The Multiplicative Schwarz Preconditioned Inexact Newton (MSPIN) algorithm is presented as a complement to Additive Schwarz Preconditioned Inexact Newton (ASPIN). At an algebraic level, ASPIN and MSPIN are variants of the same strategy to improve the convergence of systems with unbalanced nonlinearities; however, they have natural complementarity in practice. MSPIN is naturally based on partitioning of degrees of freedom in a nonlinear PDE system by field type rather than by subdomain, where a modest factor of concurrency can be sacrificed for physically motivated convergence robustness. ASPIN, originally introduced for decompositions into subdomains, is natural for high concurrency and reduction of global synchronization. The ASPIN framework, as an option for the outermost solver, successfully handles strong nonlinearities in computational fluid dynamics, but is barely explored for the highly nonlinear models of complex multiphase flow with capillarity, heterogeneity, and complex geometry. In this dissertation, the fully implicit ASPIN method is demonstrated for a finite volume discretization based on incompressible two-phase reservoir simulators in the presence of capillary forces and gravity. Numerical experiments show that the number of global nonlinear iterations is not only scalable with respect to the number of processors, but also significantly reduced compared with the standard inexact Newton method with a backtracking technique. Moreover, the ASPIN method, in contrast with the IMPES method, saves overall execution time because of the savings in timestep size. We consider the additive and multiplicative types of inexact Newton algorithms in the field-split context, and we augment the classical convergence theory of ASPIN for the multiplicative case. Moreover, we provide the convergence analysis of the MSPIN algorithm. Under suitable assumptions, it is shown that MSPIN is locally convergent, and desired superlinear or even quadratic convergence can be obtained when the forcing terms are picked suitably. Numerical experiments show that MSPIN can be significantly more robust than Newton methods based on global linearizations, and that MSPIN can be more robust than ASPIN, and maintain fast convergence even for challenging problems, such as high-Reynolds number Navier-Stokes equations.

Submitted by Cláudia Bueno (claudiamoura18@gmail.com) on 2015-10-29T19:04:41Z No. of bitstreams: 2 Tese - Tiberio Bittencourt de Oliveira Martins.pdf: 1155588 bytes, checksum: add1eac74c4397efc29678341b834448 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2015-11-03T14:25:04Z (GMT) No. of bitstreams: 2 Tese - Tiberio Bittencourt de Oliveira Martins.pdf: 1155588 bytes, checksum: add1eac74c4397efc29678341b834448 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2015-11-03T14:25:04Z (GMT). No. of bitstreams: 2 Tese - Tiberio Bittencourt de Oliveira Martins.pdf: 1155588 bytes, checksum: add1eac74c4397efc29678341b834448 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2015-06-26 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Apresentamos, nesta tese, uma an álise da convergência do m étodo de Newton inexato com tolerância de erro residual relativa e uma an alise semi-local de m etodos de Newton robustos exato e inexato, objetivando encontrar uma singularidade de um campo de vetores diferenci avel de nido em uma variedade Riemanniana completa, baseados no princ pio majorante a m invariante. Sob hip oteses locais e considerando uma fun ção majorante geral, a Q-convergância linear do m etodo de Newton inexato com uma tolerância de erro residual relativa xa e provada. Na ausência dos erros, a an alise apresentada reobtem o teorema local cl assico sobre o m etodo de Newton no contexto Riemanniano. Na an alise semi-local dos m etodos exato e inexato de Newton apresentada, a cl assica condi ção de Lipschitz tamb em e relaxada usando uma fun ção majorante geral, permitindo estabelecer existência e unicidade local da solu ção, uni cando previamente resultados pertencentes ao m etodo de Newton. A an alise enfatiza a robustez, a saber, e dada uma bola prescrita em torno do ponto inicial que satifaz as hip oteses de Kantorovich, garantindo a convergência do m etodo para qualquer ponto inicial nesta bola. Al em disso, limitantes que dependem da função majorante para a taxa de convergência Q-quadr atica do m étodo exato e para a taxa de convergência Q-linear para o m etodo inexato são obtidos. / A local convergence analysis with relative residual error tolerance of inexact Newton method and a semi-local analysis of a robust exact and inexact Newton methods are presented in this thesis, objecting to nd a singularity of a di erentiable vector eld de ned on a complete Riemannian manifold, based on a ne invariant majorant principle. Considering local assumptions and a general majorant function, the Q-linear convergence of inexact Newton method with a xed relative residual error tolerance is proved. In the absence of errors, the analysis presented retrieves the classical local theorem on Newton's method in Riemannian context. In the semi-local analysis of exact and inexact Newton methods presented, the classical Lipschitz condition is also relaxed by using a general majorant function, allowing to establish the existence and also local uniqueness of the solution, unifying previous results pertaining Newton's method. The analysis emphasizes robustness, being more speci c, is given a prescribed ball around the point satisfying Kantorovich's assumptions, ensuring convergence of the method for any starting point in this ball. Furthermore, the bounds depending on the majorant function for Q-quadratic convergence rate of the exact method and Q-linear convergence rate of the inexact method are obtained.

