• Journal of the Korean Mathematical Society
• /
• v.61 no.3
• /
• pp.427-444
• /
• 2024

The matrix completion problem is to predict missing entries of a data matrix using the low-rank approximation of the observed entries. Typical approaches to matrix completion problem often rely on thresholding the singular values of the data matrix. However, these approaches have some limitations. In particular, a discontinuity is present near the thresholding value, and the thresholding value must be manually selected. To overcome these difficulties, we propose a shrinkage and thresholding function that smoothly thresholds the singular values to obtain more accurate and robust estimation of the data matrix. Furthermore, the proposed function is differentiable so that the thresholding values can be adaptively calculated during the iterations using Stein unbiased risk estimate. The experimental results demonstrate that the proposed algorithm yields a more accurate estimation with a faster execution than other matrix completion algorithms in image inpainting problems.

• International Journal of Aeronautical and Space Sciences
• /
• v.18 no.4
• /
• pp.662-674
• /
• 2017

This paper describes the multidisciplinary design optimization (MDO) process of a tailless unmanned combat aerial vehicle (UCAV) using global variable fidelity aerodynamic analysis. The developed tailless UAV design framework combines multiple disciplines that are based on low-fidelity and empirical analysis methods. An automated high-fidelity aerodynamic analysis is efficiently integrated into the MDO framework. Global variable fidelity modeling algorithm manages the use of the high-fidelity analysis to enhance the overall accuracy of the MDO by providing the initial sampling of the design space with iterative refinement of the approximation model in the neighborhood of the optimum solution. A design formulation was established considering a specific aerodynamic, stability and control design features of a tailless aircraft configuration with a UCAV specific mission profile. Design optimization problems with low-fidelity and variable fidelity analyses were successfully solved. The objective function improvement is 14.5% and 15.9% with low and variable fidelity optimization respectively. Results also indicate that low-fidelity analysis overestimates the value of lift-to-drag ratio by 3-5%, while the variable fidelity results are equal to the high-fidelity analysis results by algorithm definition.

• Structural Engineering and Mechanics
• /
• v.69 no.6
• /
• pp.615-626
• /
• 2019

The 3-noded metric Timoshenko beam element with an offset of the internal node from the element centre is used here to demonstrate the best-fit paradigm using function space formulation under locking and mesh distortion. The best-fit paradigm follows from the projection theorem describing finite element analysis which shows that the stresses computed by the displacement finite element procedure are the best approximation of the true stresses at an element level as well as global level. In this paper, closed form best-fit solutions are arrived for the 3-noded Timoshenko beam element through function space formulation by combining field consistency requirements and distortion effects for the element modelled in metric Cartesian coordinates. It is demonstrated through projection theorems how lock-free best-fit solutions are arrived even under mesh distortion by using a consistent definition for the shear strain field. It is shown how the field consistency enforced finite element solution differ from the best-fit solution by an extraneous response resulting from an additional spurious force vector. However, it can be observed that when the extraneous forces vanish fortuitously, the field consistent solution coincides with the best-fit strain solution.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.25 no.3
• /
• pp.537-547
• /
• 2001

The paper proposes the fuzzy logic global approximate optimization strategies in optimal sizing of automotive A-pillar trim with rib structures for occupant head protection. Two different strategies referred to as evolutionary fuzzy modeling (EFM) and neuro-fuzzy modeling (NFM) are implemented in the context of global approximate optimization. EFM and NFM are based on soft computing paradigms utilizing fuzzy systems, neural networks and evolutionary computing techniques. Such approximation methods may have their promising characteristics in a case where the inherent nonlinearity in analysis model should be accommodated over the entire design space and the training data is not sufficiently provided. The objective of structural design is to determine the dimensions of rib in A-pillar, minimizing the equivalent head injury criterion HIC(d). The paper describes the head-form modeling and head impact simulation using LS-DYNA3D, and the approximation procedures including fuzzy rule generation, membership function selection and inference process for EFM and NFM, and subsequently presents their generalization capabilities in terms of number of fuzzy rules and training data.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.685-686
• /
• 2007

This paper presents an application of power system fault diagnostic algorithm for the PEFP Proton Accelerator Research Center using neural network. Proposed fault diagnostic system is constructed by the radial basis function (RBF) neural network because it has the capabilities of the pattern classification and function approximation of any nonlinear function. Proposed system identifies faulted section in the power system based on information about the operation of protection devices such as relays and circuit breakers. In this paper, parameters of the RBF neural networks are tuned by the GA-TS algorithm, which has the global optimal solution searching capabilities. To show the validity of the proposed method, proposed algorithm has been tested with a practical power system in Proton Accelerator Research Center of PEFP.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.654-656
• /
• 1998

This paper presents an optimal identification method of nonlinear and complex system that is based on fuzzy-neural network(FNN). The FNN used simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. And we use a HCM Algorithm to find initial parameters of membership function. And then to obtain optimal parameters, we use the genetic algorithm. Genetic algorithm is a random search algorithm which can find the global optimum without converging to local optimum. The parameters such as membership functions, learning rates and momentum coefficients are easily adjusted using the genetic algorithms. Also, the performance index with weighted value is introduced to achieve a meaningful balance between approximation and generalization abilities of the model. To evaluate the performance of the FNN, we use the time series data for 9as furnace and the sewage treatment process.

