• ETRI Journal
• /
• v.15 no.2
• /
• pp.35-51
• /
• 1993

This paper presents function approximation based on nonparametric estimation. As an estimation model of function approximation, a three layered network composed of input, hidden and output layers is considered. The input and output layers have linear activation units while the hidden layer has nonlinear activation units or kernel functions which have the characteristics of bounds and locality. Using this type of network, a many-to-one function is synthesized over the domain of the input space by a number of kernel functions. In this network, we have to estimate the necessary number of kernel functions as well as the parameters associated with kernel functions. For this purpose, a new method of parameter estimation in which linear learning rule is applied between hidden and output layers while nonlinear (piecewise-linear) learning rule is applied between input and hidden layers, is considered. The linear learning rule updates the output weights between hidden and output layers based on the Linear Minimization of Mean Square Error (LMMSE) sense in the space of kernel functions while the nonlinear learning rule updates the parameters of kernel functions based on the gradient of the actual output of network with respect to the parameters (especially, the shape) of kernel functions. This approach of parameter adaptation provides near optimal values of the parameters associated with kernel functions in the sense of minimizing mean square error. As a result, the suggested nonparametric estimation provides an efficient way of function approximation from the view point of the number of kernel functions as well as learning speed.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.07a
• /
• pp.680-683
• /
• 2001

The output characteristics of step-down piezoelectric transformer is changed by a structure of layers. In this paper, we simulated output characteristics of multi-layer piezoelectric transformers with variation of output layers. Also, fabricated piezoelectric transformers were compared with simulated data. From simulated piezoelectric transformers, the output voltage decreased with increasing number of layers. From these results, piezoelectric transformers were made and the output electrical power of the transformers was measured at resonance frequency and at other frequency. The electrical power of transformers was measured on each transformer's resonance mode. However, measured value of 12-layed transformer's output power was smaller than that of 6-layered transformer's one. It is supposed that internal capacitance and reactance of the piezoelectric transformer's were effected in this result. Therefore we need to connect other road resistance and capacitance in output circuit, in order to increase electrical power of transformers.

• Journal of the Optical Society of Korea
• /
• v.18 no.2
• /
• pp.167-173
• /
• 2014

A novel multi-oxide layer structure for vertical cavity surface emitting laser (VCSEL) structures is proposed to achieve higher single mode output power. The structure has four oxide layers with different aperture sizes and thicknesses. The oxide layer thicknesses are optimized simultaneously to reach the highest single mode output power. A heuristic method is proposed for plotting the influence of these variable changes on the operation of optical output power. A comprehensive optical-electrical thermal-gain self-consistent VCSEL model is used to simulate the continuous-wave operation of the multi-layer oxide VCSELs. A comparison between optimized VCSELs with different structures is presented. The results show that by using multi-oxide layers with different thicknesses, higher single-mode optical output power could be achieved in comparison with multi-oxide layer structures with the same thicknesses.

• Journal of Korean Association for Spatial Structures
• /
• v.18 no.4
• /
• pp.69-80
• /
• 2018

In this study, an algorithm applying deep learning to the truss structures was proposed. Deep learning is a method of raising the accuracy of machine learning by creating a neural networks in a computer. Neural networks consist of input layers, hidden layers and output layers. Numerous studies have focused on the introduction of neural networks and performed under limited examples and conditions, but this study focused on two- and three-dimensional truss structures to prove the effectiveness of algorithms. and the training phase was divided into training model based on the dataset size and epochs. At these case, a specific data value was selected and the error rate was shown by comparing the actual data value with the predicted value, and the error rate decreases as the data set and the number of hidden layers increases. In consequence, it showed that it is possible to predict the result quickly and accurately without using a numerical analysis program when applying the deep learning technique to the field of structural analysis.

• Journal of Environmental Science International
• /
• v.13 no.7
• /
• pp.611-618
• /
• 2004

The Neural Network Models which mathematically interpret human thought processes were applied to resolve the uncertainty of model parameters and to increase the model's output for the streamflow forecast model. In order to test and verify the flood discharge forecast model eight flood events observed at Kumho station located on the midstream of Kumho river were chosen. Six events of them were used as test data and two events for verification. In order to make an analysis the Levengerg-Marquart method was used to estimate the best parameter for the Neural Network model. The structure of the model was composed of five types of models by varying the number of hidden layers and the number of nodes of hidden layers. Moreover, a logarithmic-sigmoid varying function was used in first and second hidden layers, and a linear function was used for the output. As a result of applying Neural Networks models for the five models, the N10-6model was considered suitable when there is one hidden layer, and the Nl0-9-5model when there are two hidden layers. In addition, when all the Neural Network models were reviewed, the Nl0-9-5model, which has two hidden layers, gave the most preferable results in an actual hydro-event.