Submitted by Cássia Santos (cassia.bcufg@gmail.com) on 2017-03-16T12:01:01Z No. of bitstreams: 2 Dissertação - Stefan Alberto Gómez Guevara - 2017.pdf: 2201042 bytes, checksum: bd12be92bd41bae24c13758a1fc1a73d (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-03-20T13:11:14Z (GMT) No. of bitstreams: 2 Dissertação - Stefan Alberto Gómez Guevara - 2017.pdf: 2201042 bytes, checksum: bd12be92bd41bae24c13758a1fc1a73d (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-03-20T13:11:14Z (GMT). No. of bitstreams: 2 Dissertação - Stefan Alberto Gómez Guevara - 2017.pdf: 2201042 bytes, checksum: bd12be92bd41bae24c13758a1fc1a73d (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-03-08 / In this work we consider the problem of finding a singularity of a field of differentiable vectors X on a Riemannian manifold. We present a local analysis of the convergence of Newton's method to find a singularity of field X on an increasing condition. The analysis shows a relationship between the major function and the vector field X. We also present a semi-local Kantorovich type analysis in the Riemannian context under a major condition. The two results allow to unify some previously unrelated results. / Neste trabalho consideramos o problema de encontrar uma singularidade de um campo de vetores diferenciável X sobre uma variedade Riemanniana. Apresentamos uma análise local da convergência do método de Newton para encontrar uma singularidade do Campo X sobre uma condição majorante. A análise mostra uma relação entre a função majorante e o campo de vetores X. Também apresentamos uma análise semi-local do tipo Kantorovich no contexto Riemanniana sob uma condição majorante. Os dois resultados permitem unificar alguns resultados não previamente.

Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2015-03-05T14:28:50Z No. of bitstreams: 2 Dissertação - Ademir Alves Aguiar - 2014.pdf: 1975016 bytes, checksum: 31320b5840b8b149afedc97d0e02b49b (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2015-03-06T10:38:03Z (GMT) No. of bitstreams: 2 Dissertação - Ademir Alves Aguiar - 2014.pdf: 1975016 bytes, checksum: 31320b5840b8b149afedc97d0e02b49b (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2015-03-06T10:38:03Z (GMT). No. of bitstreams: 2 Dissertação - Ademir Alves Aguiar - 2014.pdf: 1975016 bytes, checksum: 31320b5840b8b149afedc97d0e02b49b (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2014-12-18 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / In this dissertation we present a semi-local convergence analysis for the Gauss-Newton method to solve a special class of systems of non-linear equations, under the hypothesis that the derivative of the non-linear operator satisfies a majorant condition. The proofs and conditions of convergence presented in this work are simplified by using a simple majorant condition. Another tool of demonstration that simplifies our study is to identify regions where the iteration of Gauss-Newton is “well-defined”. Moreover, special cases of the general theory are presented as applications. / Nesta dissertação apresentamos uma análise de convergência semi-local do método de Gauss-Newton para resolver uma classe especial de sistemas de equações não-lineares, sob a hipótese que a derivada do operador não-linear satisfaz uma condição majorante. As demonstrações e condições de convergência apresentadas neste trabalho são simplificadas pelo uso de uma simples condição majorante. Outra ferramenta de demonstração que simplifica o nosso estudo é a identificação de regiões onde a iteração de Gauss-Newton está “bem-definida”. Além disso, casos especiais da teoria geral são apresentados como aplicações.

