• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.5
• /
• pp.755-765
• /
• 2015

This paper is concerned with Interval Type-2 Radial Basis Function Neural Network classifier realized with the aid of Census Transform(CT) and (2D)2LDA methods. CT is considered to improve performance of face recognition in a variety of illumination variations. (2D)2LDA is applied to transform high dimensional image into low-dimensional image which is used as input data to the proposed pattern classifier. Receptive fields in hidden layer are formed as interval type-2 membership function. We use the coefficients of linear polynomial function as the connection weights of the proposed networks, and the coefficients and their ensuing spreads are learned through Conjugate Gradient Method(CGM). Moreover, the parameters such as fuzzification coefficient and the number of input variables are optimized by Artificial Bee Colony(ABC). In order to evaluate the performance of the proposed classifier, Yale B dataset which consists of images obtained under diverse state of illumination environment is applied. We show that the results of the proposed model have much more superb performance and robust characteristic than those reported in the previous studies.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.6 no.3
• /
• pp.216-225
• /
• 1994

By use of laboratory experiment data set for an elliptic shoal by Berkhoff et al. (1982), both accuracy and Performance tests of numerical results between PCGM (Preconditioned Conjugate Gradient Method) and PA(Parabolic Approximation) are compared. Although both results show good agreement with the experimental data the PA model gives better reproduction of the relatively high amplitudes in the section 4-5 downwave of the shoal, in comparison with the PCGM. The PA model has been proved to be a useful tool for predicting wave transformationsin large shallow water region, but it can be applied only to the case of negligible reflection. On the other hand, there is a need to improve the computational efficiency of the PCGM model which is a finite difference scheme directly derived from the mild slope equation and can handle reflection. By taking the results of th PA model as an input data of the PCGM, the CPU time can be reduced by about 40%.

• Computers and Concrete
• /
• v.2 no.5
• /
• pp.345-366
• /
• 2005

This work presents an assessment of the computational performance of a vector-parallel implementation of probabilistic model for concrete cracking in 3D. This paper shows the continuing efforts towards code optimization as reported in earlier works Paz, et al. (2002a,b and 2003). The probabilistic crack approach is based on the direct Monte Carlo method. Cracking is accounted by means of 3D interface elements. This approach considers that all nonlinearities are restricted to interface elements modeling cracks. The heterogeneity governs the overall cracking behavior and related size effects on concrete fracture. Computational kernels in the implementation are the inexact Newton iterative driver to solve the non-linear problem and a preconditioned conjugate gradient (PCG) driver to solve linearized equations, using an element by element (EBE) strategy to compute matrix-vector products. In particular the paper analyzes code behavior using OpenMP directives in parallel vector processors (PVP), such as the CRAY SV1 and CRAY T94. The impact of the memory architecture on code performance, and also some strategies devised to circumvent this issue are addressed by numerical experiment.

• Computers and Concrete
• /
• v.20 no.6
• /
• pp.627-634
• /
• 2017

Predicting the compressive strength of concrete is important to assess the load-carrying capacity of a structure. However, the use of blended cements to accrue the technical, economic and environmental benefits has increased the complexity of prediction models. Artificial Neural Networks (ANNs) have been used for predicting the compressive strength of ordinary Portland cement concrete, i.e., concrete produced without the addition of supplementary cementing materials. In this study, models to predict the compressive strength of blended cement concrete prepared with a natural pozzolan were developed using regression models and single- and 2-phase learning ANNs. Back-propagation (BP), Levenberg-Marquardt (LM) and Conjugate Gradient Descent (CGD) methods were used for training the ANNs. A 2-phase learning algorithm is proposed for the first time in this study for predictive modeling of the compressive strength of blended cement concrete. The output of these predictive models indicates that the use of a 2-phase learning algorithm will provide better results than the linear regression model or the traditional single-phase ANN models.

