Practical
Insight into
Multiphase Now Matering
MPFM’s
can contribute significantly to cost savings if they are used as a replacement
for conventional test separators, elaborate Lex (A.M.) Scheers, Khamis
Busaidi, Mike Harper, Martin Halvorsen and Tor Wideroe.
The
FlowSys concept is discussed in more detail and a summary of the results
obtained in 3 test programs will be presented.
FlowSys
MPFM concept.
The major parts of the FlowSys TopFlow meter are the Venturi and the
electrodes incorporated inside the throat of the Venturi insert. The
flow rates of oil, water and gas are calculated based on the measurements
obtained by the electrodes and the measurement of the differential pressure
across the Venturi inlet. No separation devices, flow conditioners,
mixers, by-pass lines or radioactive sources are used in the TopFlow
concept.
The DP is measured across the inlet and throat of the Venturi insert.
The electrodes inside the Venturi throat measure the capacitance or
conductance of the mixture flowing through the Venturi insert. The velocity
is found from cross-correlating the high-resolution time signals from
pairs of electrodes within the Venturi insert.
As there is no gamma densitometer, the FlowSys meter does not measure
the fluid mixture density directly, unlike most other meters. However,
the density is determined indirectly, through the momentum equation
of the Venturi.Note
that capacitance and oil permittivity (eoil) apply for oil-continuous
mixtures while conductance and water conductivity (swater) applies for
water-continuous mixtures. The capacitance or conductance, DP and velocity
are the input measurements and are being measured by the electrodes
and DP transmitter. The permittivity of the oil (eoil), water conductivity
(swater) and densities of oil, water and gas (ro, rw, rg) are input
parameters entered into the user interface.
The cross sectional
area of the pipe (A) is known based on the geometry of the Venturi,
the velocity (V) is measured by cross-correlating the electrical signals
from the electrodes and the fractions of oil, water and gas are found
from the set of three equations for the fractions, as indicated earlier.
Test
Results
Earlier NEL and CMR test on the FlowSys meter showed very good results
(see figure 3 and 4) and based on those test results two Shell operating
companies, Petroleum Development Oman (PDO) and Shell Gabon, together
with Shell Global Solution International (SGSI) have decided to further
field trial the meter. Main driver for this extensive testing was the
promising technical concept, the small and compact configuration of
the meter, the fact that the meter doesn’t require a radioactive
source and the potential for sub-sea and downhole applications. The
Shell Gabon installation was at a production gathering station where
the meter was installed is series with a test separator and a test tank.
The PDO testing was done in a testloop in Daqing (China) where live
crude oil, actual production water and natural gas could be used.
*Shell
Gabon Experience
(Rabi-A Production Station, May 2001)
A 4” FlowSys prototype meter was tested at a gathering station
in series with a two-phase test separator and a test tank. Extensive
work was carried out to prepare and calibrate the test separator and
its associated measurement equipment, this to ensure proper reference
measurements. The Gas Volume Fractions (GVF) on some of the Rabi wells
are not only very high but also fluctuate a lot due to long horizontal
pipelines between well heads and gathering station. These long horizontal
pipelines also generate severe slugging and fluctuating water cut, which
is a challenge for any multiphase meter. The operating pressure for
the multiphase meter at the gathering station was about 10-12 bar.
Prior to the start of the Shell test campaign, FlowSys discovered that
the meter requires a minimum liquid flow rate to operate accurately,
this is because with upward flow through the MPFM liquid might fall
back at the low velocities. This discovery resulted in a reduction in
the operating envelope of the meter and unfortunately more of the Rabi
wells fell below this limit. Consequently not all wells could be tested,
however, the wells whose rates were within the operating envelope and
with sufficiently low GVF’s (<90%), the FlowSys meter could
measure the liquid within plus or minus 10%. For the higher GVF’s
the deviation between the references and the FlowSys meter were higher.
For the water cut in particular it was seen that also by comparing the
two available references (well head samples and tank dipping) a ±10%
difference were also seen between the references (see figure 5).
One has to bear in mind that this was a prototype and that further developments
have been carried out on the software and hardware after this test program.
In general, it should be noted that MPFM’s work better at the lower
or moderate GVF’s and this calls for installation upstream of the
choke. However, this again requires MPFM installation per well, which
again is only feasible with low-cost MPFM’s.
*SGSI/PDO
Experience
(Done in DOD Test Loop in China, Oct 2001)
The same 4" FlowSys meter was tested at DOD during September/October
2001. A 4" meter is a bit large when considering the operating
envelope of the DOD test facility. Consequently only liquid flow rates
in the low range of the operating envelope of the 4" meter were
tested. The test consisted primarily of low liquid flow rates and GVF’s
of 60% and above (see figure 6). The operating pressure at the DOD test
facility was low (2-5 bar). The trial at DOD revealed a need for further
development of the software models especially for high water cuts and
in the transition between oil and water-continuous flow.
