Modeling pump dicharge regardless of waterdemand

Hello Gashin,

You can model the pump as a Variable Speed Pump (VSP) with "VSP Type" as "Fixed Flow". This would ensure that there is a fixed flow discharge through the pump if that is what you are looking for.

Ideally though you should try and update the model with the latest data on demands and pressures in the network. If you have the latest demands a system head curve can help you determine the required pump curve for your model.

If this is not what you are looking for, please elaborate on your query.

Hope this helps.

Another idea to consider is to apply a multiplication factor to your base demands to raise them by an estimated percentage, until the flow matches. This could be applied globally or different for sets of nodes (for example based on zone or pattern type). To preserve the 3-year-old demands, first create a child demand alternative, assign it to a child scenario and make that new scenario current. Alternatively if you choose to use the demand adjustment calculation option, you would duplicate your current calculation options and use the new one with the demand adjustments in your new child scenario. See related articles:

Viewing and global editing a list of Demands

Modeling scenarios with different demand conditions

If you are trying to use measured pump flow data to get the model results to match the real system, this may be more of a calibration exercise, where you could use Darwin Calibrator to help adjust demands to get the pump flow to match the measurements. Care must be taken in this case though, see: Water Model Calibration Tips 

I'm not sure what you mean by subestimation and what you are trying to accomplish?

The way that you place the demands and input the pump curve the hydraulics (i.e. the model) determines what the pump discharge it.

Do you actually have a variable speed pump that you can control to provide a known flow?

Is there a tank in the service area? If not, you can end up with convergence problems. If the pump is set to 60 L/s but the demand is 70 L/s and there is no other source, the model struggles with how to deal with 60=70. "Imposing" a fixed demand is not a good approach. What is yuor goal?

Dear Tom and Jesse,

Indeed my objective is not to modify the demands (as the 1st option). there are 4 pumps and at the inlet of this pumping unit the flow is 1500 m3/h (demand is 1500m3/h in the pic hour of the day).
One of the pumps is different than other three smaller pumps and alone it can dispatch 1500 m3/h. When simulating, the model uns only that bigger pump but in the field all 4 pumps are in duty 24h/24h. 

So my aim in modelling is to make all 4 pumps in duty at the same time with relative participation regarding their capacity. 

How can I deal with this situation and model the same condition?

thank you in advance for your response. 

Gashin

Hello Gashin,

I think this is a case of modeling VSP's in parallel. VSP’s in parallel work on the principle of “lead” and “lag”. For a certain number of VSP’s in parallel, there will be one pump which functions as the “lead” VSP which first turns on to attempt to meet the target pressure or flow.

However, in this case your lead pump is the one single pump which is capable of supplying 1500 m3/hr. It would satisfy the target flow itself and as a result the "lag" pumps would not be triggered. 

If you still wish to run all 4 pumps you could setup the lead pump for a lower target flow. So although it can satisfy the demands downstream it wont be able to due to the limit enforced. This could then activate the 3 smaller pumps based on how you set them. See the below article which discusses VSP's in parallel in detail;

Modeling variable speed pumps (VSP’s) in parallel

Hope this helps.