A Study of Single and Multi Step Methods To Solve Differential Equations

dc.contributor.advisorHossain, Prof. Dr. Mohammad Arif
dc.contributor.authorKhanam, Mary
dc.date.accessioned2018-08-13T10:02:35Z
dc.date.available2018-08-13T10:02:35Z
dc.date.issued2011-04
dc.descriptionThis thesis is submitted to the Department of Mathematics, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Philosophy in Mathematics, April 2011.
dc.descriptionCataloged from PDF Version of Thesis.
dc.descriptionIncludes bibliographical references (pages 76-76).
dc.description.abstractHere the task was to study single and multi step methods to solve differential equations. As Rungc-Kutta method. a single step method has some property to represent a larnily so attention has been given to that method. It was found that to establish the method of any order the number of unknowns are more than the number of available equations. Thus there are options to choose certain values as someone wants (keeping in mind about the conditions). The opportunity has been taken and FIVE formulas of different orders, two fifth, two sixth and one seventh, has been proposed. Problems are solved to verify their capability and strength and it is found that the percentage errors with respect to the exact values are comparable in magnitude with respect to some available same order methods. In case of multi step methods, extensions in both explicit and implicit methods in terms of order are done. It is done keeping in mind that the extension in order will reduce the error. Extensions in Adams- Bash forth and Adams-Moulton formulas up to TENth order are done and with eighth order of both of them a problem is solved to demonstrate the strength of the extension of these predictor-corrector methods.
dc.identifier.otherID 0551556
dc.identifier.otherhttp://dspace.kuet.ac.bd/handle/20.500.12228/367
dc.identifier.urihttp://hdl.handle.net/20.500.12228/367
dc.language.isoen_US
dc.publisherKhulna University of Engineering & Technology (KUET), Khulna, Bangladesh.
dc.sourceKUET Institutional Repository
dc.subjectSingle and Multi Step Methods
dc.subjectDifferential Equations
dc.subjectRunge-Kutta Method
dc.titleA Study of Single and Multi Step Methods To Solve Differential Equations
dc.typeThesis

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
Full Thesis.pdf
Size:
10.8 MB
Format:
Adobe Portable Document Format

Collections