Initial Tank Levels

This case has been marked as "Will not implement", but really we are planning to solve this while providing a better connection with the underlying sensor data that drives these varying initial tank levels.

I would think this would be much easier to do if InfoWater used a real database as opposed to lots of dbf files. For me this is just another reason to improve the InfoWater database.

Interesting response. I know that WaterGEMS is capable of assigning initial tank levels independent of the tank geometry set. And I know you can export a WaterGEMS model to EPANET, so I wonder how they deal with the EPANET export issue.

This has always been a bit of a challenge to users and would be a nice change if doable. This would eliminate the need to make most "Tank Sets" due solely to the initial tank level. The challenge to making the change is likely how to keep models with data stored in the old way (the tank set) to still work and now store the new data in a new table as any change would have to be backward compatible to old functionality. Unfortunately since EPANET cannot give an initial status this way there is no way to assign an initial status for a tank level without potentially confusing users when they look at the model in EPANET, but if there was away to do this and I would include the data with the control set as that is where initial status data is stored this would be the best dataset to use for it. The element modeling/hydraulic tables generally cover the modeling data that can be scenario specific and likely the reason this was put into the tank hydraulic/modeling tables, but like you in my 16 years of consulting I really wished this was part of the control set rather than the tank set. But given that the data is not incorporated as an initial status in EPANET (an unfortunate circumstance in EPANET) this almost had to be incorporated in the hydraulic.modeling data for the tanks even though it was generally the only value that changed for most tank sets. If there is a way to adjust this though it would be very useful, but given the way the data is stored and the need for backward compatibility it may unfortunately be difficult or not feasible to make this change.