Identification of a multi-dimensional space-dependent heat source from boundary data

M.S. Hussein, D. Lesnic, B.T. Johansson, A. Hazanee

    Research output: Contribution to journalArticlepeer-review

    Abstract

    We investigate the linear but ill-posed inverse problem of determining a multi-dimensional space-dependent heat source in the parabolic heat equation from Cauchy boundary data. This model is important in practical applications where the distribution of internal sources is to be monitored and controlled with care and accuracy from non-invasive and non-intrusive boundary measurements only. The mathematical formulation ensures that a solution of the inverse problem is unique but the existence and stability are still issues to be dealt with. Even if a solution exists it is not stable with respect to small noise in the measured boundary data hence the inverse problem is still ill-posed. The Landweber method is developed in order to restore stability through iterative regularization. Furthermore, the conjugate gradient method is also developed in order to speed up the convergence. An alternating direction explicit finite-difference method is employed for discretising the well-posed problems resulting from these iterative procedures. Numerical results in two-dimensions are illustrated and discussed.
    Original languageEnglish
    Pages (from-to)202-220
    Number of pages19
    JournalApplied Mathematical Modelling
    Volume54
    Early online date21 Sept 2017
    DOIs
    Publication statusPublished - Feb 2018

    Bibliographical note

    © 2017, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
    http://creativecommons.org/licenses/by-nc-nd/4.0/.

    Keywords

    • Heat equation
    • inverse source problem
    • iterative regularization

    Fingerprint

    Dive into the research topics of 'Identification of a multi-dimensional space-dependent heat source from boundary data'. Together they form a unique fingerprint.

    Cite this