• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.210-210
• /
• 2000

In this paper, we have investigated the properties of multi-layer perceptron (MLP) for odour patterns classification and concentration estimation simultaneously. When the MLP may be has a fast convergence speed with small error and excellent mapping ability for classification, it can be possible to use for classification and concentration prediction of volatile chemicals simultaneously. However, the conventional MLP, which is back-Propagation of error based on the steepest descent method, was difficult to use for odour classification and concentration estimation simultaneously, because it is slow to converge and may fall into the local minimum. We adapted the Levenberg-Marquardt(LM) algorithm [4,5] having advantages both the steepest descent method and Gauss-Newton method instead of the conventional steepest descent method for the simultaneous classification and concentration estimation of odours. And, We designed the artificial odour sensing system(Electronic Nose) and applied LM-BP algorithm for classification and concentration prediction of VOC gases.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.11 no.6
• /
• pp.939-945
• /
• 2000

A cylindrical object buried in the lossy half space is reconstructed from the measured scattered fields above the lossy half space. The position, the size and the medium parameters i.e. relative dielectric constants and conductivity of the buried object as well as the medium parameters of the background lossy half space are obtained from the scattered fields by using the iterative inversion method and the optimization hybrid algorithm combining the genetic algorithm and the Levenberg-Marquardt algorithm. Illposedness of the inversion due to the measurement errors in the scattered fields are regularized by filtering out the evanescent modes in the spatial frequency spectrum domain.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.3
• /
• pp.439-444
• /
• 2016

The method for extracting the SPICE model parameter of Schottky diode under DC forward bias is proposed. A method for improving the accuracy of the SPICE model parameter at various temperatures is proposed. Three analysis steps according to the magnitude of the current is used in order to extract the parameters effectively. At each analysis step, initial parameters are calculated by using the current-voltage equations and the Levenberg-Marquardt analysis is proceeded. To verify the validity of the proposed method, the SPICE model parameters for the BAT45 and FSV1045 under DC forward bias is extracted. Schottky diode currents obtained from the proposed method shows the average relative error of 6.1% and 9% compared with the measured data for the BAT45 and FSV1045 sample at various temperatures.

• Steel and Composite Structures
• /
• v.7 no.5
• /
• pp.355-376
• /
• 2007

The purpose of this study is to predict and investigate the time-dependent creep behavior of composite materials. For this, firstly the evaluation method for the modulus of elasticity of whole fiber and matrix is presented from the limited information on fiber volume fraction using the singular value decomposition method. Then, the effects of fiber volume fraction on modulus of elasticity of GFRP are verified. Also, as a creep model, the nonlinear curve fitting method based on the Marquardt algorithm is proposed. Using the existing Findley's power creep model and the proposed creep model, the effect of fiber volume fraction on the nonlinear creep behavior of composite materials is verified. Then, for the time-dependent analysis of a composite material subjected to uniaxial tension and simple shear loadings, a user-provided subroutine UMAT is developed to run within ABAQUS. Finally, the creep behavior of center loaded beam structure is investigated using the Hermitian beam elements with shear deformation effect and with time-dependent elastic and shear moduli.

• Journal of Korea Water Resources Association
• /
• v.34 no.6
• /
• pp.701-711
• /
• 2001

Forecasting of water quality variation is not an easy process due to the complicated nature of various water quality factors and their interrelationships. The objective of this study is to test the applicability of neural network models to the forecasting of the water quality at Gongju station in Geum River. This is done by forecasting monthly water qualities such as DO, BOD, and TN, and comparing with those obtained by ARIMA model. The neural network models of this study use BP(Back Propagation) algorithm for training. In order to improve the performance of the training, the models are tested in three different styles ; MANN model which uses the Moment-Adaptive learning rate method, LMNN model which uses the Levenberg-Marquardt method, and MNN model which separates the hidden layers for judgement factors from the hidden layers for water quality data. the results show that the forecasted water qualities are reasonably close to the observed data. And the MNN model shows the best results among the three models tested

• The KIPS Transactions:PartB
• /
• v.10B no.7
• /
• pp.803-810
• /
• 2003

In this paper, we describe a fast image mosaic method for constructing a large-scale image with video image captured from cameras that are arranged in radial shape. In the first step, we adopt the phase correlation algorithm to estimate the horizontal and vertical displacement between two adjacent images. Secondly, we calculate the accurate transform matrix among those cameras with Levenberg-Marquardt method. In the last step, those images are stitched into one large scale image in real-time by applying the transform matrix to the texture mapping function of DirectX. The feature of the method is that we do not need to use special hardware devices or write machine-level programs for Implementing a real-time mosaic system since we use conventional graphic APIs (Application Programming Interfaces), DirectX for image synthesis process.

