• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.438-441
• /
• 2004

In this study, we introduce a noble neurogenetic approach to the design of fuzzy controller. The design procedure dwells on the use of Computational Intelligence (CI), namely genetic algorithms and Fuzzy-Set based Neural Networks (FSNN). The crux of the design methodology is based on the selection and determination of optimal values of the scaling factors of the fuzzy controllers, which are essential to the entire optimization process. First, the tuning of the scaling factors of the fuzzy controller is carried out by using GAs, and then the development of a nonlinear mapping for the scaling factors is realized by using GA based FSNN. The developed approach is applied to a nonlinear system such as an inverted pendulum where we show the results of comprehensive numerical studies and carry out a detailed comparative analysis.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.3
• /
• pp.625-636
• /
• 2013

In this paper we study four kernel machines for estimating expected shortfall, which are constructed through combinations of support vector quantile regression (SVQR), restricted SVQR (RSVQR), least squares support vector machine (LS-SVM) and support vector expectile regression (SVER). These kernel machines have obvious advantages such that they achieve nonlinear model but they do not require the explicit form of nonlinear mapping function. Moreover they need no assumption about the underlying probability distribution of errors. Through numerical studies on two artificial an two real data sets we show their effectiveness on the estimation performance at various confidence levels.

• Journal of Electrical Engineering and Technology
• /
• v.5 no.4
• /
• pp.653-665
• /
• 2010

In this study, we introduce a neurofuzzy approach to the design of fuzzy controllers. The development process exploits key technologies of Computational Intelligence (CI), namely, genetic algorithms (GA) and neurofuzzy networks. The crux of the design methodology deals with the selection and determination of optimal values of the scaling factors of fuzzy controllers, which are essential to the entire optimization process. First, the tuning of the scaling factors of the fuzzy controller is carried out. Next, we form a nonlinear mapping for the scaling factors, which are realized by GA-based neurofuzzy networks by using a fuzzy set or fuzzy relation. The proposed approach is applied to control nonlinear systems like the inverted pendulum. Results of comprehensive numerical studies are presented through a detailed comparative analysis.

• East Asian mathematical journal
• /
• v.26 no.5
• /
• pp.703-714
• /
• 2010

A new class of nonlinear set-valued mixed variational inclusions involving (A, ${\eta}$)-accretive mappings in Banach spaces is introduced and studied, which includes many kind of variational inclusion (inequality) and complementarity problems as special cases. By using the resolvent operator associated with (A, ${\eta}$)-accretive operator due to Lan-Cho-Verma, the existence of solution for this kind of variational inclusion is proved, and a new hybrid proximal point algorithm is established and suggested, the convergence and stability theorems of iterative sequences generated by new iterative algorithms are also given in q-uniformly smooth Banach spaces.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.3
• /
• pp.362-373
• /
• 2004

In this study, we introduce a noble neurogenetic approach to the design of the fuzzy controller. The design procedure dwells on the use of Computational Intelligence (CI), namely genetic algorithms and neurofuzzy networks (NFN). The crux of the design methodology is based on the selection and determination of optimal values of the scaling factors of the fuzzy controllers, which are essential to the entire optimization process. First, tuning of the scaling factors of the fuzzy controller is carried out, and then the development of a nonlinear mapping for the scaling factors is realized by using GA based NFN. The developed approach is applied to an inverted pendulum nonlinear system where we show the results of comprehensive numerical studies and carry out a detailed comparative analysis.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.4
• /
• pp.543-552
• /
• 1994

This paper presents a decentralized quadratic regulation architecture with feedforward neural networks for the control problem of complex systems. In this method, the decentralized technique was used to treat several simple subsystems instead of a full complex system in order to reduce training time of neural networks, and the neural networks' nonlinear mapping ability is exploited to handle the nonlinear interaction variables between subsystems. The decentralized regulating architecture is composed of local neuro-controllers, local neuro-identifiers and an overall interaction neuro-identifier. With the interaction neuro-identifier that catches interaction characteristics, a local neuro-identifier is trained to simulate a subsystem dynamics. A local neuro-controller is trained to learn how to control the subsystem by using generalized Backprogation Through Time(BTT) algorithm. The proposed neural network based decentralized regulating scheme is applied in the power System Stabilization(PSS) control problem for an imterconnected power system, and compared with that by a conventional centralized LQ regulator for the power system.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2003.10a
• /
• pp.429-436
• /
• 2003

A new linkage framework between elastic shell element with finite rotation and computar-aided geometric design (CAGD) (or surface is developed in the present study. The framework of shell finite element is based on the generalized curved two-parametric coordinate system. To represent free-form surface, cubic B-spline tensor-product functions are used. Thus the present finite element can be directly linked into the geometric modeling produced by surface generation tool in CAD software. The efficiency and accuracy of the Previously developed linear elements hold for the nonlinear element with finite rotations. To handle the finite rotation behavior of shells, exponential mapping in the SO(3) group is employed to allow the large incremental step size. The integrated frameworks of shell geometric design and nonlinear computational analysis can serve as an efficient tool in shape and topological design of surfaces with large deformations.

• Nonlinear Functional Analysis and Applications
• /
• v.26 no.2
• /
• pp.411-432
• /
• 2021

In this paper, an inertial Halpern-type forward backward iterative algorithm for approximating solution of a monotone inclusion problem whose solution is also a fixed point of some nonlinear mapping is introduced and studied. Strong convergence theorem is established in a real Hilbert space. Furthermore, our theorem is applied to variational inequality problems, convex minimization problems and image restoration problems. Finally, numerical illustrations are presented to support the main theorem and its applications.

• Honam Mathematical Journal
• /
• v.42 no.4
• /
• pp.667-680
• /
• 2020

Our paper deals with the following neutral nonlinear Levin-Nohel integro-differential with variable delay $${\frac{d}{dt}x(t)}+{\normalsize\displaystyle\smashmargin{2}{\int\nolimits_{t-r(t)}}^t}a(t,s)x(s)ds+{\frac{d}{dt}}g(t,x(t-{\tau}(t)))=0.$$ By using Krasnoselskii's fixed point theorem we obtain the existence of periodic and positive periodic solutions and by contraction mapping principle we obtain the existence of a unique periodic solution. An example is given to illustrate this work.

• Nonlinear Functional Analysis and Applications
• /
• v.26 no.5
• /
• pp.995-1009
• /
• 2021

Many iterative algorithms like that Picard, Mann, Ishikawa and S-iteration are very useful to elucidate the fixed point problems of a nonlinear operators in various topological spaces. The recent trend for elucidate the fixed point via inertial iterative algorithm, in which next iterative depends on more than one previous terms. The purpose of the paper is to establish convergence theorems of new inertial Picard normal S-iteration algorithm for nonexpansive mapping in Hilbert spaces. The comparison of convergence of InerNSP and InerPNSP is done with InerSP (introduced by Phon-on et al. [25]) and MSP (introduced by Suparatulatorn et al. [27]) via numerical example.