• Structural Engineering and Mechanics
• /
• v.83 no.3
• /
• pp.353-361
• /
• 2022

Linear Elastic Fracture Mechanics (LEFM) has been developed by applying stress analysis to determine the stress intensity factor (SIF, K). The finite element method (FEM) is widely used as a standard tool for evaluating the SIF for various crack configurations. The prediction accuracy can be achieved by applying an adaptive Delaunay triangulation combined with a FEM. The solution can be solved using either direct or iterative solvers. This work adopts the element-by-element preconditioned conjugate gradient (EBE-PCG) iterative solver into an adaptive FEM to solve the solution to heal problem size constraints that exist when direct solution techniques are applied. It can avoid the formation of a global stiffness matrix of a finite element model. Several numerical experiments reveal that the present method is simple, fast, and efficient compared to conventional sparse direct solvers. The optimum convergence criterion for two-dimensional LEFM analysis is studied. In this paper, four sample problems of a two-edge cracked plate, a center cracked plate, a single-edge cracked plate, and a compact tension specimen is used to evaluate the accuracy of the prediction of the SIF values. Finally, the efficiency of the present iterative solver is summarized by comparing the computational time for all cases.

• Wind and Structures
• /
• v.36 no.6
• /
• pp.423-434
• /
• 2023

Aerodynamic force coefficients are generally obtained from traditional wind tunnel tests or computational fluid dynamics (CFD). Unfortunately, the techniques mentioned above can sometimes be cumbersome because of the cost involved, such as the computational cost and the use of heavy equipment, to name only two examples. This study proposed to build a deep neural network model to predict the aerodynamic force coefficients based on data collected from CFD simulations to overcome these drawbacks. Therefore, a series of CFD simulations were conducted using different geometric parameters to obtain the aerodynamic force coefficients, validated with wind tunnel tests. The results obtained from CFD simulations were used to create a dataset to train a multilayer perceptron artificial neural network (ANN) model. The models were obtained using three optimization algorithms: scaled conjugate gradient (SCG), Bayesian regularization (BR), and Levenberg-Marquardt algorithms (LM). Furthermore, the performance of each neural network was verified using two performance metrics, including the mean square error and the R-squared coefficient of determination. Finally, the ANN model proved to be highly accurate in predicting the force coefficients of similar bridge sections, thus circumventing the computational burden associated with CFD simulation and the cost of traditional wind tunnel tests.

• Geomechanics and Engineering
• /
• v.33 no.6
• /
• pp.567-581
• /
• 2023

The prediction and achievement of a proper rock fragmentation size is the main challenge of blasting operations in surface mines. This is because an optimum size distribution can optimize the overall mine/plant economics. To this end, this study attempts to develop four improved artificial intelligence models to predict rock fragmentation through cascaded forward neural network (CFNN) and radial basis function neural network (RBFNN) models. In this regards, the CFNN was trained by the Levenberg-Marquardt algorithm (LMA) and Conjugate gradient backpropagation (CGP). Further, the RBFNN was optimized by the Dragonfly Algorithm (DA) and teaching-learning-based optimization (TLBO). For developing the models, the database required was collected from the Midouk copper mine, Iran. After modeling, the statistical functions were computed to check the accuracy of the models, and the root mean square errors (RMSEs) of CFNN-LMA, CFNN-CGP, RBFNN-DA, and RBFNN-TLBO were obtained as 1.0656, 1.9698, 2.2235, and 1.6216, respectively. Accordingly, CFNN-LMA, with the lowest RMSE, was determined as the model with the best prediction results among the four examined in this study.

