• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.878-882
• /
• 2004

In this paper, we propose a hierarchical hybrid neural network for detecting faults of induction motor. Implementing the classifier based on the input and output data, we apply appropriate transform and classification method at each step. In the proposed method, after obtaining the current of state of motor for each period, we transform it by Principle Component Analysis(PCA) to reduce its dimension. Before the training process, we use the conditional Fuzzy C-means(FCM) for obtaining the initial parameters of neural network for more effective learning procedure. From the various simulations, we find that the proposed method shows better performance to detect and diagnosis of induction motor and compare than other methods.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2004.03b
• /
• pp.442-449
• /
• 2004

This paper describes a new concept of finite element analysis, which is based on neural network based material models (NNCMs) without invoking any pre-chosen mathematical framework. NNCMs have several advantages over conventional constitutive models (CCMs) and once plugged in a finite element (FE) engine, can be used for FE analysis in a manner similar to CCMs. The paper demonstrates a FE framework in which NNCMs are incorporated and also proposes a strategy for data enhancement by invoking the assumption of isotropy of the material. It is shown through some illustrative examples that this provides a better training environment for a generalized NNCM in which stress and strain components are used as effects and cause. Form this study, it appears that there is a prima facia case for developing NNCMs for materials for which mathematical theories become too complex and a large number of material parameters and constants have to be identified or determined.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.5C
• /
• pp.601-610
• /
• 2004

In this paper, the performance of an adaptive array antenna using Constant Modulus Algorithm (CMA) based on the signal structure for the IMT-2000 3GPP specification reverse link of an asynchronous direct sequence code division multiple access (DS-CDMA) system are evaluated. In addition, the performance is compared with the array antenna using Least Mean Square (LMS) based on the training signal. The simulation parameters such as the number of multipath, mu10pa1h intensity profiles between path, spreading gain and multiuser etc., are considered in the Monte Carlo simulation. Simulation results demonstrate an adaptive array antenna using CMA may give more capacity gain than the amy antenna using LMS in the case of multipath fading channel.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.32A no.1
• /
• pp.14-25
• /
• 1995

The main contribution of this paper can be summarized in two items as follws. Firstly, a modelling of mobile communication channel with Rayleigh fading characteristics is presented. Actual signal environments can be approximated as being real measurements by a proper mathematical representation of fluctuation of channel parameters due to Doppler effect, that is determined by the relative speed between transmitter and receiver, and noises, that vary at each sampling time. Secondly, an alternative procedure of synthesizing an adaptive equalizers is presented for recovering original signals that have been corrupted through the modelled channel. In order to compute the optimal tap coefficients for a high speed data(512 k symbol/sec) on a real-time basis, the CGM that guarantees fast and stable convergency is adopted during the training period of each frame. The coefficients obtained by the CGM are used as initial values for the LMS algorithm to trace the optimal coefficients during the data period that vary at each sampling time due to the mobility and noise at the receiver. Using the modelling presented in this paper, distributions of received signal power in various signal environments are demonstrated. The performance of the eqalizer proposed in this paper is shown as a function of BER under the various signal circumstances of mobile communications.

• BMB Reports
• /
• v.45 no.4
• /
• pp.259-264
• /
• 2012

Accumulation of modified nucleotides is defective to various cellular processes, especially those involving DNA and RNA. To be viable, organisms possess a number of (deoxy)nucleotide phosphohydrolases, which hydrolyze these nucleotides removing them from the active NTP and dNTP pools. Deamination of purine bases can result in accumulation of such nucleotides as ITP, dITP, XTP and dXTP. E. coli RdgB has been characterised as a deoxyribonucleoside triphosphate pyrophosphohydrolase that can act on these nucleotides. S. cerevisiae homologue encoded by YJR069C was purified and its (d)NTPase activity was assayed using fifteen nucleotide substrates. ITP, dITP, and XTP were identified as major substrates and kinetic parameters measured. Inhibition by ATP, dATP and GTP were established. On the basis of experimental and published data, modelling and simulation of ITP, dITP, XTP and dXTP metabolism was performed. (d)ITP/(d)XTPase is a new example of enzyme with multiple substrate-specificity demonstrating that multispecificity is not a rare phenomenon

• Structural Engineering and Mechanics
• /
• v.54 no.2
• /
• pp.337-362
• /
• 2015

This paper presents a feasibility study on structural damage alarming and localization of long-span cable-supported bridges using multi-novelty indices formulated by monitoring-derived modal parameters. The proposed method which requires neither structural model nor damage model is applicable to structures of arbitrary complexity. With the intention to enhance the tolerance to measurement noise/uncertainty and the sensitivity to structural damage, an improved novelty index is formulated in terms of auto-associative neural networks (ANNs) where the output vector is designated to differ from the input vector while the training of the ANNs needs only the measured modal properties of the intact structure under in-service conditions. After validating the enhanced capability of the improved novelty index for structural damage alarming over the commonly configured novelty index, the performance of the improved novelty index for damage occurrence detection of large-scale bridges is examined through numerical simulation studies of the suspension Tsing Ma Bridge (TMB) and the cable-stayed Ting Kau Bridge (TKB) incurred with different types of structural damage. Then the improved novelty index is extended to formulate multi-novelty indices in terms of the measured modal frequencies and incomplete modeshape components for damage region identification. The capability of the formulated multi-novelty indices for damage region identification is also examined through numerical simulations of the TMB and TKB.

