Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 94
PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON ENGINEERING COMPUTATIONAL TECHNOLOGY Edited by:
Paper 40
A Dynamic and Analytical Study of the Damped Newton's Method J.M. Gutiérrez, A.A. Magreñán and N. Romero
Department of Mathematics and Computation, University of La Rioja, Spain , "A Dynamic and Analytical Study of the Damped Newton's Method", in , (Editors), "Proceedings of the Seventh International Conference on Engineering Computational Technology", Civil-Comp Press, Stirlingshire, UK, Paper 40, 2010. doi:10.4203/ccp.94.40
Keywords: Newton's method, Kantorovich theorem, majorizing sequences, damped Newtons's method.
Summary
This paper begins by introducing the Newton's method, the most known method to solve systems of non-linear equations [1,2,3]. Kantorovich introduced this method to solve functional equations [4] and gives conditions to guarantee the convergence of this method. After that introduction we explain the main framework of our study, the damped Newton's method. We make a study of this method taking into account the Newton-Kantorovich theory (see [4,5,6] for more details). Taking advantage of this theorem we can construct a majorizing sequence that allow us to improve the condition that explains the parameters for the existence of the inverse and the Lipschitz's condition [7], is to said, Kantorovich established that h has to be less or equal than 0.5, but we give another condition taking into account the damping factor. In the next section we make a dynamical study of the damped Newton's method, this study consists of realizing how the fractal dimension [8] varies with different damping factors and linking it with the continuous Newton's method [9]. Finally we provide numerical experiments using our results.
References
purchase the full-text of this paper (price £20)
go to the previous paper |
|