AAA Abstract and Applied Analysis 1687-0409 1085-3375 Hindawi Publishing Corporation 139530 10.1155/2013/139530 139530 Research Article A Spectral Deferred Correction Method for Fractional Differential Equations Xin Jia 1 http://orcid.org/0000-0002-3793-1277 Huang Jianfei 1 Zhao Weijia 1 Zhu Jiang 2 Baleanu Dumitru 1 College of Mathematics Qingdao University Qingdao 266071 China qdu.edu.cn 2 National Laboratory for Scientific Computing MCTI Avenida Getulio Vargas 333 25651-075 Petropolis RJ Brazil lncc.br 2013 22 10 2013 2013 31 05 2013 07 08 2013 19 09 2013 2013 Copyright © 2013 Jia Xin et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A spectral deferred correction method is presented for the initial value problems of fractional differential equations (FDEs) with Caputo derivative. This method is constructed based on the residual function and the error equation deduced from Volterra integral equations equivalent to the FDEs. The proposed method allows that one can use a relatively few nodes to obtain the high accuracy numerical solutions of FDEs without the penalty of a huge computational cost due to the nonlocality of Caputo derivative. Finally, preliminary numerical experiments are given to verify the efficiency and accuracy of this method.

1. Introduction

Fractional calculus, almost as old as the familiar integer-order calculus, has recently gained considerable popularity and importance due to its attractive applications in widespread fields of science and engineering (e.g., ), such as control theory , viscoelasticity , image processing , electromagnetism , anomalous diffusion [10, 11], and hydrology [12, 13]. The overwhelming advantage is that fractional calculus provides an excellent instrument for the description of memory and hereditary properties of various materials and processes. These applications greatly highlight distinct superiorities and unsubstitutability of fractional calculus.

Similar to the integer-order differential equations, it is usually difficult to obtain the analytical solution for fractional differential equations (FDEs) (e.g., [14, 15]), and one has to resort to numerical methods. Thus, there has been a growing interest to develop numerical approaches in solving FDEs, and many and varied methods have been considered, for example . In 2000, Podlubny  suggested a matrix form representation of discrete analogues of various forms of fractional differentiation and fractional integration. Based on the so-called triangular strip matrices, an approach was presented to significant simplification of the numerical solution of fractional integral and differential equations. Diethelm et al. [17, 18] discussed an Adams-type predictor-corrector method for the numerical solution of FDEs and gave a detailed error analysis for this algorithm, including error bounds under various types of assumptions on the equation and asymptotic expansions for the error. This Adams-type method is convergent with order at least one if the analytical solution y(t) is twice continuously differentiable. In 2007, Lin and Liu  developed a kind of linear multistep methods for fractional initial value problems based on Lubich’s high-order approximations  to fractional derivatives and integrals. And they proved the consistence, convergence, and stability of these methods. In 2006, Kumar and Agrawal  utilized the equivalent Volterra integral equation and extended a block-by-block method to some kinds of FDEs. Numerical examples have shown the efficiency and stability of this scheme. However, it is a pity that the error estimate and convergence order analysis of this scheme was neglected. Thus Huang et al.  derived error estimate and precise convergence order of the block-by-block method under certain assumptions and tested the order via numerical experiments. It is shown that this block-by-block method is convergent with order at least 3 for any fractional order index α>0. Li  derived a Chebyshev wavelet operational matrix of the fractional integration and used it to solve a nonlinear FDEs. In 2011, Scherer et al.  discussed finite difference schemes for the approximation of FDEs based on the Grünwald-Letnikov definition of fractional derivatives. Then, the asymptotic stability, the absolute stability, error representations, and estimates of the proposed explicit and implicit methods were obtained.

As for fractional partial differential equations (FPDEs), maybe the most famous model is the fractional diffusion equation used to describe the anomalous diffusion in porous medium. Further applications and numerical methods for fractional diffusion equation were discussed (e.g.,  and references therein).

It is well known that the key difficulty in numerically solving FDEs or FPDEs is to discretize the nonlocal fractional operators, due to the requirement of using the values of the numerical solution for all the previous times at which the solution is calculated. This makes numerical methods slow and hugely memory demanding when the number of nodes n is large. The problem is so acute that a great deal of effort has been devoted to overcoming it. In , Yuste and Quintana-Murillo presented an implicit finite difference method with nonuniform timesteps for discretizing Caputo derivative. This method allows one to build adaptive methods where the size of the timesteps is adjusted to the behavior of the solution. In this paper, a spectral deferred correction method for classical ordinary differential equations [31, 32] is extended and reconstructed to solve FDEs based on accelerating the convergence of lower-order schemes. For the new method, a relatively few Gauss-Legendre-Lobatto points are needed, and the high accuracy numerical solutions of FDEs can be easily obtained without a huge computational cost.

