Unconditionally Optimal Error Analysis of Crank–Nicolson Galerkin FEMs for a Strongly Nonlinear Parabolic System

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

65 Scopus Citations
View graph of relations


  • Dongfang Li
  • Jilu Wang

Related Research Unit(s)


Original languageEnglish
Pages (from-to)892-915
Journal / PublicationJournal of Scientific Computing
Issue number2
Online published15 Feb 2017
Publication statusPublished - Aug 2017


In this paper, we present unconditionally optimal error estimates of linearized Crank–Nicolson Galerkin finite element methods for a strongly nonlinear parabolic system in Rd (d = 2, 3). However, all previous works required certain time-step conditions that were dependent on the spatial mesh size. In order to overcome several entitative difficulties caused by the strong nonlinearity of the system, the proof takes two steps. First, by using a temporal-spatial error splitting argument and a new technique, optimal 2 error estimates of the numerical schemes can be obtained under the condition τ h, where τ denotes the time-step size and h is the spatial mesh size. Second, we obtain the boundedness of numerical solutions by mathematical induction and inverse inequality when τ h. Then, optimal 2 and 1 error estimates are proved in a different way for such case. Numerical results are given to illustrate our theoretical analyses.

Research Area(s)

  • A strongly nonlinear parabolic system, Galerkin FEMs, Linearized Crank–Nicolson scheme, Unconditionally optimal error analysis