Skip to content

An extension of T-IFISS to diffusion problems with random inputs

License

Notifications You must be signed in to change notification settings

albespalov/Stochastic_T-IFISS

Repository files navigation

Stochastic T-IFISS

Stochastic T-IFISS extends the core version of T-IFISS to cover stochastic Galerkin approximations of diffusion problems with random coefficients, the associated a posteriori error estimation and adaptive algorithms, including goal-oriented adaptivity and multilevel adaptivity.

The adaptive algorithms for computing single-level stochastic Galerkin approximations are discussed in the following paper:

The goal-oriented error estimation strategy and the associated goal-oriented adaptive algorithm are presented in the following paper:

An a posteriori error estimator for multilevel stochastic Galerkin approximations and adaptive algorithms for computing multilevel approximations are discussed in this paper:

The MATLAB code in this repository provides the complete T-IFISS package, including the stochastic extension. The package is compatible with Windows, Linux, and MacOS computers. The toolbox has been tested using MATLAB version 9.1 and it is compatible with versions as far back as 6.5.

To run the software, download the file tifiss1.2_plus_stoch.zip. After unpacking the files, start MATLAB in the directory tifiss1.2. Type install_tifiss and follow the on-screen instructions. After installation, type setpath and then helpme to get started.

The routines implementing stochastic Galerkin approximations for problems with random inputs are included in the sub-directory stoch_diffusion. For help with setting up test problems, run helpme_stoch_diffusion.m.

Some sample sessions are saved in the following diary files:
stoch_diffusion_testresults_ex2.txt
stoch_adapt_testresults_ex5.txt
stoch_goafem_testresults_ex4.txt
stoch_multilevel_testresults_ex9.txt

To cite Stochastic T-IFISS in papers, please use:
Alex Bespalov and Leonardo Rocchi, Stochastic T-IFISS, January 2022. Available online at https://github.com/albespalov/Stochastic_T-IFISS/.