• Journal of the Korean Operations Research and Management Science Society
• /
• v.31 no.3
• /
• pp.41-51
• /
• 2006

Clustering is to group similar objects into clusters. Until now there are a lot of approaches using Self-Organizing feature Maps (SOFMS) But they have problems with a small output-layer nodes and initial weight. For example, one of them is a one-dimension map of c output-layer nodes, if they want to make c clusters. This approach has problems to classify elaboratively. This Paper suggests one-dimensional output-layer nodes in SOFMs. The number of output-layer nodes is more than those of clusters intended to find and the order of output-layer nodes is ascending in the sum of the output-layer node's weight. We un find input data in SOFMs output node and classify input data in output nodes using Euclidean distance. The proposed algorithm was tested on well-known IRIS data and TSPLIB. The results of this computational study demonstrate the superiority of the proposed algorithm.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 1999.11a
• /
• pp.358-361
• /
• 1999

This paper describes on the fabrication and characteristics of hot-film type micro-flowsensors integrated with Pt-RTD\`s and micro-heater on the Si substrate, in which MgO thin-films were used as medium layer in order to improve adhesion of Pt thin-films to SiO$_2$ layer The MgO layer improved adhesion of Pt thin-films to SiO$_2$` layer without any chemical reactions to Pt thin-films under high as gas flow rate and its conductivity increased due to increase of heat-loss from sensor to external. Output voltage was 82 mV at N2 flow rate of 2000 sccm/min, heating power of 1.2W. The respons time was about 100 msec when input flow was step-input

• Journal of the Korean Ceramic Society
• /
• v.36 no.9
• /
• pp.937-945
• /
• 1999

In an effort to protect a RBSC(reaction -bonded SiC) tube SiC films from methyltrichlorosilane(MTS) by low pressure chemical vapor deposition were deposited in hydrogen atmosphere on the RBSC(reaction-bonded SiC) substrates over a range of input gas ratio(${\alpha}=P_{H2}/P_{MTS}=Q_{H2}/Q_{MTS}$=1 to 10) and deposition temperatures(T=1050~1300$^{\circ}C$). At the temper-ature of 1250$^{\circ}C$ the growth rate of SiC films increased and then decreased with decreasing the input gas ratio. The microstructure of SiC films was changed from granular type structure exhibiting (111) preferred orientation in the high input gas ratios to faceted columnar grain structure showing (220) in the low input gas ratios. The similar microstructure change was obtained by increasing the deposition temperature. These results were closely related to a change of deposition mechanism. Double layer structure having granular type and faceted ciolumnar grain structure from the manipulation of mechanism. Double layer structure having granular type and faceted columnar grain structure from the manipulation of the input gas ratio without changing the deposition temperatue was successfully fabricated through in -site process.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1759-1771
• /
• 2017

In this paper, an improved continuous integral variable structure systems(ICIVSS) with the prescribed control performance is designed for simple regulation controls of uncertain general linear systems. An integral sliding surface with an integral state having a special initial condition is adopted for removing the reaching phase and predetermining the ideal sliding trajectory from a given initial state to the origin in the state space. The ideal sliding dynamics of the integral sliding surface is analytically obtained and the solution of the ideal sliding dynamics can predetermine the ideal sliding trajectory(integral sliding surface) from the given initial state to the origin. Provided that the value of the integral sliding surface is bounded by certain value by means of the continuous input, the norm of the state error to the ideal sliding trajectory is analyzed and obtained in Theorem 1. A corresponding discontinuous control input with the exponential stability is proposed to generate the perfect sliding mode on the every point of the pre-selected sliding surface. For practical applications, the discontinuity of the VSS control input is approximated to be continuous based on the proposed modified fixed boundary layer method. The bounded stability by the continuous input is investigated in Theorem 3. With combining the results of Theorem 1 and Theorem 3, as the prescribed control performance, the pre specification on the error to the ideal sliding trajectory is possible by means of the boundary layer continuous input with the integral sliding surface. The suggested algorithm with the continuous input can provide the effective method to increase the control accuracy within the boundary layer by means of the increase of the $G_1$ gain. Through an illustrative design example and simulation study, the usefulness of the main results is verified.

• Journal of radiological science and technology
• /
• v.30 no.4
• /
• pp.419-426
• /
• 2007

In this paper, the input pattern vectors are extracted and the learning algorithms is designed to recognize taste(bitter, sweet, sour and salty) pattern vectors. The signal intensity of taste are used to compose the input pattern vectors. The SOM(Self Organizing Maps) algorithm for taste pattern recognition is used to learn initial reference vectors and the ot-star learning algorithm is used to determine the class of the output neurons of the sunclass layer. The weights of the proposed algorithm which is between the input layer and the subclass layer can be learned to determine initial reference vectors by using SOM algorithm and to learn reference vectors by using LVQ(Learning Vector Quantization) algorithm. The pattern vectors are classified into subclasses by neurons in the subclass layer, and the weights between subclass layer and output layer are learned to classify the classified subclass, which is enclosed a class. To classify the pattern vectors, the proposed algorithm is simulated with ones of the conventional LVQ, and it is confirmed that the proposed learning method is more successful classification than the conventional LVQ.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.5 s.43
• /
• pp.95-103
• /
• 2006

