Pasar al contenido principal

PRECONDICIONADORES PARA SISTEMAS DE ECUACIONES LINEALES, PROBLEMAS DE MINIMOS CUADRADOS, CALCULO DE VALORES PROPIOS Y APLICACIONES TECNOLOGICAS

MTM2014-58159-P

•
Nombre agencia financiadora Ministerio de Economía y Competitividad
Acrónimo agencia financiadora MINECO
Programa Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D (Excelencia / Generación de Conocimiento)
Año convocatoria 2014
Plan Estatal 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Unidad de gestión Secretaría de Estado de Educación, Formación Profesional y Universidades
Centro beneficiario UNIVERSITAT POLITÈCNICA DE VALÈNCIA
Identificador persistente http://dx.doi.org/10.13039/501100003329

Resultados relacionados

Resultados totales (Incluyendo duplicados): 18
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The Hadamard Product of a Nonsingular General H--Matrix and Its Inverse Transpose Is Diagonaly Dominant

  • Bru García, Rafael
  • Gasso Matoses, María Teresa
  • Gimenez Manglano, María Isabel
  • Scott, José A.
[EN] We study the combined matrix of a nonsingular H-matrix. Iese matrices can belong to two diRerent H-matrices classes: the most common, invertible class, and one particular class named mixed class. DiRerent results regarding diagonal dominance of the inverse matrix and the combined matrix of a nonsingular H-matrix belonging to the referred classes are obtained. We conclude that the combined matrix of a nonsingular H-matrix is always diagonally dominant and then it is an H-matrix. In particular, the combined matrix in the invertible class remains in the same class., Ie authors thank the referee for suggesting changes that have improved the presentation of the paper. Iis research was supported by Spanish DGI Grant no. MTM2014-58159-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Spatial modes for the neutron diffusion equation and their computation

  • Carreño, Amanda|||0000-0003-2302-1157
  • Vidal-Ferràndiz, Antoni|||0000-0001-5449-7356
  • Ginestar Peiro, Damián|||0000-0003-1243-6648
  • Verdú Martín, Gumersindo Jesús|||0000-0001-5098-080X
[EN] Different spatial modes can be defined for the neutron diffusion equation such as the k; a and c-modes. These modes have been successfully used for the analysis of nuclear reactor characteristics. In this work, these modes are studied using a high order finite element method to discretize the equations and also different methods to solve the resulting algebraic eigenproblems, are compared. Particularly, Krylov subspace methods and block-Newton methods have been studied. The performance of these methods has been tested in several 3D benchmark problems: a homogeneous reactor and several configurations of NEACRP reactor., This work has been partially supported by Spanish Ministerio de Economia y Competitividad under projects ENE2014-59442-P, MTM2014-58159-P and BES-2015-072901.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Low-rank updates of balanced incomplete factorization preconditioners

  • Cerdán Soriano, Juana Mercedes
  • Marín Mateos-Aparicio, José|||0000-0002-7825-2836
  • Mas Marí, José|||0000-0002-2835-974X
[EN] Let Ax = b be a large and sparse system of linear equations where A is a nonsingular matrix. An approximate solution is frequently obtained by applying preconditioned terations. Consider the matrix B = A + PQT where P,Q ∈ Rn×k are full rank matrices. In this work, we study the problem of updating a previously computed preconditioner for A in order to solve the updated linear system Bx = b by preconditioned iterations. In particular, we propose a method for updating a Balanced Incomplete Factorization preconditioner. The strategy is based on the computation of an approximate Inverse Sherman-Morrison decomposition for an equivalent augmented linear system. Approximation properties of the preconditioned matrix and an analysis of the computational cost of the algorithm are studied. Moreover the results of the numerical experiments with different types of problems show that the proposed method contributes to accelerate the convergence., This work was supported by the Spanish Ministerio de Economia y Competitividad under grant MTM2014-58159-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Compact quasi-Newton preconditioners for symmetric positive definite linear systems