• Journal of Korean Society on Water Environment
• /
• v.21 no.1
• /
• pp.21-28
• /
• 2005

The training and prediction modeling using an artificial neural network was implemented to predict the turbidity of treated water as well as to estimate the optimized feed concentration of polyaluminium chloride (PACl) in a water treatment plant. The parameters used in the input layers were pH, temperature, turbidity and alkalinity, while those in output layers were PACl and turbidity of treated water. Levenberg-Marquadt method of feedforward back-propagation perceptron in the neural network toolbox of MATLAB program was used in this study. Correlation coefficients of the training data with the measured data were 0.9997 for PACl and 0.6850 for turbidity and those of the testing data with measured data were 0.9140 for PACl and 0.3828 for turbidity, when four parameters at input layer, 12-12 nodes each at both the first and the second hidden layers, and two parameters(PACl and turbidity) at output layer were used. Although the predictability of PACl was improved, compared to that of the previous studies to use the only coagulant dose as output layer, turbidity in treated water could not be predicted well. Acquisition of more data through several years obtained with the advanced on-line measuring system could make the artificial neural network useful and practical in actual water treatment plants.

• Structural Engineering and Mechanics
• /
• v.67 no.1
• /
• pp.33-43
• /
• 2018

The dynamic output-feedback robust control method based on linear matrix inequality (LMI) method is presented for suppressing vibration response of a functionally graded material (FGM) beam with piezoelectric actuator/sensor layers in this paper. Based on the reduced model obtained by using direct mode truncation, the linear fractional state space representation of a piezoelectric FGM beam with material properties varying through the thickness is developed by considering both the inherent uncertainties in constitution material properties as well as material distribution and the model error due to mode truncation. The dynamic output-feedback robust H-infinity control law is implemented to suppress the vibration response of the piezoelectric FGM beam and the LMI method is utilized to convert control problem into convex optimization problem for efficient computation. In numerical studies, the flexural vibration control of a cantilever piezoelectric FGM beam is considered to investigate the accuracy and efficiency of the proposed control method. Compared with the efficient linear quadratic regulator (LQR) widely employed in literatures, the proposed robust control method requires less control voltage applied to the piezoelectric actuator in the case of same control performance for the controlled closed-loop system.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2000.04a
• /
• pp.529-533
• /
• 2000

This paper present a new sort of multilayer piezoelectric ceramic transformer for AC-adapt. This piezoelectric transformer operates n the second thickness extensional vibration mode. The output voltage of the multilayer piezoelectric transformer was simulated using ANSYS. As results, the maximum stress appeared at the two points where one is the middle of input and the other is middle of output side in Second thickness extensional vibration mode. And output voltage was inversely decreased by increasing number of output layers.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.34D no.12
• /
• pp.69-75
• /
• 1997

A Si-channle MESFET using .delta.-doped layers was designed and the considerable enhancement of the current driving capability of the device was observed by simulation. The channel consists of double .delta.-doped layers separated by a low-doped spacer. Cariers are spilt from the .delta.-doped layers and are accumulated in the spacer. The saturation current is enhanced by the contribution of the carriers in the spacer. Among the design parameters that affect the peformance of the device, the thickness of the spacer and the ratio of the doping concentrations of the two .delta.-doped layers were studied. The spacer thickenss of 300~500.angs. and the doping ratio of 3~4 were shown to be the optimized values. The saturation current was observed to be increased by 75% compared with a bulk-channel MESFET. The performances of transconductance, output resistance, and subthreshold swing were also enhanced.

• Journal of Ocean Engineering and Technology
• /
• v.28 no.1
• /
• pp.55-63
• /
• 2014

In this paper, we propose an threshold-based Schur algorithm for estimating the media characteristics of sub-bottom multi-layers by using the signal generated by a parametric array transducer. We use the KZK model to generate a parametric array signal, and use the proposed threshold-based Schur algorithm for estimating the reflection coefficients of multiple sea bottom layers. Using computer simulation, we verify that the difference frequency component generated by the KZK model prevails over the signals of primary frequencies at long range. For the simulation, we use the transmit signal generated by the KZK and the reflected signal obtained from a lattice filter model for the seawater and sub-bottom of multi-level non-homogeneous layers. Through the simulation, we verify that the proposed threshold-based Schur algorithm can give much more accurate and efficient estimates of the reflection coefficients than methods using received signal, matched filter output signal, and normal Schur algorithm output.