MULTIPHASE METER TO UNDERGO NORTH SEA TESTS

April 17, 1995
Production and pipeline economics are driving development of multiphase pumps and meters aimed at reducing capital and operating costs for marginal fields, mostly subsea. One such meter will undergo subsea testing offshore Europe this summer and is a refinement of an earlier model currently in use in Oman. The North Sea field site for the testing has yet to be determined.

One such meter will undergo subsea testing offshore Europe this summer and is a refinement of an earlier model currently in use in Oman. The North Sea field site for the testing has yet to be determined.

Kongsberg Offshore A.S., Kongsberg, Norway, has developed and is conducting field tests on its KOS MCF (multi-component flow) 351 multiphase meter that will handle both gas and oil-external emulsion (that is, water droplets in oil) as well as gas and water-external emulsions (oil droplets in water) up to 100% water cut.

The company earlier developed its KOS MCF 350 for oil-external emulsions. This meter was tested in the North Sea (Maersk Oil & Gas, Denmark offshore), Gabon (Shell Gabon, Rabi field onshore), and Oman (Petroleum Development Oman, Marmul and Ramlat Rawl fields).

It is currently operating in the Ramlat Rawl field and the Lekhwair field Station B in Oman (Fig. 1).

The KOS MCF 350 meter was developed in cooperation with A/S Norske Shell and Shell Research and is based on a Shell patent. Shell provided laboratory facilities at Shell Research, The Netherlands, as well as field-test facilities in Gabon.

Kongsberg is wholly owned by the FMC Corp.

At least one other multiphase meter for high water cut has been developed by another manufacturer and field tested, this on Gullfaks B in the North Sea under sponsorship of British Petroleum Norway A.S., Elf Aquitaine Norge A.S., Phillips Petroleum Co. Norway, and Total Norge A.S. (OGJ, May 17,1993, p. 54).

MULTIPHASE METERING

During multiphase flow, distribution of liquid and gas across and along the pipe is not homogeneous but varies over time even when both liquid (oil and water) and gas flow rates are constant.

With changing flow rates, distribution of the two phases changes considerably, and various distinct flow regimes have been identified (Fig. 2)(52964 bytes).

Stratified flow has two distinct layers, one liquid and one gas. In slug or intermittent flow, the pipe cross section is filled alternatively with gas and liquid. Two other regimes commonly encountered are bubble and annular flow.

The most common flow regime in production pipelines, slug flow, is the basis for the KOS MCF meter.

Kongsberg says that most flow lines, because of fluid-transport capacity considerations, are designed sc) that flow will occur in the slugging regime. Slugging can also be induced by introduction of slug-generating devices upstream of the meter or by a slight inclination of the flow line.

Eliminating the need to separate produced gas, oil, and water streams for well testing, the company also says, can cut production costs. An element in realizing such savings has been the introduction of multiphase flow meters to replace conventional well-test equipment consisting of test separators and associated test lines and test manifolds.

Kongsberg says that its MCF meters permit continuous monitoring of low-producing onshore and offshore wells without costly traditional test equipment. Monitoring of subsea satellite developments is also possible.

Continuous well monitoring can lead to increased production rates through optimized production systems. Additionally, Kongsberg says such monitoring presents the opportunity for improved "tuning" of well streams that will reduce energy costs for production equipment because of better utilization of equipment.

Kongsberg notes three approaches to multiphase metering:

1. The traditional well-testing method involves separation and metering in which oil, gas, and water are separated and subsequently measured by volumetric meters in individual flow lines.

The size and weight of conventional test separators make them expensive, particularly for offshore applications where size and weight are at a premium. For satellite developments, they also require expensive manifolds and test lines.

These separators, says Kongsberg, and meters are also less representative because they sample flow rates only for fractions of the total production time of a well and not continuously.

2. Hybrid systems measure the density of the flow mixture in the main stream, while a fraction of the flow is diverted in a slip stream where gas and liquid are separated in a mini-separator. The distribution of water and oil in the liquid is then measured separately.

These measurements, made with gamma densitometers, combined with temperature and pressure measurements permit indirect calculation of volumetric flow for the three phases.

Hybrid systems are smaller and lighter than conventional test separators but still bulky, costly, and complex, says Kongsberg. And the problem of discontinuous sampling remains.