This paper is organized as follows. In Section 2, the equivalent Volterra integral equation is given; then a residual function and an error equation are defined. In Section 3, based on Gauss-Legendre-Lobatto points, a spectral approximation of the residual function is obtained. In Section 4, a fractional Adams method is used as a preconditioner to solve the error equation; then we describe the new spectral deferred correction method for FDEs. In Section 5, preliminary numerical experiments are given to verify the efficiency and accuracy of the proposed method.

2. Residual Function and Error Equation

In this paper, we discuss the numerical solution of the following FDE initial value problem (IVP): (1)D*αy(t)=f(t,y(t)),0tT,0<α1,y(0)=y0, where D*α is fractional Caputo derivative of order α and defined as (2)D*αy(t)=1Γ(1-α)0t(t-τ)-αdy(τ)dτdτ. Assume that the right-hand side function f(t,y) is continuous with respect to two variables t and y. Then, according to Lemma 1 of , IVP (1) is equivalent to the following Volterra integral equation of the second kind: (3)y(t)=y0+1Γ(α)0t(t-τ)α-1f(τ,y(τ))dτ. Obviously, if 0<α<1, the kernel in the integral of (3) is singular, and if α=1, then (3) becomes the classical Picard integral equation formulation.

Suppose now that an approximate solution y0(t) to (3) has been obtained by a low-order method. A measure of the quality of the approximation is given by the residual function (4)ε(t)=y0+1Γ(α)×0t(t-τ)α-1f(τ,y0(τ))dτ-y0(t). And the error can be defined by (5)δ(t)=y(t)-y0(t). However, in (5), y(t) is the exact solution of IVP (1), and we often cannot know it beforehand. Next, a relation between the residual function and the error will be deduced.

Substituting (5) into (3), (3) can be rewritten as (6)y0(t)+δ(t)=y0+1Γ(α)×0t(t-τ)α-1f(τ,y0(τ)+δ(τ))dτ; then after some algebraic manipulation, it becomes (7)δ(t)=1Γ(α)0t(t-τ)α-1×[f(τ,y0(τ)+δ(τ))-f(τ,y0(τ))]dτ+ε(t)1Γ(α)0t(t-τ)α-1G(τ,y0(τ),δ(τ))dτ+ε(t).

In (7), once the residual function is fixed, then a low-order method, such as fractional Adams method presented in , will be used to solve (7). In the following, we specify how the values ε(ti) are actually computed. For this, some stable and high-order accurate methods for interpolation and integration are required.

3. Spectral Approximation of the Residual Function

In this section, a spectral method is designed for obtaining high-order numerical approximation of the residual function.

Suppose that s0<s1<<sn are the n+1 Gauss-Legendre-Lobatto nodes on [-1,1], where s0=-1 and sn=1. Noticing that the interval of IVP (1) is [0,T], we can obtain n+1 Gauss nodes on [0,T] through the following linear transformation: (8)tk=T2sk+T2,k=0,1,,n.

Now, a low-order method is used to solve IVP (1) and get the initial approximation y0(t). Here, we take the following fractional explicit Euler method; namely, for k=0,1,,n-1, (9)yk+1=y0+j=0kbj,k+1f(tj,yj), where yk+1 is the approximations of solution function y(t) at nodes tk+1 (k=0,1,,n-1), the weight coefficients bj,k+1 are computed by (10)bj,k+1=1Γ(α+1)((tk+1-tj)α-(tk+1-tj+1)α),j=0,1,,k.

After obtaining the numerical solutions y1,y2,,yn, using the nodes t0,t1,,tn as the nodes of Lagrange interpolation for the initial approximation y(t) and the right-hand side function f(t,y(t)), we have (11)y0(t)=k=0nykLk(t),(12)f(t,y0(t))=k=0nfkLk(t), where Lk(t) is the Lagrange interpolation polynomial of order n, given as (13)Lk(t)=i=0,iknt-titk-ti,fk=f(tk,y0(tk)).

In specific computations, the fractional Riemann-Liouville integral of Lagrange interpolation polynomial is needed. To this end, Lk(t) should be transformed into following expressions: (14)Lk(t)=j=0ncj,ktj,k=0,1,,n, where the coefficients cj,k can be computed by its matrix form, namely, (15)(t00t01t0nt10t11t1ntn0tn1tnn)(c0,0c0,1c0,nc1,0c1,1c1,ncn,0cn,1cn,n)=In+1, here In+1 is the identity matrix of order n+1.