[ES] La resolución de ecuaciones y sistemas no lineales es un tema de gran interés teórico-práctico, pues muchos modelos matemáticos de la ciencia o de la industria se expresan mediante sistemas no lineales o ecuaciones diferenciales o integrales que, mediante técnicas de discretización, dan lugar a dichos sistemas. Dado que generalmente es difícil, si no imposible, resolver analíticamente las ecuaciones no lineales, la herramienta más extendida son los métodos iterativos, que tratan de obtener aproximaciones cada vez más precisas de las soluciones partiendo de determinadas estimaciones iniciales. Existe una variada literatura sobre los métodos iterativos para resolver ecuaciones y sistemas, que abarca conceptos como, eficiencia, optimalidad, estabilidad, entre otros importantes temas. En este estudio obtenemos nuevos métodos iterativos que mejoran algunos conocidos en términos de orden o eficiencia, es decir que obtienen mejores aproximaciones con menor coste computacional. La convergencia de los métodos iterativos suele estudiarse desde el punto de vista local. Esto significa que se obtienen resultados de convergencia imponiendo condiciones a la ecuación en un entorno de la solución. Obviamente, estos resultados no son aplicables si no la conocemos. Otro punto de vista, que abordamos en este trabajo, es el estudio semilocal que, imponiendo condiciones en un entorno de la estimación inicial, proporciona un entorno de dicho punto que contiene la solución y garantiza la convergencia del método iterativo a la misma. Finalmente, desde un punto de vista global, estudiamos el comportamiento de los métodos iterativos en función de la estimación inicial, mediante el estudio de la dinámica de las funciones racionales asociadas a estos métodos. La presente memoria recoge los resultados de varios artículos de nuestra autoría, en los que se tratan distintos aspectos de la materia, como son, las peculiaridades de la convergencia en el caso de raíces múltiples, la posibilidad de aumentar el orden de un método óptimo de orden cuatro a orden ocho, manteniendo la optimalidad en el caso de raíces múltiples, el estudio de la convergencia semilocal en un método de alto orden, así como el comportamiento dinámico de algunos métodos iterativos. / [CA] La resolució d'equacions i sistemes no lineals és un tema de gran interés teoricopràctic, perquè molts models matemàtics de la ciència o de la indústria s'expressen mitjançant sistemes no lineals o equacions diferencials o integrals que, mitjançant tècniques de discretizació, donen lloc a aquests sistemes. Atés que generalment és difícil, si no impossible, resoldre analíticament les equacions no lineals, l'eina més estesa són els mètodes iteratius, que tracten d'obtindre aproximacions cada vegada més precises de les solucions partint de determinades estimacions inicials. Existeix una variada literatura sobre els mètodes iteratius per a resoldre equacions i sistemes, que abasta conceptes com ordre d'aproximació, eficiència, optimalitat, estabilitat, entre altres importants temes. En aquest estudi obtenim nous mètodes iteratius que milloren alguns coneguts en termes d'ordre o eficiència, és a dir que obtenen millors aproximacions amb menor cost computacional. La convergència dels mètodes iteratius sol estudiar-se des del punt de vista local. Això significa que s'obtenen resultats de convergència imposant condicions a l'equació en un entorn de la solució. Òbviament, aquests resultats no són aplicables si no la coneixem. Un altre punt de vista, que abordem en aquest treball, és l'estudi semilocal que, imposant condicions en un entorn de l'estimació inicial, proporciona un entorn d'aquest punt que conté la solució i garanteix la convergència del mètode iteratiu a aquesta. Finalment, des d'un punt de vista global, estudiem el comportament dels mètodes iteratius en funció de l'estimació inicial, mitjançant l'estudi de la dinàmica de les funcions racionals associades a aquests mètodes. La present memòria recull els resultats de diversos articles de la nostra autoria, en els quals es tracten diferents aspectes de la matèria, com són, les peculiaritats de la convergència en el cas d'arrels múltiples, la possibilitat d'augmentar l'ordre d'un mètode òptim d'ordre quatre a ordre huit, mantenint l'optimalitat en el cas d'arrels múltiples, l'estudi de la convergència semilocal en un mètode d'alt ordre, així com el comportament dinàmic d'alguns mètodes iteratius. / [EN] The resolution of nonlinear equations and systems is a subject of great theoretical and practical interest, since many mathematical models in science or industry are expressed through nonlinear systems or differential or integral equations that, by means of discretization techniques, give rise to such systems. Since it is generally difficult, if not