For and End For
From the Programming submenu for solver sequences, you can select For to add two nodes to the end of the sequence, a For node () and an End For node (). You can then move these nodes where you want to create a for loop that iterates some part of the solver sequence. When the sequence is run, the for loop runs the nodes between the For and End For nodes a fixed number of steps. You can add more than one For node to create nested for loops. The model tree displays the loop structure by indenting the description text for the nodes. The for loops must be balanced; otherwise an error occurs. For loops can be useful, for example, for solving particle-field interactions in particle tracing by iterating between a stationary and a time-dependent solver.
The End For node has no settings. The Settings window for the For node contains the following setting:
General
Select an option from the Defined by study step list. The default is User defined. If any instances of the Bidirectionally Coupled Particle Tracing or Bidirectionally Coupled Ray Tracing study step are present in the study then they can also be selected from the list.
Select an option from the Termination method list: Fixed number of iterations (the default) or Convergence of global variable. For Fixed number of iterations enter a positive integer for the Number of iterations (default 5). Any nodes between this For node and the following End For node will be repeated a number of times equal to the specified Number of iterations.
For Convergence of global variable enter the following:
Global variable (default 1)
Relative tolerance (default 0.001)
Maximum number of iterations (positive integer, default 25)
Minimum number of iterations (positive integer, default 1)
At the end of each iteration, the relative error is
where g0 is the value of the global variable from the previous iteration, g1 is the value of the global variable from the current iteration, and gt is the Relative tolerance threshold. A new iteration of the loop is started if the iteration number is less than the Maximum number of iterations, and if either the relative error is less than the Relative tolerance or the iteration number is less than the Minimum number of iterations.