In cases where I've used the CAN controlled PVG actuators I assign node ID based on function. That seems the simplest way to me. That way I know the valve for Function A always has node ID X, and so on. In cases where they need to do a field replacement I've either come up with a programming routine that only works if there's one new valve plugged in at a time. If I remember correctly I did this by monitoring for the default node ID, and there was I think a way to determine if there are multiple new sections plugged in.
