dimensions and hull factors in a conceptual design. The neurofuzzy system for a ship design(NeFHull) applies a information about given input-output data to fuzzy theories and deals these fuzzificated values with neural networks, e.g. first, redefines normalized input-output data ad membership functions and then executes these fuzzficated information with backpropagation neural networks. We use a hybrid learning algorithm in the training of neural networks and examine the usefulness of suggested method through mathematical and mechanical examples.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.2
• /
• pp.180-187
• /
• 2010

In this paper, a global path generation method is suggested using multiple grid maps connected with the mission type of unmanned combat vehicle(UCV). In order to carry out a mission for UCV, it is essential to find a global path which is coincident with the characteristics of the mission. This can be done by considering various combat circumstances represented as grid maps such as velocity map, threat map and communication map. Cost functions of multiple grid maps are linearly combined and normalized to them simultaneously for the path generation. The proposed method is realized using $A^*$, a well known search algorithm, and cost functions are normalized in the ratio of the traverse time which is one of critical information should be provided with the operators using the velocity map. By the experiments, it is checked found global paths match with the mission type by reflecting input data of grid maps properly and the computation time is short enough to regenerate paths in real time as combat circumstances change.

• Journal of Welding and Joining
• /
• v.23 no.6
• /
• pp.61-66
• /
• 2005

A probabilistic approach for evaluating failure risk is suggested in this paper. Probabilistic fracture analyses were performed for a pressurized pipe of a Cr-Mo steel reflecting variation of material properties at high temperature. A crack was assumed to be located along the weld fusion line. Probability density functions of major variables were determined by statistical analyses of material creep and creep crack growth data measured by the previous experimental studies by authors. Distributions of these variables were implemented in Monte Carlo simulation of this study. As a fracture parameter for characterizing growth of a fusion line crack between two materials with different creep properties, $C_t$ normalized with $C^*$ was employed. And the elapsed time was also normalized with tT, Resultingly, failure probability as a function of operating time was evaluated fur various cases. Conventional deterministic life assessment result was turned out to be conservative compared with that of probabilistic result. Sensitivity analysis for each input variable was conducted to understand the most influencing variable to the analysis results. Internal pressure, creep crack growth coefficient and creep coefficient were more sensitive to failure probability than other variables.

• Structural Engineering and Mechanics
• /
• v.85 no.4
• /
• pp.469-484
• /
• 2023

A deep recursive bidirectional Cuda Deep Neural Network Long Short Term Memory (Bi-CuDNNLSTM) layer is recruited in this paper to predict the entire force time histories, and the corresponding hysteresis and backbone curves of reinforced concrete (RC) bridge piers using experimental fast and slow cyclic tests. The proposed stacked Bi-CuDNNLSTM layers involve multiple uncertain input variables, including horizontal actuator displacements, vertical actuators axial loads, the effective height of the bridge pier, the moment of inertia, and mass. The functional application programming interface in the Keras Python library is utilized to develop a deep learning model considering all the above various input attributes. To have a robust and reliable prediction, the dataset for both the fast and slow cyclic tests is split into three mutually exclusive subsets of training, validation, and testing (unseen). The whole datasets include 17 RC bridge piers tested experimentally ten for fast and seven for slow cyclic tests. The results bring to light that the mean absolute error, as a loss function, is monotonically decreased to zero for both the training and validation datasets after 5000 epochs, and a high level of correlation is observed between the predicted and the experimentally measured values of the force time histories for all the datasets, more than 90%. It can be concluded that the maximum mean of the normalized error, obtained through Box-Whisker plot and Gaussian distribution of normalized error, associated with unseen data is about 10% and 3% for the fast and slow cyclic tests, respectively. In recapitulation, it brings to an end that the stacked Bi-CuDNNLSTM layer implemented in this study has a myriad of benefits in reducing the time and experimental costs for conducting new fast and slow cyclic tests in the future and results in a fast and accurate insight into hysteretic behavior of bridge piers.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.133-134
• /
• 2014

Conventional stereo matching algorithms fail when they deal with anaglyph image as its input because anaglyph image does not have similar intensity on both view images. To ameliorate such problems, we propose a robust method to obtain accurate disparity maps. The novel Absolute Adaptive Normalized Cross Correlation (AANCC) for anaglyph data cost is introduced in this paper. Then, it is followed by occlusion detection and segmentation-based plane fitting to achieve accurate depth map acquisition. Experimental results confirm that the proposed anaglyph data cost is robust and gives accurate disparity maps.

