Depends on what you need it for. Most multi phase flow correlations (including the much-vaunted OLGA) have so many invalid assumptions that they'll never provide repeatable results in any flow. At best they can give you an indication of an order of magnitude. At worst they fail to reflect real flows by an order of magnitude. My best advice is that if you have to have something and you have to have it now then the best thing to do is purchase the cheapest program you can find--it will be wrong, but no more so than the expensive programs.
Multi-phase flow is so replete with random flow-responses to unpredictable events that typically the performance is worse than flipping a coin. Write a program some day to predict the next throw of a pair of six-sided dice. If you do better than the odds in Vegas, then your randomizing algorithm is busted. Same with multi-phase flow, you simply never know when and at what frequency your real flow will move around the flow-regime map.
**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)
OK, just contributing to the list. I use it for LPG distribution systems where, true, diameters are usually 50 to 150 mm.
**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)
Well, at the end of the day we still need to design two phase pipes, and the following methods are used daily in the petrochemical industry:
- Beggs+Brill
Beggs+Brill was developed for small pipe size water air systems - I dont think thats extremely valid for HC systems.
Regarding the ramdom thing - well at the contrary to Vegas doing a lot of tosses of the coin and then establishing your design margins is not so foolish imho.
Morten,
I agree that we have to design piping for multi-phase flow, we always have. My problem is that we pay a bunch of money for something like OLGA to burn and chug over the problem and then feel like we can reduce safety margins because of our "state of the art" software. I've seen slug catchers eliminated from designs and pigging equipment dropped because the FEA models predicted a slug size that was manageable without slug catchers or pigging. This is a bad use of technology.
Random is still random. The last toss of the dice does not influence the next toss. Duckler and Flannigan are pretty easy to implement and the millions of miles of pipe that have been designed using these "simplistic" approaches are every bit as "valid" as a pipe designed using a mechinistic finite element model, but the report is not nearly as pretty.
The gas/oil/water ratios change so much over the life of a field that what they "do do" right isnt't valid for a long time anyway.
**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)
1) The best of two phase flow calculation still always 20% error from the real situation in specified operation range. Then if you have so wide operate envelop error could be 40-60%..
2) Many formular is good only for horizontal pipe not vertical pipe
3) Liquid/Gas ratio & its densidty in each pipe section will change when temperature/pressure change, so if Pi of each section change over 15%, you will get the mess, so try to make section more detail
4) always check /calculate new thermodynamic to create the correct fluid property
5)All commercial software has its own defualt, we must put correlated factor after calculation.
6) slug management require proper flow assurance full scope review a long the field life ( steady,dynamic,corrosion management,etc)
Well on the other hand ive seen some (a lot?) of project that wouldnt have been developed (and has worked fine) if we had insisted on giat slug catchers off-shore.
That is also true, but sometimes getting rid of the safety factors is actually unsafe. It can be a close thing, and I just keep seeing people taking Program output to 9 decimal places and betting the farm on it. Often the bet pays off, sometimes it doesn't.
After some extensive research and keeping in mind the correlations available in Hysis pipe segment calcs, it's clear that the best correlation to use for horizontal pipe is Gregory, Aziz and Mandhame. For horizontal pipe or angled pipe, Beggs and Brill is good too. For vertical pipe, Orkiszewski is the best.
If I had a licence for OLGAS2000, I would use it for 2 phase calcs.
