Similar to Jeremy Tinkler's Jeremy Tinkler's conundrum, I have an issue with circular references.
I need to model an electricity distribution network with a number of generators, busbars, cable runs, spurs, switches and breakers. Depending on the positions of the qwitches abd breakers and which generators are generating, the power could be energizing these components and cables from either direction. I need to calculate which components are live.so to figure out if B is live, I need to check both A and C. But to see if C is live, I need to check B and D.So B depends on C and C depends on B and we have a circular calculation.
I did think we could do this by modelling each component as having separate wires for each possible path with corresponding analyses looking 'upstream' on that path. Then a rollup for the overall component. I believe this would avoid circular references at the cost of lots more tags and analyses.
Is this a viable solution? How have others modelled such networks?