  • BERGAMASCHI, LUCA
  • MARTINEZ, ANGELES
  • Marín Mateos-Aparicio, José|||0000-0002-7825-2836
[EN] In this paper, preconditioners for the conjugate gradient method are studied to solve the Newton system with symmetric positive definite Jacobian. In particular, we define a sequence of preconditioners built by means of Symmetric Rank one (SR1) and Broyden-Fletcher-Goldfarb-Shanno (BFGS) low-rank updates. We develop conditions under which the SR1 update maintains the preconditioner symmetric positive definite. Spectral analysis of the SR1 preconditioned Jacobians shows an improved eigenvalue distribution as the Newton iteration proceeds. A compact matrix formulation of the preconditioner update is developed which reduces the cost of its application and is more suitable to parallel implementation. Some notes on the implementation of the corresponding Inexact Newton method are given and some numerical results on a number of model problems illustrate the efficiency of the proposed preconditioners., This work was supported by the Spanish Ministerio de Economia y Competitividad under grants MTM2014-58159-P, MTM2017-85669-P, and MTM2017-90682-REDT. The first and third authors have been also partially supported by the INdAM Research group GNCS, 2020 Project: Optimization and advanced linear algebra for problems arising from PDEs.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Nikodym boundedness property for webs in sigma-algebras

  • López Alfonso, Salvador|||0000-0003-1655-2320
  • Mas Marí, José|||0000-0002-2835-974X
  • Moll López, Santiago Emmanuel|||0000-0003-3388-5135
[EN] A subset B of an algebra A of subsets of Omega is said to have property N if a B-pointwise bounded subset M of ba(A ) is uniformly bounded on A , where ba(A ) is the Banach space of the real (or complex) finitely additive measures of bounded variation defined on A with the norm variation. Moreover B is said to have property sN if for each increasing countable covering (B_m)_m of B there exists B_n which has property N and B is said to have property wN if given the increasing countable coverings (B_m_1 )_m_1 of B and (B_m_1m_2...m_pm_(p+1) )_m_(p+1) of B_m_1m_2...m_p , for each p,m_i &#8712; N, 1<= i <= p + 1, there exists a sequence (n_i )_i such that each B_n_1n_2...n_r , r &#8712; N, has property N. For a &#963;-algebra S of subsets of Omega it has been proved that S has property N (Nikodym Grothendieck), property sN (Valdivia) and property w(sN) (Kakol López-Pellicer). We give a proof of property wN for a &#963;-algebra S which is independent of properties N and sN. This result and the equivalence of properties wN and w2N enable us to give some applications to localization of bounded additive vector measures., This work was supported for the second named author by the Spanish Ministerio de Economía y Competitividad under Grant MTM2014-58159-P
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Block Preconditioning Matrices for the Newton Method to Compute the Dominant lambda-Modes Associated with the Neutron Diffusion Equation

  • Carreño, Amanda|||0000-0003-2302-1157
  • Vidal-Ferràndiz, Antoni|||0000-0001-5449-7356
  • Ginestar Peiro, Damián|||0000-0003-1243-6648
  • Verdú Martín, Gumersindo Jesús|||0000-0001-5098-080X
  • Bergamaschi, Luca
  • Martinez, Angeles
[EN] In nuclear engineering, the lambda-modes associated with the neutron diffusion equation are applied to study the criticality of reactors and to develop modal methods for the transient analysis. The differential eigenvalue problem that needs to be solved is discretized using a finite element method, obtaining a generalized algebraic eigenvalue problem whose associated matrices are large and sparse. Then, efficient methods are needed to solve this problem. In this work, we used a block generalized Newton method implemented with a matrix-free technique that does not store all matrices explicitly. This technique reduces mainly the computational memory and, in some cases, when the assembly of the matrices is an expensive task, the computational time. The main problem is that the block Newton method requires solving linear systems, which need to be preconditioned. The construction of preconditioners such as ILU or ICC based on a fully-assembled matrix is not efficient in terms of the memory with the matrix-free implementation. As an alternative, several block preconditioners are studied that only save a few block matrices in comparison with the full problem. To test the performance of these methodologies, different reactor problems are studied., This work has been partially supported by the Spanish Ministerio de Economia y Competitividad under Projects ENE2014-59442-P, MTM2014-58159-P, and BES-2015-072901.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A two-level ILU preconditioner for electromagnetic applications