By (11), the fractional integral of Lk(t) is calculated as (16)1Γ(α)0t(t-τ)α-1Lk(τ)dτ=j=0ncj,k1Γ(α)0t(t-τ)α-1τjdτ=j=0ncj,kΓ(j+1)Γ(j+1+α)tj+α.

Substituting (11) and (12) into (4) and using (16), the residual function (4) is approximated as follows: (17)ε(t)=y0+1Γ(α)×0t(t-τ)α-1f(τ,y0(τ))dτ-y0(t)=y0+1Γ(α)×0t(t-τ)α-1k=0nfkLk(τ)dτ-y0(t)=y0+k=0nfkj=0ncj,k1Γ(α)×0t(t-τ)α-1τjdτ-y0(t)=y0+k=0nfkj=0ncj,kΓ(j+1)Γ(j+1+α)×tj+α-y0(t).

4. Spectral Deferred Correction Method

Once the residual function is computed by (17), the error equation (7) can be numerically solved by the fractional Adams method ; that is, for k=0,1,,n-1(18)δk+1p=ε(tk+1)+j=0kbj,k+1G(tj,y0(tj),δj),δk+1=ε(tk+1)+(j=0kaj,k+1G(tj,y0(tj),δj)+ak+1,k+1G(tk+1,y0(tk+1),δk+1p)j=0k), where δk+1p is the so-called predictor, and aj,k+1 are defined as (19)a0,k+1=ω0,k+1,ak+1,k+1=γk+1,k+1,aj,k+1=γj,k+1+ωj,k+1,j=1,2,,k, with (20)γj+1,k+1=1tj+1-tj×((tk+1-tj)α+1Γ(α+2)-(tk+1-tj+1)α+1Γ(α+2)+(tk+1-tj+1)α(tj-tj+1)Γ(α+1)(tk+1-tj)α+1Γ(α+2)-(tk+1-tj+1)α+1Γ(α+2)),ωj,k+1=1tj+1-tj×((tk+1-tj+1)α+1Γ(α+2)-(tk+1-tj)α+1Γ(α+2)+(tk+1-tj)α(tj+1-tj)Γ(α+1)(tk+1-tj+1)α+1Γ(α+2)-(tk+1-tj)α+1Γ(α+2)).

Let the initial approximation vector Y=(y1,y2,,yn)T, the error vector δ=(δ1,δ2,,δn)T; then the initial approximation solution is updated with (21)Y=Y+δ.

Based on the new solution Y, we can continue to compute the new residual function and the new error. And this procedure can be repeated until the high-accuracy solution is obtained. The pseudocode for spectral deferred correction method is given by the following.

Step 1 (initialization).

Set a small parameter etol>0. Use fractional explicit Euler method (9) to compute an initial solution yky(tk) at nodes t1,t2,,tn on the interval [0,T].

Step 2.

Based on the initial approximation yk, compute the approximate residual function εk  (k=1,2,,n) by using (17).

Step 3.

Use fractional Adams method (18) to solve the error equation and obtain the approximation δk  (k=1,2,,n).

Step 4.

Update the approximate solution yk=yk+δk   (k=1,2,,n). If δ<etol,  then stop, and the approximation yk is the solution we get finally. If δ>etol,  let yk=yk, and go to Step 2.

It can be checked that each correction procedure in this algorithm can improve the accuracy of the method, as long as such improvement has not gone beyond the degree of the underlying interpolating polynomial and the quadrature rules.

5. Numerical Examples

In this section, we verify the performance and high accuracy of the proposed method by following two examples. In these computations, the small parameter etol=1e-14. All codes are written in Matlab 2010b and run on a personal computer with Intel(R) Core(TM)2 Duo CPU P7350 processor 2.00 GHz, 2 GB memory.

Example 1.

Consider the following FDEs: (22)D*αy(t)=Γ(5+α)24t4+t8+2α-y2(t) with initial condition y(0)=0. The exact solution of this problem is given as (23)y(t)=t4+α.

In Table 1, we list the number of iterations denoted by “Iter.,” the CPU time in second “Time[s]” and the errors in the maximum norm for different nodes n and fractional order index α. It demonstrates that when the new spectral deferred correction method is used to solve Example 1, only a relatively few nodes are used, and the high-order accuracy numerical solution is obtained with small computation complexity and computing time. To illustrate this point, we use fractional block-by-block method  for comparison. To achieve the accuracy of 10−9, fractional block-by-block method needs about 320 nodes and 20.031599 seconds. Figure 1 shows that the error fast converge to machine precision, and it almost shows an exponential decay with increasing n.

