Reading the initial conditions from one or more existing results files ("Initial Values Files") will often provide a flexible way to initialize a run and to provide initial conditions. See Initial Condition Modeling for some examples of circumstances when this may be useful. Most simulations will require the use of only one Initial Values File.

When you read initial conditions for a run from an existing results file, then there are various controls that affect how the results in the results file are used to provide the necessary initial conditions. These are described in the sections Continuing the History, Using Multiple Files to Provide Initial Conditions, Using the Mesh from the Initial Values File, and Interpolation Mapping.

If the Initial Values File contains particle data from a Lagrangian Particle Tracking Simulation, then some limitations apply. See Using an Initial Values File that Contains Particles for details.

Sometimes the new simulation will require the initialization of a variable that is not present in the Initial Values File. Wherever there is a variable missing from the Initial Values File(s), the CFX-Solver will fall back to the initial values specified in CFX-Pre on the Initialization panel - it will use the specified value if the Automatic with Value option was chosen in CFX-Pre, or a solver-calculated value if the Automatic option was chosen. For example, the Initial Values File might contain the results of a simulation with just a single fluid "Water", but the new simulation might have two fluids "Water" and "Air" in an inhomogeneous multiphase set-up. In this case, the CFX-Solver will initialize all the variables it can from the Initial Values File, including variables such as Pressure and Water.Velocity. The initial conditions for the remaining variables, including Water.Volume Fraction, Air.Volume Fraction and Air.Velocity (which are not present in the Initial Values Files), will all be read from the CFX-Pre settings.

When multiple fluids (or components of fluids) are present in the Initial Values File(s) and/or the new CFX-Solver Input file, especially if more than one Initial Values File is used, the logic used to decide which variables in the Initial Values File(s) map to which variables in the CFX-Solver Input file may not be obvious. This is described in detail in Using the CFX-Interpolator.

The specification of an Initial Values Files is part of the Run Definition rather than part of the physics set-up. An Initial Values File can be specified in any of the following locations;

Unless the setting Use Mesh From (described in Using the Mesh from the Initial Values File is set to Initial Values, then behind-the-scenes the CFX-Interpolator will be called at the start of any run using an Initial Values File in order to extract the appropriate initial values, and you will see a section in the CFX-Solver Output file headed

 +--------------------------------------------------------------------+
 |                                                                    |
 |                  Interpolation of Initial Values                   |
 |                                                                    |
 +--------------------------------------------------------------------+

You should check this carefully to ensure that the diagnostic information presented represents what you might expect given your new run settings, your existing results file and any settings made to control how the initial values are used. Details on how the CFX-Interpolator works are given in Using the CFX-Interpolator.

Visualization of the initial values used (typically using CFD-Post) is recommended for any situation where the initial conditions are important (for example, for a transient run) and you are not sure what to expect, or where the diagnostic output from the CFX-Interpolator is unexpected or reports a problem. The best way to visualize the initial conditions actually used by the CFX-Solver is to view the transient or backup file written before the first timestep or outer iteration of the new run begins. The backup file at start parameter controls whether a backup file is written at the start of every run.