  • Cerdán Soriano, Juana Mercedes
  • Marín Mateos-Aparicio, José|||0000-0002-7825-2836
  • Mas Marí, José|||0000-0002-2835-974X
[EN] Computational electromagnetics based on the solution of the integral form of Maxwell s equations with boundary element methods require the solution of large and dense linear systems. For large-scale problems the solution is obtained by using iterative Krylov-type methods provided that a fast method for performing matrix vector products is available. In addition, for ill-conditioned problems some kind of preconditioning technique must be applied to the linear system in order to accelerate the convergence of the iterative method and improve its performance. For many applications it has been reported that incomplete factorizations often suffer from numerical instability due to the indefiniteness of the coefficient matrix. In this context, approximate inverse preconditioners based on Frobenius-norm minimization have emerged as a robust and highly parallel alternative. In this work we propose a two-level ILU preconditioner for the preconditioned GMRES method. The computation and application of the preconditioner is based on graph partitioning techniques. Numerical experiments are presented for different problems and show that with this technique it is possible to obtain robust ILU preconditioners that perform competitively compared with Frobenius-norm minimization preconditioners., This work was supported by the Spanish Ministerio de Economía y Competitividad under grant MTM2014-58159-P and MTM2015-68805-REDT.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Statistical tools and control of internal lubricant content of inhalation grade HPMC capsules during manufacture

  • Ayala, Guillermo
  • Diez, Fernando
  • Gasso Matoses, María Teresa
  • Jones, Brian E.
  • Martin-Portugues, Rafae
  • Ramiro-Aparicio, Juan
The internal lubricant content (ILC) of inhalation grade HPMC capsules is a key factor to ensure good powder release when the patient inhales a medicine from a dry powder inhaler (DPI). Powder release from capsules has been shown to be influenced by the ILC. The characteristics used to measure this are the emitted dose, fine particle fraction and mass median aerodynamic diameter. In addition the ILC level is critical for capsule shell manufacture because it is an essential part of the process that cannot work without it. An experiment has been applied to the manufacture of inhalation capsules with the required ILC. A full factorial model was used to identify the controlling factors and from this a linear model has been proposed to improve control of the process. (C) 2016 Elsevier B.V. All rights reserved., This research was supported by Qualicaps Europe, S.A.U.; Spanish DGI Grant no. MTM2014-58159-P (M.T. Gasso) and G. Ayala (DPI2013-47279-C2-1-R),
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

HP Finite element method for the simplified Pn equations

  • Vidal-Ferràndiz, Antoni|||0000-0001-5449-7356
  • Carreño, Amanda|||0000-0003-2302-1157
  • Ginestar Peiro, Damián|||0000-0003-1243-6648
  • Verdú Martín, Gumersindo Jesús|||0000-0001-5098-080X
[EN] The simplified spherical harmonics approximation to the neutron transport equation defines the neutron distribution inside a nuclear reactor core more accurately than the widely used neutron diffusion equation. To solve this equation, a goal oriented h-adaptable finite element method is employed. The proposed adaptive mesh refinement (AMR) method splits mesh cells whose errors are large and whose importance towards the computational goal is high. In this work, the computational goal is set as the maximum averaged power in an assembly of the reactor as this usually is a design constraint. The goal oriented error estimator is compared with the usual Kelly error indicator and also with a uniform global refinement. The goal oriented error estimator demonstrates to be the most accurate strategy in order to minimise the error in the defined quantity of interest and it is as much efficient as standard error estimators to compute the fundamental eigenvalue., The work has been partially supported by the Ministerio de Economía y Competitividad under projects ENE 2014-59442-P and MTM2014-58159-P, the Generalitat Valenciana under PROMETEO II/2014/008 and the Universitat Politecnica de València under FPI-2013
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