• Journal of Korea Water Resources Association
• /
• v.33 no.4
• /
• pp.505-515
• /
• 2000

THe purpose of this study is to forecast of runoff hydrographs according to rainfall event in a stream. The neural network theory as a hydrologic blackbox model is used to solve hydrological problems. The Back-Propagation(BP) algorithm by the Levenberg-Marquardt(LM) techniques and Radial Basis Function(RBF) network in Neural Network(NN) models are used. Runoff hydrograph is forecasted in Bocheongstream basin which is a IHP the representative basin. The possibility of a simulation for runoff hydrographs about unlearned stations is considered. The results show that NN models are performed to effective learning for rainfall-runoff process of hydrologic system which involves a complexity and nonliner relationships. The RBF networks consist of 2 learning steps. The first step is an unsupervised learning in hidden layer and the next step is a supervised learning in output layer. Therefore, the RBF networks could provide rather time saved in the learning step than the BP algorithm. The peak discharge both BP algorithm and RBF network model in the estimation of an unlearned are a is trended to observed values.

• Journal of Korea Water Resources Association
• /
• v.31 no.1
• /
• pp.45-57
• /
• 1998

The purpose of this paper is to develop a neural network model in order to forecast flood inflow into the reservoir that has the nature of uncertainty and nonlinearity. The model has the features of multi-layered structure and parallel multi-connections. To develop the model. backpropagation learning algorithm was used with the Momentum and Levenberg-Marquardt techniques. The former technique uses gradient descent method and the later uses gradient descent and Gauss-Newton method respectively to solve the problems of local minima and for the speed of convergency. Used data for learning are continuous fixed real values of input as well as output to emulate the real physical aspects. after learning process. a reservoir inflows forecasting model at flood period was constructed. The data for learning were used to calibrate the developed model and the results were very satisfactory. applicability of the model to the Chungju Mlultipurpose Reservoir proved the availability of the developed model.

• Korean Journal of Applied Biomechanics
• /
• v.23 no.1
• /
• pp.53-62
• /
• 2013

The methods of fitting a circle to measured data, geometric fit and algebraic fit, have been studied profoundly in various areas of science. However, they have not been applied exactly to a biomechanics discipline for locating the center of rotation of a human joint. The purpose of this study was to generalize the methods to fitting spheres to the points in 3-dimension, and to estimate the center of rotation of a hip joint by three of geometric fit methods(Levenberg-Marquardt, Landau, and Sp$\ddot{a}$th) and four of algebraic fit methods(Delogne-K${\aa}$sa, Pratt, Taubin, and Hyper). 1000 times of simulation experiments for flexion/extension and ad/abduction at an artificial hip joint with four levels of range of motion(10, 15, 30, and $60^{\circ}$) and three levels of angular velocity(30, 60, and $90^{\circ}$/s) were executed to analyze the responses of the estimated center of rotation. The results showed that the Sp$\ddot{a}$th estimate was very sensitive to the marker near the center of rotation. The bias of Delogne-K${\aa}$sa estimate existed in an even larger range of motion. The Levenberg-Marquardt algorithm of geometric fit and the Pratt of algebraic fit showed the best results. The combination of two methods, using the Pratt's estimate as initial values of the Levenberg-Marquardt algorithm, could be a candidate of more valid estimator.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.10 no.1 s.47
• /
• pp.33-39
• /
• 2006

Parameters including the seismic sources and the elastic wave propagation characteristics were analysed using the observed ground motions from 3 Kyoungju region earthquakes. The Levenberg-Marquardt algorithm was applied to invert all the variables non-linearly and simultaneously with S wave energy In frequency domain. Average stress drop of 3 events and local attenuation parameter ${\kappa}$ were estimated to about 48-bar and 0.0312 respectively. Regional attenuation parameter, Qo and ${\eta}$, were also estimated to be about 417 and 0.83. ${\kappa}$ values are much higher than that of EUS, even though smaller than that of WUS. All these values resultant from this study show that there are differences in some parameters of other studios due to differences in the governing equation. of acceleration motions