impossible, to solve nonlinear equations analytically, the most widely used tool is iterative methods, which try to obtain increasingly precise approximations of the solutions based on certain initial estimates. There is a varied literature on iterative methods for solving equations and systems, which covers concepts of order of approximation, efficiency, optimality, stability, among other important topics. In this study we obtain new iterative methods that improve some known ones in terms of order or efficiency, that is, they obtain better approximations with lower computational cost. The convergence of iterative methods is usually studied locally. This means that convergence results are obtained by imposing conditions on the equation in a neighbourhood of the solution. Obviously, these results are not applicable if we do not know it. Another point of view, which we address in this work, is the semilocal study that, by imposing conditions in a neighbourhood of the initial estimation, provides an environment of this point that contains the solution and guarantees the convergence of the iterative method to it. Finally, from a global point of view, we study the behaviour of iterative methods as a function of the initial estimation, by studying the dynamics of the rational functions associated with these methods. This report collects the results of several articles of our authorship, in which different aspects of the matter are dealt with, such as the peculiarities of convergence in the case of multiple roots, the possibility of increasing the order of an optimal method from order four to order eight, maintaining optimality in the case of multiple roots, the study of semilocal convergence in a high-order method, as well as the dynamic behaviour of some iterative methods. / Cevallos Alarcón, FA. (2023). Métodos iterativos para la resolución de problemas aplicados transformados a sistemas no lineales [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/193495

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2017-03-22T20:23:25Z No. of bitstreams: 2 Tese - Gilson do Nascimento Silva - 2017.pdf: 2015008 bytes, checksum: e0148664ca46221978f71731aeabfa36 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-03-23T11:30:21Z (GMT) No. of bitstreams: 2 Tese - Gilson do Nascimento Silva - 2017.pdf: 2015008 bytes, checksum: e0148664ca46221978f71731aeabfa36 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-03-23T11:30:21Z (GMT). No. of bitstreams: 2 Tese - Gilson do Nascimento Silva - 2017.pdf: 2015008 bytes, checksum: e0148664ca46221978f71731aeabfa36 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-03-13 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / We consider Newton’s method for solving a generalized equation of the form f(x) + F(x) 3 0, where f : Ω → Y is continuously differentiable, X and Y are Banach spaces, Ω ⊆ X is open and F : X ⇒ Y has nonempty closed graph. Assuming strong regularity of the equation and that the starting point satisfies Kantorovich’s conditions, we show that the method is quadratically convergent to a solution, which is unique in a suitable neighborhood of the starting point. In addition, a local convergence analysis of this method is presented. Moreover, using convex optimization techniques introduced by S. M. Robinson (Numer. Math., Vol. 19, 1972, pp. 341-347), we prove a robust convergence theorem for inexact Newton’s method for solving nonlinear inclusion problems in Banach space, i.e., when F(x) = −C and C is a closed convex set. Our analysis, which is based on Kantorovich’s majorant technique, enables us to obtain convergence results under Lipschitz, Smale’s and Nesterov-Nemirovskii’s self-concordant conditions. / N´os consideraremos o m´etodo de Newton para resolver uma equa¸c˜ao generalizada da forma f(x) + F(x) 3 0, onde f : Ω → Y ´e continuamente diferenci´avel, X e Y s˜ao espa¸cos de Banach, Ω ⊆ X ´e aberto e F : X ⇒ Y tem gr´afico fechado n˜ao-vazio. Supondo regularidade forte da equa¸c˜ao e que o ponto inicial satisfaz as hip´oteses de Kantorovich, mostraremos que o m´etodo ´e quadraticamente convergente para uma solu¸c˜ao, a qual ´e ´unica em uma vizinhan¸ca do ponto inicial. Uma an´alise de convergˆencia local deste m´etodo tamb´em ´e apresentada. Al´em disso, usando t´ecnicas de otimiza¸c˜ao convexa introduzida por S. M. Robinson (Numer. Math., Vol. 19, 1972, pp. 341-347), provaremos um robusto teorema de convergˆencia para o m´etodo de Newton inexato para resolver problemas de inclus˜ao n˜ao–linear em espa¸cos de Banach, i.e., quando F(x) = −C e C ´e um conjunto convexo fechado. Nossa an´alise, a qual ´e baseada na t´ecnica majorante de Kantorovich, nos permite obter resultados de convergˆencia sob as condi¸c˜oes Lipschitz, Smale e Nesterov-Nemirovskii auto-concordante.

