Hi.
I have a basic question w.r.t to modelling.
I am having a DMA with 2 inlets.
How do I model the flows through these inlets?
I am trying using FCV at D/s of Both the reservoirs and giving them the known volume/day value.
Is this a correct practice?
If I try to model without FCVs, then only single inlet will supply the entire flow, which isn't correct.
Please suggest the correct method to model multiple inlets for a DMA.
Thanks.
What exactly are you trying to accomplish with this model?
If you have a model of the entire system and the DMA is a small part of it, then you don't need to do anything special. The model will calculate the flow into the DMA and pressures in the DMA.
If it is fed through some kind of control valves (e.g. PRV), you need to model that valve correctly. I doubt that an FCV would be the correct model element. Only use an FCV if you have a valve that behaves like the FCV element. See blog post on this topic below.
blog.virtuosity.com/death-to-flow-control-valves
Using a Flow Control Valve (FCV)
Hello Chetan,
Ideally a DMA should have a single point of source / connection which can be monitored / metered. However, in your case if there are multiple legitimate connections using an FCV for one of them should suffice (for flow control). The balance flow should be consumed from the other reservoir.
However, it is difficult to keep track of water in such cases as multiple inlets can cause bigger problems if the system has existing leakages and ultimately increase NRW.
Regards,
Yashodhan Joshi
Thanks for the reply Tom.
I am having a single DMA with me. This DMA has 2 inlets. One inlet is at higher elevation, whereas the other is at the lower elevation. Also, the HGL difference is such that, If I do not put any control valve, then complete flow is being taken up from Inlet-1.
I am trying to model this DMA. For this I am trying to see if putting an FCV and assigning it with the flow pattern mutliplier is the way to model the DMA.
Thanks for the link. I went through it completely, and found myself at Point 4.
That's what I am doing.
I guess it's a bad modelling practice.
But I will have to find another way out of it. In order to get the better results.
Thanks Yashodhan.