tem_convergence_load Subroutine

public subroutine tem_convergence_load(me, conf, parent, steady_state)

Load the convergence definition table The convergence object must be part of a convergence object, for which the format has been set to format = 'convergence' In the convergence table, you then must define a norm:

• simple: just check against the state value of the last check, and reach convergence if below the defined threshold
• average: build the average over a defined set of last checks with nvals stop, if the difference to the current state value is below the given threshold
• nvals: define, how many last checks should be taken into account for averaging procedure

The error is by default calculated to be a relative error. If an absolute error is desired, choose absolute=true in the convergence object

The stopping criterion is defined as a general condition object, where the threshold and the operator has to be given

  condition = { threshold = 1.E-6, operator = '<' }

A sample convergence object with a convergence definition can look as follows (within time_control table):

  abort_criteria = {
   stop_file = 'stop',
   steady_state = true,
   convergence = {
     variable = {'pressure','velocity'},
     shape = {kind = 'all'},
     time_control = {
       min = {iter=0},
       max = {iter=tmax},
       interval = {iter=10*dt}},
     reduction = {'average','average'},
     norm='average', nvals = 100, absolute = true,
     condition = {
        { threshold = 1.e-15, operator = '<=' },
        { threshold = 1.e-12, operator = '<=' }
     }
   }
  }

Or another sample:

  abort_criteria = {
    stop_file     = 'stop',
    steady_state  = true,
    convergence   = {
      variable = {'pressure_phy'},
      shape = {
        {kind = 'canoND', object = {origin ={0.15-dx,0.2,zpos} }},
        {kind = 'canoND', object = {origin ={0.25+dx,0.2,zpos} }}
      },
      time_control = {min = 0, max = tmax, interval = 10*dt},
      reduction = {'average'},
      norm      = 'average',
      nvals     = 50,
      absolute  = true,
      condition = { threshold = 1.e-10, operator = '<=' }
    }
  }

Arguments