<p><span style="left: 81.5833px; top: 720.322px; font-size: 17.5px; font-family: serif; transform: scaleX(1.07268);">U doktorskoj disertaciji razmatrani su iterativni postupci za re&scaron;avanje nelinearnih komplementarnih problema (NCP). Problemi ovakvog tipa javljaju se u teoriji optimizacije, inženjerstvu i ekonomiji. Matematički modeli mnogih prirodnih, dru&scaron;tvenih i tehničkih procesa svode se takođe na ove probleme. Zbog izuzetno velike zastupljenosti NCP problema, njihovo re&scaron;avanje je veoma aktuelno. Među mnogobrojnim numeričkim postupcima koji se koriste u tu svrhu, u ovoj disertaciji posebna pažnja posvećena je<br />generalizovanim postupcima Njutnovog tipa i iterativnim postupcima sa re-gularizacijom matrice jakobijana. Definisani su novi postupci za re&scaron;avanje NCP i dokazana je njihova lokalna ili globalna konvergencija. Dobijeni teorijski rezultati testirani su na relevantnim numeričkim primerima. </span></p> / <p>Iterative methods for nonlinear complementarity problems (NCP) are con-sidered in this doctoral dissertation. NCP problems appear in many math-ematical models from economy, engineering and optimization theory. Solv-ing NCP is very atractive in recent years. Among many numerical methods for NCP, we are interested in generalized Newton-type methods and Jaco-bian smoothing methođs. Several new methods for NCP are defined in this dissertation and their local or global convergence is proved. Theoretical results are tested on relevant numerical examples.</p>

<p>U doktorskoj diseratciji posmatrani su singularni nelinearni problemi. U prvom&nbsp;poglavlju predstavljene su oznake i osnovne definicije i teoreme koje se koriste u&nbsp;disertaciji. U drugom poglavlju prikazani su poznati postupci i njihovo pona&scaron;anje&nbsp;u slučajevima da je re&scaron;enje regularno ili singularno. Takođe su pokazane poznate&nbsp;modifikacije ovih postupaka kako bi se pobolj&scaron;ala konvergencija. Posebno su&nbsp;predstavljena četiri kvazi-Njutnova metoda i predložene njihove modifikacije u&nbsp;slučaju singularnosti re&scaron;enja. U trećem poglavlju predstavljeni su teorijski okvir&nbsp;pri definisanju graničnih sistema i neki poznati algoritmi za njihovo re&scaron;avanje i&nbsp;definisan je novi algoritam koji je podjednako efikasan ali jeftiniji za rad jer ne&nbsp;uključuje izračunavanje izvoda. Takođe, predložena je kombinacija definisanog&nbsp;algortitma sa metodom negativnog gradijenta, kao i algoritam koji predstavlja&nbsp;primenu poznatog algoritma na definisani granični sistem. U četvrtom poglavlju&nbsp;predstavljeni su numerički rezultati dobijeni primenom definisanih algoritama na&nbsp;relevantne primere i potvrđeni su teorijski dobijeni rezultati.</p> / <p>In this doctoral thesis nonlinear singular problems were observed. The first&nbsp;chapter presents basic definitions and theorems that are used in the thesis. The&nbsp;second chapter presents several methods that are commonly used and their&nbsp;behavior if the solution is regular or singular. Also, some known modifications to&nbsp;these methods are presented in order to improve convergence. In addition four&nbsp;quasi-Newton methods and their modifications in the case the singularity of the&nbsp;solution. The third chapter consists of the theoretical foundation for defining the&nbsp;bordered system, some known algorithms for solving them and new algorithm is&nbsp;defined to accelerate convergence to a singular solution. New algorithm is&nbsp;efficient but cheaper for the use since there is no derivative evaluations in it. It is&nbsp;presented synthesis of new algorithm with negative gradient method and using&nbsp;one of well known method on the bordered system as well. The fourth chapter&nbsp;presents the numerical results obtained by the defined algorithms on the relevant&nbsp;examples and theoretical results are confirmed.</p>