1. 09 Jan, 2017 2 commits
  2. 04 Jan, 2017 5 commits
  3. 03 Jan, 2017 2 commits
  4. 30 Dec, 2016 1 commit
  5. 29 Dec, 2016 3 commits
  6. 20 Dec, 2016 3 commits
    • 's avatar
      fix tests · 140c69c2
    • 's avatar
      merge master · 0f115a1b
    • Dave Moxey's avatar
      Merge branch 'feature/meshing_tools' into 'master' · 32a754b8
      Dave Moxey authored
      Feature/meshing release
      This merge request is the massive clean up, reformatting and documenting (WIP) of the mesh generation routines.
      This is the release candidate we can iterate on with smaller branches in the future.
      See merge request !704
  7. 19 Dec, 2016 12 commits
  8. 15 Dec, 2016 3 commits
  9. 14 Dec, 2016 1 commit
  10. 13 Dec, 2016 3 commits
  11. 09 Dec, 2016 5 commits
    • Michael Turner's avatar
      some changes not working · 5833f3ed
      Michael Turner authored
    • Chris Cantwell's avatar
      Merge branch 'feature/NewHOBCs' into 'master' · 8a902627
      Chris Cantwell authored
      Added in a new convective like outflow boundary condition as well as a velocity correction scheme using a weak pressure Poisson equations. 
      To complement this I have also tried to restructure the description in the user guide to use the Guermond and Shen approach to motivate both the standard velocity correction scheme and the weak pressure form.  
      See merge request !713
    • Chris Cantwell's avatar
      Merge branch 'master' into feature/NewHOBCs · 3e801554
      Chris Cantwell authored
    • Chris Cantwell's avatar
      Merge branch 'fix/Stability3D' into 'master' · f57dfa9d
      Chris Cantwell authored
      Tidy linearised advection and fix full 3D
      This MR simplifies the logic of `LinearisedAdvection::v_Advect`, reducing the dependency on the dimension and handling all components in a single loop. In doing so, I think this fixes the full 3D case, which was not working. Also, the gradient of the base flow is now computed in a separate function and stored in `m_gradBase`, avoiding the need to recompute this every time-step for steady base flows.
      The adjoint advection is now derived from LinearisedAdvection, eliminating a lot of duplicate code. It only needs to redefine v_Advect, which now follows a similar structure to the new implementation from LinearisedAdvection.
      See merge request !708
    • Chris Cantwell's avatar
      Merge branch 'master' into fix/Stability3D · 8d010c7f
      Chris Cantwell authored