TY - JOUR
T1 - Finite difference/spectral-Galerkin method for a two-dimensional distributed-order time–space fractional reaction–diffusion equation
AU - Guo, Shimin
AU - Mei, Liquan
AU - Zhang, Zhengqiang
AU - Jiang, Yutao
N1 - Publisher Copyright:
© 2018 Elsevier Ltd
PY - 2018/11
Y1 - 2018/11
N2 - In this letter, we consider the numerical approximation of a two-dimensional distributed-order time–space fractional reaction–diffusion equation. The time- and space-fractional derivatives are considered in the senses of Caputo and Riesz, respectively. By using the composite mid-point quadrature, the original fractional problem is approximated by a multi-term time–space fractional differential equation. Then the multi-term Caputo fractional derivatives are discretized by the L2-1σ formula. We apply the Legendre–Galerkin spectral method for the spatial approximation. Two numerical experiments with smooth and non-smooth initial conditions, respectively, are performed to illustrate the robustness of the proposed method. The results show that: our scheme can arrive at the spectral accuracy (resp. algebraic accuracy) in space for the problem with smooth (resp. non-smooth) initial condition. For both of these two cases, our scheme can lead to the second-order accuracies in time. Additionally, the convergence rates in both spatial and temporal distributed-order variables are two.
AB - In this letter, we consider the numerical approximation of a two-dimensional distributed-order time–space fractional reaction–diffusion equation. The time- and space-fractional derivatives are considered in the senses of Caputo and Riesz, respectively. By using the composite mid-point quadrature, the original fractional problem is approximated by a multi-term time–space fractional differential equation. Then the multi-term Caputo fractional derivatives are discretized by the L2-1σ formula. We apply the Legendre–Galerkin spectral method for the spatial approximation. Two numerical experiments with smooth and non-smooth initial conditions, respectively, are performed to illustrate the robustness of the proposed method. The results show that: our scheme can arrive at the spectral accuracy (resp. algebraic accuracy) in space for the problem with smooth (resp. non-smooth) initial condition. For both of these two cases, our scheme can lead to the second-order accuracies in time. Additionally, the convergence rates in both spatial and temporal distributed-order variables are two.
KW - Distributed-order fractional derivative
KW - Finite difference
KW - Reaction–diffusion equation
KW - Spectral approximation
UR - https://www.scopus.com/pages/publications/85048763109
U2 - 10.1016/j.aml.2018.06.005
DO - 10.1016/j.aml.2018.06.005
M3 - 文章
AN - SCOPUS:85048763109
SN - 0893-9659
VL - 85
SP - 157
EP - 163
JO - Applied Mathematics Letters
JF - Applied Mathematics Letters
ER -