• Korean Chemical Engineering Research
• /
• v.57 no.5
• /
• pp.723-729
• /
• 2019

A theoretical analysis was conducted of convective instability driven by buoyancy forces under transient temperature fields in an annular porous medium bounded by coaxial vertical cylinders. Darcy's law and Boussinesq approximation are used to explain the characteristics of fluid motion and linear stability theory is employed to predict the onset of buoyancy-driven motion. The linear stability equations are derived in a global domain, and then cast into in a self-similar domain. Using a spectral expansion method, the stability equations are reformed as a system of ordinary differential equations and solved analytically and numerically. The critical Darcy-Rayleigh number is founded as a function of the radius ratio. Also, the onset time and corresponding wavelength are obtained for the various cases. The critical time becomes smaller with increasing the Darcy-Rayleigh number and follows the asymptotic relation derived in the infinite horizontal porous layer.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.2
• /
• pp.194-202
• /
• 2003

In this paper, we introduce a category of Multi-FNN (Fuzzy-Neural Networks) models, analyze the underlying architectures and propose a comprehensive identification framework. The proposed Multi-FNNs dwell on a concept of fuzzy rule-based FNNs based on HCM clustering and evolutionary fuzzy granulation, and exploit linear inference being treated as a generic inference mechanism. By this nature, this FNN model is geared toward capturing relationships between information granules known as fuzzy sets. The form of the information granules themselves (in particular their distribution and a type of membership function) becomes an important design feature of the FNN model contributing to its structural as well as parametric optimization. The identification environment uses clustering techniques (Hard C - Means, HCM) and exploits genetic optimization as a vehicle of global optimization. The global optimization is augmented by more refined gradient-based learning mechanisms such as standard back-propagation. The HCM algorithm, whose role is to carry out preprocessing of the process data for system modeling, is utilized to determine the structure of Multi-FNNs. The detailed parameters of the Multi-FNN (such as apexes of membership functions, learning rates and momentum coefficients) are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization (predictive) abilities of the model. To evaluate the performance of the proposed model, two numeric data sets are experimented with. One is the numerical data coming from a description of a certain nonlinear function and the other is NOx emission process data from a gas turbine power plant.

• Ocean Systems Engineering
• /
• v.5 no.3
• /
• pp.139-160
• /
• 2015

The global performance of the 5MW OC4 semisubmersible floating wind turbine in random waves was numerically simulated by using the turbine-floater-mooring fully coupled and time-domain dynamic analysis program FAST-CHARM3D. There have been many papers regarding floating offshore wind turbines but the effects of second-order wave-body interactions on their global performance have rarely been studied. The second-order wave forces are actually small compared to the first-order wave forces, but its effect cannot be ignored when the natural frequencies of a floating system are outside the wave-frequency range. In the case of semi-submersible platform, second-order difference-frequency wave-diffraction forces and moments become important since surge/sway and pitch/roll natural frequencies are lower than those of typical incident waves. The computational effort related to the full second-order diffraction calculation is typically very heavy, so in many cases, the simplified approach called Newman's approximation or first-order-wave-force-only are used. However, it needs to be justified against more complete solutions with full QTF (quadratic transfer function), which is a main subject of the present study. The numerically simulated results for the 5MW OC4 semisubmersible floating wind turbine by FAST-CHARM3D are also extensively compared with the DeepCWind model test results by Technip/NREL/UMaine. The predicted motions and mooring tensions for two white-noise input-wave spectra agree well against the measure values. In this paper, the numerical static-offset and free-decay tests are also conducted to verify the system stiffness, damping, and natural frequencies against the experimental results. They also agree well to verify that the dynamic system modeling is correct to the details. The performance of the simplified approaches instead of using the full QTF are also tested.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.21 no.6
• /
• pp.954-960
• /
• 2012

Nowadays, rapidly changing and unstable global economic environments request a lot of roles to engineers. In this situation, product should be designed to make more profit by cost down and to satisfy distinguished performance comparing to other competitive ones. In this research, the optimization design of the industrial robot casting will be done. The weight and deflection have to be reduced as objective functions and stress has to be constrained under some constant value. To reduce time cost, CCD (Central Composite Design) will be used to make experimental design. And RSM (Response Surface Methodology) will be taken to make regression model for objective functions and constraint function. Finally, optimization will be done with Genetic Algorithm. In this problem, the objective functions are multiple, so NSGA-II which is brilliant and efficient for such a problem will be used. For the solution quality check, the diversity between Pareto solutions will be also checked.