The performance of spectral deferred correction method for solving Example 1.

n    α = 0.5 α = 0.8
Iter. Time (s) Error Iter. Time (s) Error
4 11 0.120460 1.684548 e - 09 9 0.099106 3.829423 e - 09
6 11 0.144606 7.243051 e - 11 9 0.120215 6.735190 e - 11
8 11 0.177994 2.727929 e - 12 10 0.148923 3.474332 e - 12
10 11 0.219745 9.281464 e - 14 10 0.181589 6.827872 e - 14

The errors for different n and α in Example 1.

Example 2.

Let one consider the following equations: (24)D*αy(t)=40320Γ(9-α)t8-α-3Γ(5+α/2)Γ(5-α/2)t4-α/2+94Γ(α+1)+(32tα/2-t4)3-[y(t)]3/2 subject to the initial condition y(0)=0. The exact solution is (25)y(t)=t8-3t4+α/2+94tα.

For numerical experiments of Example 2, compared with the Example 1, the similar numerical results can be obtained. However, from Table 2 and Figure 2, one should notice that the error is larger than that of Example 1’s, and its convergence speed is also relatively slow. We think the reason is that the exact solution in Example 1 is smoother than the solution in Example 2, because it is well known that the results of any numerical method often depend on the smoothness of a given problem.

The performance of spectral deferred correction method for solving Example 2.

n α = 0.5 α = 0.8
Iter. Time (s) Error Iter. Time (s) Error
5 16 0.383436 2.017025 e - 3 13 0.208751 2.177103 e - 3
7 15 0.443425 4.038208 e - 6 13 0.270258 1.418501 e - 6
9 14 0.496825 1.748804 e - 7 12 0.331811 3.051062 e - 7
11 15 0.655070 3.506529 e - 8 11 0.399406 5.716511 e - 8

The errors for different n and α in Example 2.

6. Conclusion

In this paper, a spectral deferred correction method is presented for fractional differential equations with initial value condition. Numerical experiments show it can obtain the high-order accuracy numerical solutions of FDEs without a huge computational cost caused by the nonlocality of fractional derivative. However, more efforts are needed to further the study such as detailed theoretical analysis and improve the algorithm for long-time simulation.

Acknowledgments

W. Zhao’s work was supported partially by National Natural Science Foundation of China (no. 11072120 and no. 11002075), and J. Zhu’s work was supported partially by Brazilian National Council for Scientific and Technological Development (CNPq).

