Numerisk analyse

Multilayer neural networks based on multi-valued neurons (MLMVNs) have been proposed to combine the advantages of complex-valued neural networks with a plain derivative-free learning algorithm. In addition, multi-valued neurons (MVNs) offer a multi-valued threshold logic resulting in the ability to replace multiple conventional output neurons in classification tasks. Therefore, several classes can be assigned to one output neuron. This book introduces a novel approach to assign multiple classes to numerous MVNs in the output layer. It was found that classes that possess similarities should be allocated to the same neuron and arranged adjacent to each other on the unit circle. Since MLMVNs require input data located on the unit circle, two employed transformations are reevaluated. The min-max scaler utilizing the exponential function, and the 2D discrete Fourier transform restricting to the phase information for image recognition. The evaluation was performed on the Sensorless Drive Diagnosis dataset and the Fashion MNIST dataset.

This CIME Series book provides mathematical and simulation tools to help resolve environmental hazard and security-related issues.The contributions reflect five major topics identified by the SIES (Strategic Initiatives for the Environment and Security) as having significant societal impact: optimal control in waste management, in particular the degradation of organic waste by an aerobic biomass, by means of a mathematical model; recent developments in the mathematical analysis of subwave resonators; conservation laws in continuum mechanics, including an elaboration on the notion of weak solutions and issues related to entropy criteria; the applications of variational methods to 1-dimensional boundary value problems, in particular to light ray-tracing in ionospheric physics; and the mathematical modelling of potential electromagnetic co-seismic events associated to large earthquakes.This material will provide a sound foundation for those who intend to approach similar problems from a multidisciplinary perspective.

Mathematical models cannot be solved using the traditional analytical methods for dynamic equations on time scales. These models must be dealt with using computational methods. This textbook introduces numerical methods for initial value problems for dynamic equations on time scales. Hands-on examples utilizing MATLAB and practical problems illustrate a wide variety of solution techniques.

Functional analysis is an important branch of mathematical analysis which deals with the transformations of functions and their algebraic and topological properties. Motivated by their large applicability to real life problems, applications of functional analysis have been the aim of an intensive study effort in the last decades, yielding significant progress in the theory of functions and functional spaces, differential and difference equations and boundary value problems, differential and integral operators and spectral theory, and mathematical methods in physical and engineering sciences. The present volume is devoted to these investigations. The publication of this collection of papers is based on the materials of the mini-symposium "Functional Analysis in Interdisciplinary Applications" organized in the framework of the Fourth International Conference on Analysis and Applied Mathematics (ICAAM 2018, September 6¿9, 2018). Presenting a widerange of topics and results, this book will appeal to anyone working in the subject area, including researchers and students interested to learn more about different aspects and applications of functional analysis. Many articles are written by experts from around the world, strengthening international integration in the fields covered. The contributions to the volume, all peer reviewed, contain numerous new results.This volume contains four different chapters. The first chapter contains the contributed papers focusing on various aspects of the theory of functions and functional spaces. The second chapter is devoted to the research on difference and differential equations and boundary value problems. The third chapter contains the results of studies on differential and integral operators and on the spectral theory. The fourth chapter is focused on the simulation of problems arising in real-world applications of applied sciences.

Revolutionary advances in hardware and software technology have made computer aided design and analysis a standard tool in engineering practice. While this puts a lot of power at the hands of the end user, in order to use them wisely and interpret the results correctly, the users are expected to have a sound knowledge of the relationship between the physical world and the mathematical model and that between the mathematical model and the numerical approximation. The text examines the link between the mathematical and the numerical model in a succinct manner. Numerical algorithms relevant to engineers are developed and applied to solve industrial problems.The text is intended for senior level undergraduate and first year graduate students without rigorous training in numerical analysis. The following topics are covered both in theory and practical examples.-Roots of non-linear equations.-Systems of linear algebraic equations.-Interpolation and curve fitting.-Numerical differentiation and integration.-Ordinary differential equation - initial value problems.-Ordinary differential equations - boundary value problems.-Partial differential equations.

En este libro estudiaremos una amplia variedad de algoritmos numéricos para resolver diferentes problemas de matemática continua, enfatizando en aquellos que surgen con frecuencia en las aplicaciones. En especial consideraremos la solución de sistemas lineales, por métodos directos o iterativos, la solución de ecuaciones no lineales en una o en muchas variables, el problema de interpolar un conjunto dado de datos, el ajuste de datos mediante la técnica de cuadrados mínimos lineales y no lineales, estimación de autovalores y autovectores, la solución de ecuaciones diferenciales entre otros. Para todos estos problemas y para cada algoritmo propuesto discutiremos en el capítulo asociado sus propiedades teóricas y prácticas y en especial su estabilidad, y como se ven afectados por la presencia de errores, y por la representación de los números en el computador. Por cada método estudiado ilustraremos su comportamiento con ejemplos básicos. De igual forma, cada capítulo cierra con ejercícios adicionales y aplicaciones reales relativas a los métodos mencionados. Este libro Está dirigido a estudiantes avanzados de pregrado, pero se ajusta bien a cursos iniciales de postgrado. Más aún, al cubrir una amplia variedad de temas, también puede ser usado como referencia por investigadores de diferentes disciplinas de las ciencias aplicadas que tengan la necesidad de resolver problemas prácticos que requieran el uso de métodos numéricos.

The position taken in this collection of pedagogically written essays is that conjugate gradient algorithms and finite element methods complement each other extremely well. Via their combinations practitioners have been able to solve complicated, direct and inverse, multidemensional problems modeled by ordinary or partial differential equations and inequalities, not necessarily linear, optimal control and optimal design being part of these problems.The aim of this book is to present both methods in the context of complicated problems modeled by linear and nonlinear partial differential equations, to provide an in-depth discussion on their implementation aspects. The authors show that conjugate gradient methods and finite element methods apply to the solution of real-life problems. They address graduate students as well as experts in scientific computing.

Kronecker products are used to define the underlying Markov chain (MC) in various modeling formalisms, including compositional Markovian models, hierarchical Markovian models, and stochastic process algebras. The motivation behind using a Kronecker structured representation rather than a flat one is to alleviate the storage requirements associated with the MC. With this approach, systems that are an order of magnitude larger can be analyzed on the same platform. The developments in the solution of such MCs are reviewed from an algebraic point of view and possible areas for further research are indicated with an emphasis on preprocessing using reordering, grouping, and lumping and numerical analysis using block iterative, preconditioned projection, multilevel, decompositional, and matrix analytic methods. Case studies from closed queueing networks and stochastic chemical kinetics are provided to motivate decompositional and matrix analytic methods, respectively.