Surprised to see additional Part-Surfaces in your geometry displayer wondering where this came from. Don't panic! this article is intended to clear your doubts in this regard.
I will try to explain my thoughts through the means of a sample simulation. Let us assume you have multiple parts, some of these parts are split into additional part-surfaces which you need for specifying certain boundary conditions and some of them contain only a single part-surface. It is important to note that some of these parts are contacting their counterparts and we will create strong contacts using the imprint operation which will later become interfaces and are important for exchanging specific fluxes with each other.
A rectangular part named "Large" composing of a single part-surface has the six enclosing surfaces, there are other parts namely "Block" & "Outer" have some of the part surfaces organised into individual part-surfaces. The below screenshot gives you an overview of the all the parts in this simulation.
When you create a new Displayer and add a new Geometry/Surface displayer, you will see for the part "Large" only one part surface named "Block Surface". (See PartsBasedReporting_Start.sim)
Now, let us go ahead and create the various imprints required and create the various interfaces required. I have set-up the Operations for you including the imprint and Mesh Operations, go ahead and run this (should run a couple of minutes). Ensure that the newly created contacts are set as Interfaces. (Sim-Tree: Geometry > Contacts > Block/Large(Contact of interest) > Set Interface)
Now, when you go ahead and look at the Surface Displayer, you should now see an additional node under the Block Surface and you will have new surfaces (Magic! not quite).
The additional part surface you see are generated by STAR-CCM+ automatically, technically no new part surface has been generated rather the individual faces redistributed/sorted based on the contacting surface to its neighbours. Hence, you see four sub-surfaces, three are the individual contacts and the fourth is the rest of the remaining faces which is not associated with any contact[rem]. (See PartsBasedReporting@Meshed.sim)
The change helps you in quickly using the said sub-surface for calculating the energy transfer between the two parent parts through a report. The changes are intended mainly for reporting and post-processing purposes. If you are someone who do not like the naming convention or want to have more control with the naming of the individual part surfaces, you could either go ahead and name the contacting surface according to your needs or you can ask the imprint operation to create a new part surface for the resulting contacting surface.
Hope the article help clear some of the intrigue regarding where these additional ghost surfaces came from.