Wyniki wyszukiwania - BazTech

1

Preconditiong and Inoculation of Low Sulphur Grey Iron

Hossain Md Sojib S., Bazlur B. Rashid A. K. M.

Archives of Foundry Engineering

|

2020

|

Vol. 20, iss. 1

61--66

EN

For quality grey cast iron production, the challenging issues are to avoid cementite structure and obtain the desired graphite morphology with proper matrix as well as hardness. The objective of the present research is to find out the right combination of preconditioner and inoculant that may help to overcome the challenges. In this work, sulphur content is kept low (0.01%). Two preconditioners namely metallurgical SiC and zirconium bearing FeSi with two types of inoculant are individually used to make four combinations of sample and for each case metal is poured into the green sand mould. Finally Brinell hardness and graphite morphology is observed in the thickest and thinnest portions of the castings. Metallurgical SiC with barium bearing inoculant gives better graphite morphology and hardness than strontium bearing inoculant, on the other hand zirconium bearing FeSi gives more satisfying result than SiC with every type of inoculant. Among all of the combinations Zr bearing preconditioner with Ba bearing inoculant gives good graphite morphology with best mechanical properties in both thickest and thinnest portions of the casting.

2

ILU preconditioning based on the FAPINV algorithm

Salkuyeh D. K., Rafiei A., Roohani H.

Opuscula Mathematica

|

2015

|

Vol. 35, no. 2

235--250

EN

A technique for computing an ILU preconditioner based on the factored approximate inverse (FAPINV) algorithm is presented. We show that this algorithm is well-defined for H-matrices. Moreover, when used in conjunction with Krylov-subspace-based iterative solvers such as the GMRES algorithm, results in reliable solvers. Numerical experiments on some test matrices are given to show the efficiency of the new ILU preconditioner.

3

Incomplete column-row factorization method for general sparse matrices

Saukh S.

Przegląd Elektrotechniczny

|

2010

|

R. 86, nr 1

127-129

EN

Incomplete column-row factorization method is proposed. This method does not require permutations of rows and columns in submatrices. Calculable stability of the method is provided by search of pivots in submatrices to minimize divergence on the Frobenius norm between the transformed and regenerate submatrices. Significance of entries of the factor matrices is specified by comparison of norms of rows and columns of transformed and subtracted submatrices. The new method requires less memory than AINV, ILUT and RIF-Ns methods.

PL

Artykuł proponuje niekompletną faktoryzację kolumnowo-wierszową, która nie wymaga permutacji wierszy ani kolumn podmacierzy. Stabilność obliczeń jest zapewniana przez wybór elementów dominujących aby zminimalizować normę Frobeniusa dla odpowiednich podmacierzy. Znaczenie współczynników jest określane przez porównanie norm wierszy i kolumn. Nowa metoda ma mniejsze zapotrzebowanie na pamięć, niż AINV, ILUT i RIF-N.

4

The convergence rate and parallel performance of a 3D elliptic solver

Lirkov I., Vutov V.

Systems Science

|

2006

|

Vol. 32, no 4

73-81

EN

It was shown that block-circulant preconditioners applied to a conjugate gradient method used to solve structured sparse linear systems arising from 2D or 3D elliptic problems have good numerical properties and a potential for high parallel efficiency. In this paper, the convergence rate and the parallel performance of a circulant block-factorization based preconditioner applied to a 3D problem are analyzed. A portable parallel code is developed based on Message Passing Interface (MPI) standards. The numerical tests performed on parallel computer systems demonstrate the level of efficiency of the algorithm developed.