CMMSE algorithms for constructing doubly stochastic matrices with the relative gain array (combined matrix) A circle A(-T)

  • Fuster Capilla, Robert Ricard
  • Gasso Matoses, María Teresa
  • Gimenez Manglano, María Isabel
[EN] The Combined matrix of a nonsingular matrix A is defined by phi(A)=A T where degrees means the Hadamard (entrywise) product. If the matrix A describes the relation between inputs and outputs in a multivariable process control, phi(A) describes the relative gain array (RGA) of the process and it defines the Bristol method (IEEE Trans Autom Control 1:133-134, 1966) often used for Chemical processes (McAvoy in Interaction analysis: principles and applications. Instrument Society of America, Pittsburgh, 1983; Papadourakis et al. in Ind Eng Chem Res 26(6):1259-1262, 1987; Wang et al. in Chem Eng Technol, 10.1002/ceat.201500202, 2016; Kariwala et al. in Ind Eng Chem Res 45(5):1751-1757, 10.1021/ie050790r, 2006; Golender et al. in J Chem Inf Comput Sci 21(4):196-204, 10.1021/ci00032a004, 1981). The combined matrix has been studied in several works such as Bru et al. (J Appl Math, 10.1155/2014/182354, 2014), Fiedler and Markham (Linear Algebra Appl 435:1945-1955, 2011) and Johnson and Shapiro (SIAM J Algebraic Discrete Methods 7:627-644, 1986). Since phi(A)=(cij) has the property of Sigma kcik=Sigma kckj=1,i,j, when phi(A)>= 0, phi(A) is a doubly stochastic matrix. In certain chemical engineering applications a diagonal of the RGA in wchich the entries are near 1 is used to determine the pairing of inputs and outputs for further design analysis. Applications of these matrices can be found in Communication Theory, related with the satellite-switched time division multiple-access systems, and about a doubly stochastic automorphism of a graph. In this paper we present new algorithms to generate doubly stochastic matrices with the Combined matrix using Hessenberg matrices in Sect.3 and orthogonal/unitary matrices in Sect.4. In addition, we discuss what kind of doubly stochastic matrices are obtained with our algorithms and the possibility of generating a particular doubly stochastic matrix by the map phi., This work has been supported by Spanish Ministerio de Economia y Competitividad Grants MTM2014-58159-P, MTM2017-85669-P and MTM2017-90682-REDT.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Diagonal entries of the combined matrix of a totally negative matrix

  • Bru García, Rafael
  • Gasso Matoses, María Teresa
  • Gimenez Manglano, María Isabel
  • Santana-De Asis, Máximo De Jesús
[EN] The combined matrix of a nonsingular matrix A is the Hadamard (entrywise) product . This paper deals with the characterization of the diagonal entries of a combined matrix C(A) of a given nonsingular real matrix A. A partial answer describing the diagonal entries of C(A) in the positive definite case was given by Fiedler in 1964. Recently in 2011, Fiedler and Markham characterized the sequence of diagonal entries of the combined matrix C(A) for any totally positive matrix A when the size is 3. For this case, we characterize totally negative matrices and we find necessary and sufficient conditions for the sequence of diagonal entries of C(A), in both cases, symmetric and nonsymmetric., This work was supported by Spanish DGI [grant number MTM-2014-58159-P]; Dominican Republic FONDOCYT [grant number 2015-1D2-166].
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Preconditioners for nonsymmetric linear systems with low-rank skew-symmetric part

