East Asian J. Appl. Math., 12 (2022), pp. 145-162.
Published online: 2021-10
Cited by
- BibTex
- RIS
- TXT
An efficient least-square support vector machine (LS-SVM) method for a two time-scale variable-order time-fractional diffusion equation is developed. The method is particularly suitable for problems defined on complex physical domains or in high spatial dimensions. The problem is discretised by the L1 scheme and the Euler method. The temporal semi-discrete problem obtained is reformulated as a minimisation problem. The Karush-Kuhn-Tucker optimality condition is used to determine the minimiser of the optimisation problem and, hence, the solution sought. Numerical experiments show the efficiency and high accuracy of the method.
}, issn = {2079-7370}, doi = {https://doi.org/10.4208/eajam.310121.120821}, url = {http://global-sci.org/intro/article_detail/eajam/19925.html} }An efficient least-square support vector machine (LS-SVM) method for a two time-scale variable-order time-fractional diffusion equation is developed. The method is particularly suitable for problems defined on complex physical domains or in high spatial dimensions. The problem is discretised by the L1 scheme and the Euler method. The temporal semi-discrete problem obtained is reformulated as a minimisation problem. The Karush-Kuhn-Tucker optimality condition is used to determine the minimiser of the optimisation problem and, hence, the solution sought. Numerical experiments show the efficiency and high accuracy of the method.