• Biomedical Science Letters
• /
• v.4 no.1
• /
• pp.57-63
• /
• 1998

The objectives of this study were to produce polyclonal antibody to adipocyte plasma membrane (APM) proteins isolated from pig, and to investigate its tissue specificity. Plasma membrane proteins from adipocyte, brain, heart, kidney, liver and spleen were isolated using a self-forming Percoll gradient. Sheep (40kg) was immunized three times at three week interval with the purified APM proteins. Blood was taken from non-immunized sheep (NS) and from immunized sheep at 10 (AS-1), 12 (AS-2), and 14 (AS-3) days after the third immunization. Antisera titers and cross-reactivity against other tissues were determined by enzyme-linked immunosorbent assay (ELISA). Antisera reacted strongly to APM proteins showing detectable amounts of antibody at 1：81,000 dilution. And antisera showed much stronger reactivity to APM proteins than any other tissue plasma membrane proteins. Furthermore, tissue specificity of antisera against APM was reconfirmed by immunoblotting using anti-sheep immunoglobulin G-horseradish peroxidase conjugate as a secondary antibody Antisera to APM proteins showed adipocyte specificity compared with other tissues. In conclusion, polyclonal antibody against APM proteins isolated from pig was developed successfully in our laboratory, and these antisera showed tissue specificity with APM.

• Korean Journal of Veterinary Research
• /
• v.31 no.4
• /
• pp.491-500
• /
• 1991

This study was conducted to detect the serum antibodies in the experimentally toxoplasma infected dogs and street dogs by use the of an enzyme-linked immunosorbent assay (ELISA). And this test was performed on the polystylene microplate by coating with the tachyzoites soluble antigen of T gondii (RH strain), incubated with sera diluted then, added with HPO-conjugated rabbit anti-dog IgG and o-phenylenediamine used as a substrate. Tachyzoites of T gondii harvested from mouse peritoneal cavity were purified by 30, 40 and 50% Percoll density gradient centrifugation and used as the source of antigen. The results obtained were summarized as follows; 1. The highest ratio of positive to negative (P/N ratio) was obtained at the level of $l{\mu}g/ml$ protein concentration of antigen with the 1/4000 dilution of the conjugate measured by checker-board titration. It was regarded as the optimum concentration of the antigen and conjugate. 2. Cut-off value in this IgG ELISA was 0.375 that was determined by mean absorbance (at 492nm) of IFA negative serum added with the dauble value of the standard deviation $(mean{\pm}2S.D.)$. 3. Serum ELISA IgG antibodies to T gondii in the exyerimentally infected dogs were detected firstly at the Week 3 after inoculation and the highest titer was recognized at the Week 4, 5 and 6 after inoculation. 4. Stability of the antigen absorbed in the microplates that were preserved at $4^{\circ}C$ and $-25^{\circ}C$ separately were prolonged up to 3 weeks and 10 weeks at $4^{\circ}C$ and $-25^{\circ}C$, respectively. However the reproducibility was not reliable after the preservation of 4 weeks and longer. 5. Positive rate of the specific antibodies in 312 test sera was 28.5% and there was no significant differences between the male (27.8%) and female (29.5%), respectively. 6. The IgG ELISA was proved to be a specific procedure for the detection of antibodies to canine toxoplasma infection and also evaluated as a screening test for the large scale of test samples in laboratory.

• Journal of Animal Science and Technology
• /
• v.46 no.2
• /
• pp.137-144
• /
• 2004

Algorithms for estimating breeding values on several categorical data by using latent variables with threshold conception were developed and showed. Thresholds on each categorical trait were estimated by Newton’s method via gradients and Hessian matrix. This algorithm was developed by way of expansion of bivariate analysis provided by Quaas(2001). Breeding values on latent variables of categorical traits and observations on linear traits were estimated by preconditioned conjugate gradient(PCG) method, which was known having a property of fast convergence. Example was shown by simulated data with two linear traits and a categorical trait with four categories(CE=calving ease) and a dichotomous trait(SB=Still Birth) in threshold animal mixed model(TAMM). Breeding value estimates in TAMM were compared to those in linear animal mixed model (LAMM). As results, correlation estimates of breeding values to parameters were 0.91${\sim}$0.92 on CE and 0.87${\sim}$0.89 on SB in TAMM and 0.72~0.84 on CE and 0.59~0.70 on SB in LAMM. As conclusion, PCG method for estimating breeding values on several categorical traits with linear traits were feasible in TAMM.