  • Cerdán Soriano, Juana Mercedes
  • Marín Mateos-Aparicio, José|||0000-0002-7825-2836
  • Mas Marí, José|||0000-0002-2835-974X
  • Guerrero-Flores, Danny Joel
[EN] We present a preconditioning technique for solving nonsymmetric linear systems Ax = b, where the coefficient matrix A has a skew-symmetric part that can be well approximated with a skew-symmetric low-rank matrix. The method consists of updating a preconditioner obtained from the symmetric part of A. We present some results concerning to the approximation properties of the preconditioner and the spectral properties of the preconditioning technique. The results of the numerical experiments performed show that our strategy is competitive compared with some specific methods. (C) 2018 Elsevier B.V. All rights reserved., This work was supported by the Spanish Ministerio de Economia y Competitividad under grants MTM2014-58159-P and MTM2015-68805-REDT.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Preconditioners based on the ISM factorization

  • Bru García, Rafael
  • Cerdán Soriano, Juana Mercedes
  • Marín Mateos-Aparicio, José|||0000-0002-7825-2836
  • Mas Marí, José|||0000-0002-2835-974X
  • Tuma, Miroslav
In this paper we survey our work on preconditioners based on the Inverse Sherman-Morrison factorization. The most important theoretical results are also summarized and some numerical conclusions are provided., This work was supported by the Spanish Ministerio de Economia y Competitividad under grant MTM2014-58159-P and by the project 13-06684S of the Grant Agency of the Czech Republic.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Updating preconditioners for modified least squares problems

  • Marín Mateos-Aparicio, José|||0000-0002-7825-2836
  • Mas Marí, José|||0000-0002-2835-974X
  • Guerrero-Flores, Danny Joel
  • Hayami, K.
[EN] In this paper, we analyze how to update incomplete Cholesky preconditioners to solve least squares problems using iterative methods when the set of linear relations is updated with some new information, a new variable is added or, contrarily, some information or variable is removed from the set. Our proposed method computes a low-rank update of the preconditioner using a bordering method which is inexpensive compared with the cost of computing a new preconditioner. Moreover, the numerical experiments presented show that this strategy gives, in many cases, a better preconditioner than other choices, including the computation of a new preconditioner from scratch or reusing an existing one., Partially supported by Spanish Grants MTM2014-58159-P and MTM2015-68805-REDT.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Schwarz type preconditioners for the neutron diffusion equation

  • Vidal-Ferràndiz, Antoni|||0000-0001-5449-7356
  • Ginestar Peiro, Damián|||0000-0003-1243-6648
  • Verdú Martín, Gumersindo Jesús|||0000-0001-5098-080X
  • González Pintor, Sebastián
  • Demazière, Christophe
[EN] Domain decomposition is a mature methodology that has been used to accelerate the convergence of partial differential equations. Even if it was devised as a solver by itself, it is usually employed together with Krylov iterative methods improving its rate of convergence, and providing scalability with respect to the size of the problem. In this work, a high order finite element discretization of the neutron diffusion equation is considered. In this problem the preconditioning of large and sparse linear systems arising from a source driven formulation becomes necessary due to the complexity of the problem. On the other hand, preconditioners based on an incomplete factorization are very expensive from the point of view of memory requirements. The acceleration of the neutron diffusion equation is thus studied here by using alternative preconditioners based on domain decomposition techniques inside Schur complement methodology. The study considers substructuring preconditioners, which do not involve overlapping, and additive Schwarz preconditioners, where some overlapping between the subdomains is taken into account. The performance of the different approaches is studied numerically using two-dimensional and three-dimensional problems. It is shown that some of the proposed methodologies outperform incomplete LU factorization for preconditioning as long as the linear system to be solved is large enough, as it occurs for three-dimensional problems. They also outperform classical diagonal Jacobi preconditioners, as long as the number of systems to be solved is large enough in such a way that the overhead of building the pre-conditioner is less than the improvement in the convergence rate. (C) 2016 Elsevier B.V. All rights reserved., The work has been partially supported by the spanish Ministerio de Economía y Competitividad under projects ENE 2014-59442-P and MTM2014-58159-P, the Generalitat Valenciana under the project PROMETEO II/2014/008 and the Universitat Politècnica de València under the project FPI-2013. The work has also been supported partially by the Swedish Research Council (VR-Vetenskapsrådet) within a framework grant called DREAM4SAFER, research contract C0467701.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A Hadamard product involving inverse-positive matrices

