خسرو سایوند

This paper presents a reliable numerical technique based on Lucas polynomials for a family of fractional differential equations and multi order fractional differential equations by means of the least square method. The fractional derivative is in the Caputo sense. A relevant feature of this approach is the analyzing of the suggested technique by Gauss quadrature method and using the theory of Lagrange multipliers to solve a constrained optimization problem. An upper error bound, the convergence, and error analysis of the scheme are investigated and the CPU time used, the values of maximum errors, the numerical convergence analysis based on the proposed technique for different values of parameters are discussed. Furthermore the results of present technique are compared with the, operational matrix of hybrid basis functions, the Jacobi orthogonal functions and pseudo-spectral scheme. In order to introduce the numerical behavior of the proposed technique in case of nonsmooth solutions, this issue is discussed. In this case, the obtained results imply an elegant superiority of our proposed technique. The numerical examples illustrate the accuracy and performance of the technique. Finally extending the proposed technique to high dimensions and system of fractional differential equations can be examined as a further works.