Is there a manual on how to use Darwin Calibrator for finding possible locations for valve closure?
If you are using Darwin Calibrator, you may want to consider adding pipes are valves to the Status tab in the calibration study. A good reference for this is the sample file found at C:\Program Files (x86)\Bentley\WaterGEMS\Samples\Example5.wtg. This sample file includes a calibration study in Darwin Calibrator.
When you open Darwin Calibrator, select Calibration Study - 1 and look at the Status Elements tab. Notice that a pipe is included there, with an initial status of Open. Now run "Optimized Calibration - 1" and view the results in Solution 1. If you select the Status results, you will see that Darwin Calibrator changed the status of the pipe to Closed. That means that to get the best results, Darwin Calibrator required the pipe to be closed, which may be an indication that a valve on that pipe is closed in the field.
You may want to try something like that in your model too. Note that if you include a lot of pipes or valves in the Status Elements tab, you may need to increase the Maximum Trials value in the Options tab. See the Darwin Calibrator wiki for other tips.
When I look at the Example #5, hydrant flushing are set at junctions instead of hydrants.
If I set the Darwin calibration using hydrants, when it simulates multiple flow tests, would it open the hydrants instead of just adding the flow demand?
When I run manual calibration (Not using Darwin Calibrator) with a closed hydrant with a demand vs. an open hydrant without demand, the results were significantly different and I found the latter to be a correct setting. I want to make sure that the Darwin Calibration can run multiple hydrant tests as shown in the example by opening each hydrants.
Otherwise, I can run only one hydrant test at a time with a specific physical setting, yes? Or perhaps I need to create a junction with the same elevation as hydrants and run the analysis?
In Example5.wtg, all nodes happen to be junctions. If you have hydrants in your model that represent where the flow test is taking place, you can use these. If you open the hydrant to do the flow test, you can enter the flow at the hydrant in the Additional Demand tab for the flow test.
You can include more than one hydrant test (Example5.wtg has three at different times). You will just need to make sure you have the time and any boundary overrides included correctly. Darwin Calibrator is designed to use an EPS simulation as the representative run. It will use an EPS Snapshot (meaning that demand patterns are accurately accounted for) when it runs. The Darwin Calibrator wiki and the Help documentation should have further information on this.
In the Darwin Calibrator, it seems that flows are added as a demand - not by opening hydrants.
When I was analyzing prior to the Darwin Calibrator analysis, I had a steady state analysis with EPS Snapshot with (a) open hydrant(s). Not having similar flows were also part of calibration analysis when I use this method.
Here is my question.
Adding a hydrant flow as a demand vs. opening hydrants
Are both good methods for calibration?
If the hydrants do not produce as high flow as in the field, can I simply add the flows as demand?
The demand adjustment you enter can be the flow measured during the flow test. The demand adjustment would only be used during the calibration study, and would not need to be included in the representative scenario.