• Computers and Concrete
• /
• v.21 no.1
• /
• pp.47-54
• /
• 2018

Artificial neural network models can be successfully used to simulate the complex behavior of many problems in civil engineering. As compared to conventional computational methods, this popular modeling technique is powerful when the relationship between system parameters is intrinsically nonlinear, or cannot be explicitly identified, as in the case of concrete behavior. In this investigation, an artificial neural network model was developed to assess the residual compressive strength of self-compacted concrete at elevated temperatures ($20-900^{\circ}C$) and various relative humidity conditions (28-99%). A total of 332 experimental datasets, collected from available literature, were used for model calibration and verification. Data used in model development incorporated concrete ingredients, filler and fiber types, and environmental conditions. Based on the feed-forward back propagation algorithm, systematic analyses were performed to improve the accuracy of prediction and determine the most appropriate network topology. Training, testing, and validation results indicated that residual compressive strength of self-compacted concrete, exposed to high temperatures and relative humidity levels, could be estimated precisely with the suggested model. As illustrated by statistical indices, the reliability between experimental and predicted results was excellent. With new ingredients and different environmental conditions, the proposed model is an efficient approach to estimate the residual compressive strength of self-compacted concrete as a substitute for sophisticated laboratory procedures.

• Journal of IKEEE
• /
• v.17 no.3
• /
• pp.241-247
• /
• 2013

This paper presents the design of a System on Programmable Chip (SoPC) based on Field Programmable Gate Array (FPGA) for speech recognition in which Mel-Frequency Cepstral Coefficients (MFCC) for speech feature extraction and Vector Quantization for recognition are used. The implementing process of the speech recognition system undergoes the following steps: feature extraction, training codebook, recognition. In the first step of feature extraction, the input voice data will be transformed into spectral components and extracted to get the main features by using MFCC algorithm. In the recognition step, the obtained spectral features from the first step will be processed and compared with the trained components. The Vector Quantization (VQ) is applied in this step. In our experiment, Altera's DE2 board with Cyclone II FPGA is used to implement the recognition system which can recognize 64 words. The execution speed of the blocks in the speech recognition system is surveyed by calculating the number of clock cycles while executing each block. The recognition accuracies are also measured in different parameters of the system. These results in execution speed and recognition accuracy could help the designer to choose the best configurations in speech recognition on SoPC.

• Bulletin of the Korean Chemical Society
• /
• v.32 no.12
• /
• pp.4341-4346
• /
• 2011

Impact sensitivity, one of the most important screening factors for novel high energy density materials (HEDMs), was predicted by use of quantitative structure-property relationship (QSPR) based on the electrostatic potential (ESP) values calculated on the van der Waals molecular surface (MSEP). Among various 3D descriptors derived from MSEP, we utilized total and positive variance of MSEP, and devised a new QSPR equation by combining three other parameters. We employed 37 HEDMs bearing a benzene scaffold and nitro substituents, which were also utilized by Rice and Hare. All the molecular structures were optimized at the B3LYP/6-31G(d) level of theory and confirmed as minima by the frequency calculations. Our new QSPR equation provided a good result to predict the impact sensitivities of the molecules in the training set including zwitterionic molecules.

• Journal of the korean society of oceanography
• /
• v.33 no.1-2
• /
• pp.8-17
• /
• 1998

Paleoceanographic history of the East Sea is reconstructed based on several environmental parameters (coarse fraction content, planktonic foraminifera/benthic foraminifera ratio, fragmentation and assemblages of planktonic foraminifera, and coiling ratio of Neogloboquadrina pa-chyderma, etc,) of the late Quaternary sediments obtained from the Ulleung Basin. N. pa-chydeyma and Globigerina bulloides are dominant species (greater than 90% in abundance)among the total planktonic foraminifera assemblages in the late Quaternary sediments. The benthic foraminifera rarely occurred throughout the cores. Sinistrally-coiled specimens of N. pa-chyderma representing cold water temperature are observed more abundantly than dextrally-coil-ed ones. In addition, the sinistrally-coiled N, pachydeyma showed more the amount at the lower part of the cores than at the upper part suggesting the restriction of the Tsushima Warm Current into the East Sea during glacial period. G. bulloides, a species representative of upwelling condition, shows more abundant occurrence in the sediments of Core 941013 than those of Core 941006. This implies that Core 941013 is more influenced by upwelling than Core 941006. The upper part of the two cores contain more fragmentation of planktonic foraminifera suggesting significant dissolution by corrosive bottom wafer. Ascending CCD also played an important role for the absence of planktonic foraminifera at the upper part of the cores.