  • Gasso Matoses, María Teresa
  • Abad Rodríguez, Manuel Francisco
  • Torregrosa Sánchez, Juan Ramón|||0000-0002-9893-0761
In this paperwe study the Hadamard product of inverse-positive matrices.We observe that this class of matrices is not closed under the Hadamard product, but we show that for a particular sign pattern of the inverse-positive matrices A and B, the Hadamard product A ◦ B−1 is again an inverse-positive matrix., Research supported by Spanish DGI grant number MTM2014-58159-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Pin-wise homogenization for SPN neutron transport approximation using the finite element method

  • Vidal-Ferràndiz, Antoni|||0000-0001-5449-7356
  • Ginestar Peiro, Damián|||0000-0003-1243-6648
  • Verdú Martín, Gumersindo Jesús|||0000-0001-5098-080X
  • Gonzalez-Pintor, S.
  • Demaziere, C.
[EN] The neutron transport equation describes the distribution of neutrons inside a nuclear reactor core. Homogenization strategies have been used for decades to reduce the spatial and angular domain complexity of a nuclear reactor by replacing previously calculated heterogeneous subdomains by homogeneous ones and using a low order transport approximation to solve the new problem. The generalized equivalence theory for homogenization looks for discontinuous solutions through the introduction of discontinuity factors at the boundaries of the homogenized subdomains. In this work, the generalized equivalence theory is extended to the Simplified P-N equations using the finite element method. This extension proposes pin discontinuity factors instead of the usual assembly discontinuity factors and the use of the simplified spherical harmonics approximation rather than diffusion theory. An interior penalty finite element method is used to discretize and solve the problem using discontinuity factors. One dimensional numerical results show that the proposed pin discontinuity factors produce more accurate results than the usual assembly discontinuity factors. The proposed pin discontinuity factors produce precise results for both pin and assembly averaged values without using advanced reconstruction methods. Also, the homogenization methodology is verified against the calculation performed with reference discontinuity factors. (C) 2017 Elsevier B.V. All rights reserved., The work has been partially supported by the spanish Ministerio de Economía y Competitividad under projects ENE 2014-59442-P and MTM2014-58159-P, the Generalitat Valenciana under the project PROMETEO II/2014/008 and the Universitat Politècnica de València under the project FPI-2013. The work has also been supported partially by the Swedish Research Council (VR-Vetenskapsrådet) within a framework grant called DREAM4SAFER, research contract C0467701
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Multilevel method to compute the lambda modes of the neutron diffusion equation

  • Carreño, Amanda|||0000-0003-2302-1157
  • Vidal-Ferràndiz, Antoni|||0000-0001-5449-7356
  • Ginestar Peiro, Damián|||0000-0003-1243-6648
  • Verdú Martín, Gumersindo Jesús|||0000-0001-5098-080X
[EN] Determination of the reactor kinetic characteristics is very important for the design and development of a new reactor system. In this sense, the computation of lambda modes associated to a nuclear power reactor has interest since these modes can be used to analyze the reactor criticality and to develop modal methods to analyze transient situations in the reactor. In this paper, the lambda problem has been discretized using a high order finite element method to obtain a generalized algebraic eigenvalue problem. A multilevel method is proposed to solve this generalized eigenvalue problem combining a hierarchy of meshes with a Modified Block Newton method. The Krylov-Schur method is used to compare the efficiency of the multilevel method solving several benchmark problems., This work has been partially supported by Spanish Ministerio de Economía y Competitividad under projects ENE2014-59442-P, MTM2014-58159-P and BES-2015-072901