Multi-phase pipelines and pumps

[confidential]

Dear friend

I want to ask qusations about Multi-phase

i developed a computer programm to simulate a multi-phase
flow in inclined piplines

I used beggs and brill method for predicting
pressure loss and hold-up ?but is this method valid
for large diameters pipeline (20-40 inch)(trunk lines).

I suppose that the two phase pipeline system,is along distance large diameter pipeline this line transports the oil and gas mixture (coming from a gathering sytem which
transports the fluid of many wells).At the head of this line thers is a pumping station,in this station the
mixture is pumped into the trunkline and transported to processing facilities.?
does such a scheme exist in the fields?and if not how should i consider the scheme?

I want to ask about multi-phase pump, that will be used in
pumping station?is there any book or paper which deal with this issue(like horse power calculation ...etc).

if any one can provide me with a good explanation for these issues ,and a web sits that offers a free material
regarding these issues.

Your opinions would be very appreciated

Thanks

 

[zdas04]

There have been thousands of articles, theses', and technical papers on each of the issues you raise. This week's SPE ATCE in Houston had a half dozen examples of multi-phase stuff on the exhibit floor. I would start with the spe.org e-library if you're an SPE member.

The Beggs and Brill method is just one of dozens of excellent techniques for predicting pressure drop in multi-phase line. Their technique is generally pretty good in the range where they did their work. I don't beleive that the range includes 20 to 40-inch pipe.

Multi-phase flow is very tough to predict since each segment of the flow map has significantly different energy requirements to sustain flow and a real-life flow will jump from one segment to the next at pretty unpredictable times.

When I was working on my Masters in multi-phase flow I read dozens of learned papers that would go through really obscure math for many thousands of words and then say that they "have verified their results to within +/- 70%", you could do better by flipping a coin.

Good luck with your program, if it is intended to be a commercial product you need to understand that the marketplace is pretty saturated with good programs already and finding a niche will be difficult. Be prepared to meet with a lot of skepticism.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[DrillerNic]

(pasted from my reply to the same question in another thread)

Multiphase flow is a very large area.....Beggs & Brill's flow map is good for horizontal pipe (although it tends to overpredict liquid hold up). Other flow maps are good under different conditions: Dun & Ros is supposed to be better in high gas to liquid ratios; Hagedorn & Brown is supposed to be best in slug type flow etc. One problem with the flow map appraoch is that it is discontiuous: the equasions for mist flow apply until 'suddenly' the pipe is in slug flow for example. Also, most of these correlations were deveopled using water & air in small diameter clear plastic pipe....how applicable to oil & gas in steel pipe is anyones guess! Modern hydraulic simulators have mechanistic models which use flow mechanics (and lots of computation power!) such as the OLGA programme.

One observation: If you're bringing the multiphase flowlines from each well/ manifold to a pumping station and then into a single multiphase trunkline, why not put a large separator in front of the pump, separate out a lot of the gas and then have two smaller pipelines?

For Mutliphase Pumps, try talking to Framo Engineering:

http://www.framoeng.no/multiphase_booster_pumps.ht...

Also have a look in the SPE website (

http://www.spe.org)

-paper summaries are free, but you have to pay to see the full paper. There's been quite a lot of long multiphase tie back pipelines installed recently, so there will probably be some papers about operational experiences of these type of lines

 

[VinAAA]

For multiphase flow, slugging will normally be an issue. This is likely to affect your model. OLGA is one of the more acknowledged programmes that deals with this, and they base their predictions partly on theoretical calculations, and parlty on empirical data from flow loop testing. or so I am led to believe.

To accurately predict this, the topography of the pipeline will also be an issue.

Framoe, FMC kongsberg and ABB all will be happy to sell multiphase pumps, developed for wellstream service. Some experience has been had with such pumps in vastly different conditions, including several hundred meters below sea level.

To what extent they will be able to help you with theoretical pump sizing data is perhaps doubtful, but may be worth a try.