channelflow.org

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
                                                                                               
findorbit:                                                                                                                                                                
        Newton-Krylov-hookstep search for (relative) equilibrium or periodic orbit                                                                                                

options : 
  -eqb      --equilibrium                                          search for equilibrium or relative equilibrium (trav wave)
  -orb      --periodicorbit                                        search for periodic orbit or relative periodic orbit      
  -rel      --relative                                             search for relative periodic orbit or relative equilibrium
  -T        --maptime           <real>      default == 10          initial guess for orbit period or time of eqb/reqb map f^T(u)
  -es       --epsSearch         <real>      default == 1e-13       stop search if L2Norm(s f^T(u) - u) < epsEQB                 
  -ek       --epsKrylov         <real>      default == 1e-14       min. condition # of Krylov vectors                           
  -edu      --epsDuLinear       <real>      default == 1e-07       relative size of du to u in linearization                    
  -edt      --epsDtLinear       <real>      default == 1e-07       size of dT in linearization of f^T about T                   
  -eg       --epsGMRES          <real>      default == 0.001       stop GMRES iteration when Ax=b residual is < this            
  -egf      --epsGMRESfinal     <real>      default == 0.05        accept final GMRES iterate if residual is < this             
  -cd       --centerdiff                                           centered differencing to estimate differentials              
  -Nn       --Nnewton           <int>       default == 20          max number of Newton steps                                   
  -Ng       --Ngmres            <int>       default == 40          max number of GMRES iterations per restart                   
  -Nh       --Nhook             <int>       default == 20          max number of hookstep iterations per Newton                 
  -d        --delta             <real>      default == 0.01        initial radius of trust region                               
  -dmin     --deltaMin          <real>      default == 1e-12       stop if radius of trust region gets this small
  -dmax     --deltaMax          <real>      default == 0.1         maximum radius of trust region
  -df       --deltaFuzz         <real>      default == 0.01        accept steps within (1+/-deltaFuzz)*delta
  -lmin     --lambdaMin         <real>      default == 0.2         minimum delta shrink rate
  -lmax     --lambdaMax         <real>      default == 1.5         maximum delta expansion rate
  -irq      --improveReq        <real>      default == 0.001       reduce delta and recompute hookstep if improvement  is worse than this fraction of what we'd expect from gradient
  -iok      --improveOk         <real>      default == 0.1         accept step and keep same delta if improvement is better than this fraction of quadratic model
  -igd      --improveGood       <real>      default == 0.75        accept step and increase delta if improvement is better than this fraction of quadratic model
  -iac      --improveAcc        <real>      default == 0.1         recompute hookstep with larger trust region if improvement is within this fraction quadratic prediction.
  -Tsc      --Tscale            <real>      default == 20          scale time in hookstep: |T| = Ts*|u|/L2Norm(du/dt)
  -tc       --dudtConstr        <bool>      default == true        require orthogonality of step to dudt
  -os       --orthoScale        <real>      default == 1           rescale orthogonality constraint
  -sf       --symmfile          <string>                           file containing T,su,sx,sy,sz,ax,az
  -sx       --x-sign                                               change u,x sign
  -sy       --y-sign                                               change v,y sign
  -sz       --z-sign                                               change w,z sign
  -ax       --axshift           <real>      default == 0           translate by ax*Lx
  -az       --azshift           <real>      default == 0           translate by az*Lz
  -a        --anti                                                 antisymmetry instead of symmetry
  -rs       --relativeScale     <real>      default == 1           scale relative-search variables by this factor
  -xc       --dudxConstraint    <bool>      default == true        require orthogonality of step to dudx and dudz
  -b        --l2basis                                              use an explicit L2 basis
  -s1       --s1                                                   restrict to s1-symmetric subspace
  -s2       --s2                                                   restrict to s2-symmetric subspace
  -s3       --s3                                                   restrict to s3-symmetric subspace
  -R        --Reynolds          <real>      default == 400         Reynolds number
  -vdt      --variableDt                                           adjust dt to keep CFLmin<=CFL<CFLmax
  -dt       --timestep          <real>      default == 0.03125     (initial) integration timestep
  -dtmin    --dtmin             <real>      default == 0.001       minimum timestep
  -dtmax    --dtmax             <real>      default == 0.04        maximum timestep
  -dTCFL    --dTCFL             <real>      default == 1           check CFL # at this interval
  -CFLmin   --CFLmin            <real>      default == 0.4         minimum CFL number
  -CFLmax   --CFLmax            <real>      default == 0.6         maximum CFL number
  -ts       --timestepping      <string>    default == SBDF3       timestep algorithm: See README.
  -nl       --nonlinearity      <string>    default == Rotational  nonlinearity method: See README.
  -cp       --coutprec          <int>       default == 6           precison for std output
  -p        --pausing                                              pause between Newton steps
  -o        --outdir            <string>    default == ./          output directory
  <flowfield>      (trailing arg 1)                                initial guess for Newton search
channelflow.org

User Tools

Site Tools

Page Tools
