SWMM 5.2 included the ability to define named variables and math expressions within control rules, is this something we may also want to include in EPANET?
I'm not overly familiar with the codebase of SWMM or SWMM modelling so I can't comment on the difficulties of moving the code across but I can see the benefit of them being included in EPANET.
The most major use case would be rules where you look at an average of multiple sensors or a net flow across many meters.
SWMM 5.2 included the ability to define named variables and math expressions within control rules, is this something we may also want to include in EPANET?
I'm not overly familiar with the codebase of SWMM or SWMM modelling so I can't comment on the difficulties of moving the code across but I can see the benefit of them being included in EPANET.
The most major use case would be rules where you look at an average of multiple sensors or a net flow across many meters.