In this paper, the input pattern vectors are extracted and the learning algorithms is designed to recognize EEG pattern vectors. The frequency and amplitude of alpha rhythms and beta rhythms are used to compose the input pattern vectors. And the algorithm for EEG pattern recognition is used SOM to learn initial reference vectors and out-star learning algorithm to determine the class of the output neurons of the subclass layer. The weights of the proposed algorithm which is between the input layer and the subclass layer can be learned to determine initial reference vectors by using SOM algorithm and to learn reference vectors by using LVQ algorithm, and pattern vectors is classified into subclasses by neurons which is being in the subclass layer, and the weights between subclass layer and output layer is learned to classify the classified subclass, which is enclosed a class. To classify the pattern vectors of EEG, the proposed algorithm is simulated with ones of the conventional LVQ, and it was a confirmation that the proposed learning method is more successful classification than the conventional LVQ.

• Structural Engineering and Mechanics
• /
• v.77 no.1
• /
• pp.47-56
• /
• 2021

This paper deals with damage detection using a Gapped Smoothing Method (GSM) combined with deep learning. Convolutional Neural Network (CNN) is a model of deep learning. CNN has an input layer, an output layer, and a number of hidden layers that consist of convolutional layers. The input layer is a tensor with shape (number of images) × (image width) × (image height) × (image depth). An activation function is applied each time to this tensor passing through a hidden layer and the last layer is the fully connected layer. After the fully connected layer, the output layer, which is the final layer, is predicted by CNN. In this paper, a complete machine learning system is introduced. The training data was taken from a Finite Element (FE) model. The input images are the contour plots of curvature gapped smooth damage index. A free-free beam is used as a case study. In the first step, the FE model of the beam was used to generate data. The collected data were then divided into two parts, i.e. 70% for training and 30% for validation. In the second step, the proposed CNN was trained using training data and then validated using available data. Furthermore, a vibration experiment on steel damaged beam in free-free support condition was carried out in the laboratory to test the method. A total number of 15 accelerometers were set up to measure the mode shapes and calculate the curvature gapped smooth of the damaged beam. Two scenarios were introduced with different severities of the damage. The results showed that the trained CNN was successful in detecting the location as well as the severity of the damage in the experimental damaged beam.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.148-154
• /
• 2021

Text-Independent speaker verification needs to extract text-independent speaker embedding to improve generalization performance. However, deep neural networks that depend on training data have the potential to overfit text information instead of learning the speaker information when repeatedly learning from the identical time series. In this paper, to prevent the overfitting, we propose a segment unit shuffling layer that divides and rearranges the input layer or a hidden layer along the time axis, thus mixes the time series information. Since the segment unit shuffling layer can be applied not only to the input layer but also to the hidden layers, it can be used as generalization technique in the hidden layer, which is known to be effective compared to the generalization technique in the input layer, and can be applied simultaneously with data augmentation. In addition, the degree of distortion can be adjusted by adjusting the unit size of the segment. We observe that the performance of text-independent speaker verification is improved compared to the baseline when the proposed segment unit shuffling layer is applied.

• Proceedings of the KIEE Conference
• /
• 1991.11a
• /
• pp.39-42
• /
• 1991

The Application of Artificial Neural Network(ANN) to forecast a load in a power system is investigated. The load forecasting is important in the electric utility industry. This technique, methodology based on the fact that parallel structure can process very fast much information is a promising approach to a load forecasting. ANN that is highly interconnected processing element in a hierachy activated by the each input. The load pattern can be divided distinctively into two patterns, that is, weekday and weekend. ANN is composed of a input layer, several hidden layers, and a output layer and the past data is used to activate input layer. The output of ANN is the load forecast for a given day. The result of this simulation can be used as a reference to a electric utility operation.

• Journal of Bio-Environment Control
• /
• v.1 no.2
• /
• pp.162-168
• /
• 1992

As the automation of nutrient solution management proceeds in the field of hydroponics, effective supporting systems to manage the nutrient solution by computer become needed. This study was attempt to predict the EC of nutrient solution using the neural networks. The multilayer perceptron consisting of 3 layers with the back propagation learning algorithm was selected for EC prediction, of which nine variables in the input layer were the concentrations of each ion and one variable in the output layer the EC of nutrient solution. The meq unit in ion concentration was selected fir input variable in the input layer. After the 10,000 learning sweeps with 108 sample data, the comparison of predicted and measured ECs for 72 test data showed good agreements with the correlation coefficient of 0.998. In addition, the predicted ECs by neural network showed relatively equal or closer to the measured ones than those by current complicated models.