When you use the 'Create' -> 'New Data Flow' in the Directory menu option to create a new data flow, the data flow is given a unique ID by the system. Similarly, when you use the 'Import -> Legacy Data Flow' menu option to import a Lavastorm .brg file, the imported graph is given a unique ID by the system.
If you use the 'Save as...' menu within a data flow edit session you are creating a new document so it is given a new ID.
When you use the 'Export...' option in the Properties panel to export a data flow, the data flow is exported and the data flow's ID is included in the .lna file. When you then use the Import' -> 'Data Flows or Nodes' Directory menu option, the data flow is imported into the specified folder. The imported data flow retains it's original ID. You cannot have two documents with the same name or same ID within the same folder.