Oldham K. B. Spanier J. The Fractional Calculus 1974 New York, NY, USA Academic Press MR0361633 Baleanu D. Diethelm K. Scalas E. Trujillo J. J. racrional Calculus Models and Numerical Methods 2012 Singapore World Scientific 10.1142/9789814355216 MR2894576 Miller K. S. Ross B. An Introduction to the Fractional Calculus and Fractional Differential Equations 1993 New York, NY, USA Wiley MR1219954 Samko S. G. Kilbas A. A. Marichev O. I. The Fractional Integrals and Derivatives: Theory and Applications 1993 Amsterdam, The Netherlands Gordon and Breach MR1347689 Podlubny I. Fractional Differential Equations: An Introduction to Fractional Derivatives, Fractional Differential Equations, to Methods of Their Solution and Some of Their Applications 1999 San Diego, Calif, USA Academic Press MR1658022 Diethelm K. The Analysis of Fractional Differential Equations 2010 Berlin, Germany Springer 10.1007/978-3-642-14574-2 MR2680847 Bagley R. L. Calico R. A. Fractional order state equations for the control of viscoelastically damped structures Journal of Guidance, Control, and Dynamics 1991 14 2 304 311 2-s2.0-0026124343 Bai J. Feng X. C. Fractional-order anisotropic diffusion for image denoising IEEE Transactions on Image Processing 2007 16 10 2492 2502 10.1109/TIP.2007.904971 MR2467780 Tarasov V. E. Fractional vector calculus and fractional Maxwell's equations Annals of Physics 2008 323 11 2756 2778 10.1016/j.aop.2008.04.005 MR2463217 Benson D. A. Wheatcraft S. W. Meerschaert M. M. Application of a fractional advection-dispersion equation Water Resources Research 2000 36 6 1403 1412 2-s2.0-0034032484 10.1029/2000WR900031 Metzler R. Klafter J. The random walk's guide to anomalous diffusion: a fractional dynamics approach Physics Reports 2000 339 1 1 77 10.1016/S0370-1573(00)00070-3 MR1809268 Baeumer B. Meerschaert M. M. Benson D. A. Wheatcraft S. W. Subordinated advection-dispersion equation for contaminant transport Water Resources Research 2001 37 6 1543 1550 2-s2.0-0035014974 10.1029/2000WR900409 Schumer R. Benson D. A. Meerschaert M. M. Baeumer B. Multiscaling fractional advection-dispersion equations and their solutions Water Resources Research 2003 39 1 1022 1032 2-s2.0-0037490225 Yuste S. B. Acedo L. Some exact results for the trapping of subdiffusive particles in one dimension Physica A 2004 336 3-4 334 346 2-s2.0-1642632896 10.1016/j.physa.2003.12.048 Kilbas A. A. Srivastava H. M. Trujillo J. J. Theory and Applications of Fractional Differential Equations 2006 Amsterdam, The Netherlands Elsevier MR2218073 Podlubny I. Matrix approach to discrete fractional calculus Fractional Calculus & Applied Analysis 2000 3 4 359 386 MR1806310 Diethelm K. Ford N. J. Freed A. D. A predictor-corrector approach for the numerical solution of fractional differential equations Nonlinear Dynamics 2002 29 1–4 3 22 10.1023/A:1016592219341 MR1926466 Diethelm K. Ford N. J. Freed A. D. Detailed error analysis for a fractional Adams method Numerical Algorithms 2004 36 1 31 52 10.1023/B:NUMA.0000027736.85078.be MR2063572 Lin R. Liu F. Fractional high order methods for the nonlinear fractional ordinary differential equation Nonlinear Analysis: Theory, Methods & Applications 2007 66 4 856 869 10.1016/j.na.2005.12.027 MR2288436 Lubich Ch. Discretized fractional calculus SIAM Journal on Mathematical Analysis 1986 17 3 704 719 10.1137/0517050 MR838249 Kumar P. Agrawal O. P. An approximate method for numerical solution of fractional differential equations Signal Processing 2006 86 10 2602 2610 2-s2.0-33745712076 10.1016/j.sigpro.2006.02.007 Huang J. F. Tang Y. F. Vázquez L. Convergence analysis of a block-by-block method for fractional differential equations Numerical Mathematics: Theory, Methods and Applications 2012 5 2 229 241 10.4208/nmtma.2012.m1038 MR2911875 Li Y. L. Solving a nonlinear fractional differential equation using Chebyshev wavelets Communications in Nonlinear Science and Numerical Simulation 2010 15 9 2284 2292 10.1016/j.cnsns.2009.09.020 MR2602712 Scherer R. Kalla S. L. Tang Y. F. Huang J. F. The Grünwald-Letnikov method for fractional differential equations Computers & Mathematics with Applications 2011 62 3 902 917 10.1016/j.camwa.2011.03.054 MR2824679 Ben Nakhi Y. Kalla S. L. Some boundary value problems of temperature fields in oil strata Applied Mathematics and Computation 2003 146 1 105 119 10.1016/S0096-3003(02)00529-5 MR2007773 Ervin V. J. Heuer N. Roop J. P. Numerical approximation of a time dependent, nonlinear, space-fractional diffusion equation SIAM Journal on Numerical Analysis 2007 45 2 572 591 10.1137/050642757 MR2300288 Liu F. Zhuang P. Burrage K. Numerical methods and analysis for a class of fractional advection-dispersion models Computers & Mathematics with Applications 2012 64 10 2990 3007 10.1016/j.camwa.2012.01.020 MR2989329 Tadjeran C. Meerschaert M. M. Scheffler H. P. A second-order accurate numerical approximation for the fractional diffusion equation Journal of Computational Physics 2006 213 1 205 213 10.1016/j.jcp.2005.08.008 MR2203439 Basu T. S. Wang H. A fast second-order finite difference method for space-fractional diffusion equations International Journal of Numerical Analysis and Modeling 2012 9 3 658 666 MR2926542 Yuste S. B. Quintana-Murillo J. A finite difference method with non-uniform timesteps for fractional diffusion equations Computer Physics Communications 2012 183 12 2594 2600 10.1016/j.cpc.2012.07.011 MR2970362 Dutt A. Greengard L. Rokhlin V. Spectral deferred correction methods for ordinary differential equations BIT Numerical Mathematics 2000 40 2 241 266 10.1023/A:1022338906936 MR1765736 Huang J. Jia J. Minion M. Accelerating the convergence of spectral deferred correction methods Journal of Computational Physics 2006 214 2 633 656 10.1016/j.jcp.2005.10.004 MR2216607 Diethelm K. Ford N. J. Analysis of fractional differential equations Journal of Mathematical Analysis and Applications 2002 265 2 229 248 10.1006/jmaa.2000.7194 MR1876137