I often run into applications where I need to build process models for minerals processing plants in AF. Minerals processing plants can easily be broken down into a number of plant areas connected by streams. Each of these streams normally has a mass flow and an assay which determines the concentration of the minerals within that stream. Normally I create a handful of element templates that contain the site's standard calculations and then map these to the relevant inputs (flows and assays) and outputs using PI point data references and the formula data reference. I thought I might explore what AF models have to offer since they sound like the ideal tool for this application.
I've built myself a small model made up of 3 model nodes connected by flows. So now what I do with it? What I would like to be able to do is access the upstream/downstream elements (model nodes) via the ports for my calculations (eg. plant area recovery). Is there a syntax for accessing an attribute connected to an element of a port? This will allow me to build a set of standard calculations and then simply build my plant using models connected by flows. From what I can tell, this is not possible with AF out of the box. Please correct me if I am missing something here. So this begs me to question, what is the point of the AF models? Are AF models simply a feature to be leveraged by third party products? Forgive my ignorance. I couldn't find much information in the help or by searching PI Square.