• Computers and Concrete
• /
• v.10 no.4
• /
• pp.379-390
• /
• 2012

In this paper, the mechanical strength of different lightweight mortars made with 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 and 100 percentage of scoria instead of sand and 0.55 water-cement ratio and 350 $kg/m^3$ cement content is investigated. The experimental result showed 7.9%, 16.7% and 49% decrease in compressive strength, tensile strength and mortar density, respectively, by using 100% scoria instead of sand in the mortar. The normalized compressive and tensile strength data are applied for artificial neural network (ANN) generation using generalized regression neural network (GRNN). Totally, 90 experimental data were selected randomly and applied to find the best network with minimum mean square error (MSE) and maximum correlation of determination. The created GRNN with 2 input layers, 2 output layers and a network spread of 0.1 had minimum MSE close to 0 and maximum correlation of determination close to 1.

• Journal of Korea Water Resources Association
• /
• v.47 no.10
• /
• pp.935-943
• /
• 2014

Normalized Difference Vegetation Index (NDVI) used as input data for various hydrologic models plays a key role in understanding the variation of Hydrometeological parameters and Interaction between surface and atmosphere. Many studies have been conducted to estimate accurate remotely-sensed NDVI using spectral characteristics of vegetation. In this study, we conducted comparative analysis between Communication, Ocean and Meteorological Satellite and MOderate-Resolution Imaging Spectroradiometer (MODIS) NDVI. For comparison, Maximum Value Composite (MVC) was used to estimate 8-day and 16-day composite COMS NDVI. Both 8-day and 16-day COMS NDVI showed high statistical results compared with MODIS NDVI. Based on the results in this study, it can be concluded that COMS can be widely applicable for further ecological and hydrological studies.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.145-151
• /
• 2023

3D-CNN is one of the deep learning techniques for learning time series data. Such three-dimensional learning can generate many parameters, so that high-performance machine learning is required or can have a large impact on the learning rate. When learning dynamic hand-gestures in spatiotemporal domain, it is necessary for the improvement of the efficiency of dynamic hand-gesture learning with 3D-CNN to find the optimal conditions of input video data by analyzing the learning accuracy according to the spatiotemporal change of input video data without structural change of the 3D-CNN model. First, the time ratio between dynamic hand-gesture actions is adjusted by setting the learning interval of image frames in the dynamic hand-gesture video data. Second, through 2D cross-correlation analysis between classes, similarity between image frames of input video data is measured and normalized to obtain an average value between frames and analyze learning accuracy. Based on this analysis, this work proposed two methods to effectively select input video data for 3D-CNN deep learning of dynamic hand-gestures. Experimental results showed that the learning interval of image data frames and the similarity of image frames between classes can affect the accuracy of the learning model.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.18 no.3
• /
• pp.368-374
• /
• 2008

Multi-axial simulation table(MAST) is widely used in motor companies as the multi-axial excitor for vibration fatigue of target component, which provides the vibrational condition as close as the vehicle test. However, the vibration fatigue performance of target component can be guaranteed with MAST system only in case the input profile covers the required severity of the target component on field test. In this paper, the signal processing for multi-axial vibration fatigue test on vehicle component is presented, from the data acquisition of the target component to the derivation of input profile. To compare the severity of vibration condition between field and proving ground, the energy principle of a equivalent damage is proposed and then, it is determined the optimal combination of special events on proving ground using a sequential searching optimal algorithm. To explain the vibration methodology clearly, seat and door component of vehicle are selected as a example.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.7
• /
• pp.2998-3017
• /
• 2018

Massive (large-scale) MIMO (multiple-input multiple-output) is one of the key technologies in next-generation wireless communication systems. This paper proposes a high-performance low-complexity turbo receiver for SC-FDMA (single-carrier frequency-division multiple access) based MMIMO (massive MIMO) systems. Because SC-FDMA technology has the desirable characteristics of OFDMA (orthogonal frequency division multiple access) and the low PAPR (peak-to-average power ratio) of SC transmission schemes, the 3GPP LTE (long-term evolution) has adopted it as the uplink transmission to meet the demand high data rate and low error rate performance. The complexity of computing will be increased greatly in base station with massive MIMO (MMIMO) system. In this paper, a low-complexity adaptive turbo equalization receiver based on normalized minimal symbol-error-rate for MMIMO SC-FDMA system is proposed. The proposed receiver is with low complexity than that of the conventional turbo MMSE (minimum mean square error) equalizer and is also with better bit error rate (BER) performance than that of the conventional adaptive turbo MMSE equalizer. Simulation results confirm the effectiveness of the proposed scheme.