In this paper discussed the problem of constructing and training a fuzzy neural network based on fuzzy logical rules. Based on the constructed model, the developed algorithm, the objects classified with indistinct initial information. Under these conditions, traditional methods of mathematical statistics or simulation modeling do not allow building adequate models for solving data mining problems. We consider the solution of problems based on the use of intelligent modeling technologies using methods of neural networks and fuzzy logic in order to automate the construction of fuzzy rules based on methods and algorithms of machine learning.
the paper deals with the case when particles are evenly distributed in ravines, using the equilibrium coefficient to update the particle velocity. the method of optimizing a swarm of particles in the case of a "ravine" is one of the most effective approaches for multi-extremal optimization. however, in the existing methods of optimizing the swarm of particles with the "ravine" methods, the number of particles around the ravine is very different from each other, which makes it difficult to find high-quality algorithms in all the ravines. thus, the computational resources are distributed in the ravines in a more balanced way.
In this paper, qualitative properties of the reaction-diffusion equation with double nonlinearity investigated. Research carried out based on a self-similar analysis of solutions. Population models of two competing populations with double nonlinear diffusion, which described by a nonlinear system of competing individuals, are proposed.
In this paper, considered a parabolic system of two quasi linear reaction-diffusion equations with the source and absorption task and the properties of self-similar solutions of a system of quasi linear reaction-diffusion equationsfor the source and absorption task. Self-similar system of equations is constructed by the method of nonlinear splitting. Estimates of the solutions and the free boundary that arises in this case are found, which makes it possible to choose suitable initial approximations for each value of the numerical parameters. INTRODUCTIONStudy of nonlinear mathematical models of various physical, biological, chemical and other phenomena and processes is one of the important directions of mathematical modeling. As examples, we note such physical theories as non-linear quantum mechanics, nonlinear electrodynamics and optics, non-linear theory of plasmas, nonlinear acoustics, nonlinear conduction, nonlinear diffusion, and other theories based on mathematical models which are nonlinear differential equations in partial derivatives. The study of the linear mathematical models of physical processes are easy to study, since the underlying linear differential equations developed general methods for their solution [1,2]. In applied tasks, the actual physical processes are nonlinear and for their adequate description, one should use nonlinear mathematical models.Nonlinear models of mathematical physics, describing phenomena and processes in a wider range of physical parameter changes and have a better capacity of information about these phenomena and processes.Linear models are typically special cases of nonlinear models. They can give only an approximate picture of the phenomenon under study without identifying the observed effects. Studies show that the nonlinearities change not only the quantitative characteristics of the processes, but the qualitative picture of their behavior. Interestingly, from the point of view of applications to study the following classes of nonlinear differential equations in which the unknown function and the derivative of this function consists of exponential way. Then, with the comparison theorems of solutions of this class can be extended. These types of nonlinearities are often encountered in problems of the theory of filtration, diffusion, thermal conductivity, magnetohydrodynamics, biological populations, the oil industry, etc. [3,4,5,6].These models more accurately describe the physics of the process and, therefore, their research shows that there are new effects related to the nonlinearity of the studied process. So was found the effects of the finite speed of propagation of perturbations [1], localization of solutions and different modes of processes. The first effect of finite speed of propagation of perturbations was obtained and applied to the problem of nonlinear thermal conductivity, in the work of Zeldovich and A. S. Kompaneetz[1], for the problem of nonlinear filtering in the work of G. I. Barenblatt [2,3], which are independently from each other, and got this ef...
Cultural algorithm demonstrates incompetence in solving problems of multi-extremal optimization. This is due to the large dimensionality of the data and causes premature convergence. To solve this problem, a fuzzy cultural algorithm is proposed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.