3. In-line metering depends on intrinsic flow properties to derive flow rates without introduction of equipment that disturbs the flow and requires major modifications to the flow line.

The flange-mounted field unit, says Kongsberg, is installed in the flow line and results in low cost and high reliability with only minor modifications to the flow line required.

MCF meters perform continuous in-line flow measurement.

Field units containing sensor elements are installed in the flow lines between the producing wells and the production separator and eliminate the need for a conventional test separator and its dedicated oil, gas, and water flow meters.

In those instances where each well has a dedicated multiphase meter, the need for the test manifold will be eliminated, provided that adequate well sampling can be achieved by other means.

PRINCIPLE, TECHNIQUES

MCF meters gauge the liquid and gas flow rates in slugging by continuously determining the cross-sectional areas occupied by each phase and multiplying each area by the velocity of the corresponding phase.

To do this, the meter gauges the cross-section of the pipe occupied by liquid and deducts this measurement from the total cross section to determine the area which is filled with gas.

It also measures the velocity of the liquid and the velocity at which the slugs travel, which has been shown to be proportional to the velocity of the gas.

The flow rates of oil and water are finally determined through determination of the water cut. The remaining part of the liquid is then by definition oil.

Flow measurements are made with a pair of parallel sensor plates inserted into the pipe in line with the flow (Fig. 3)(46621 bytes). Etched onto the sensor plates are several electrodes that form capacitors which generate the basic signals from which the flow rates are derived through intricate data processing.

The sensor plates are 1.65 mm thick and are 4 (-m wide in the direction of the flow. They contain one column of electrodes extending from the bottom to the top of the pipe, plus six single electrodes, four near the bottom of the pipe, the other two near the top.

The capacitors in the column are connected to electronic circuits that generate output signals representing the real and imaginary part of the admittance. The imaginary part is in turn a measure of the fraction of liquid in the mixture.

This is how liquid hold-up is determined, being that part of the cross section which would be occupied by the liquid if the phases were separated. The remaining cross section will necessarily be occupied by gas.

The velocity of the liquid flowing in the film at the bottom of the pipe is measured continuously through cross correlation of the signals from two capacitors a small distance apart at the same height near the bottom of the pipe.

Similarly, slug velocity is calculated from the time difference between the signals from two electrodes 20 mm apart near the top of the pipe. The latter measure indirectly gives the gas velocity.

The water cut is determined from measurements of the dielectric constant of the liquid.

CONTROL SYSTEM

The MCF system consists of a stainless steel spool piece with sensor, which is built integral with an Exd-certified (Cenelec standards) enclosure permitting installation in dangerous zones. The spool-piece is equipped with standard ANSI flanges and can be installed in the existing flow lines.

The meter includes a control system which will calculate data for the flow rates.

The system will communicate with supervisory control and data-acquisition/distributed-control systems or any other process-control system through standard serial links, such as RS 232C, RS 485, or Ethernet. Kongsberg says a dedicated computer board is integrated in the meter for this purpose.

Commonly used standard communication protocols can be implemented. If required, a dedicated control system with a comprehensive man/machine interface (MMI) can also be supplied.

Kongsberg says the sensor plates within the spool piece have been subjected to accelerated erosion tests to ensure that the sensor will withstand sand-production at rates up to 50 ppm.

Company tests have verified the target lifetime expectancy of more than 5 years.

Field tests in Gabon's oil fields have indicated that in operations with wax producing wells, excessive wax coating of the sensor plates can be prevented by such standard operational techniques as steaming, heating, and chemical injection.

Data in Table 1 (24627 bytes) show the deviation in measurement value between the reference (test separator) and an MCF 350 meter. The typical test separator, says Kongsberg, has an accuracy of 10% and is far from representing the reference measurement accuracy experienced in a laboratory.

The graphs in Fig. 4 (64985 bytes) show field-test results at Shell Gabon's Rabi field and were performed mid-January through the end of February of this year.

Each reading on the MCF 351 multiphase flow meter (one plot point on a graph) was compared with an actual measurement (along the diagonal) from a test separator.

The results shown are from wells with a variety of water cuts, 0-80% (where % water cut = [water flowrate/liquid flowrate] 100 when liquid is oil and water). Kongsberg says the span of both the liquid and gas flow rate was about 1:10; all measurements were obtained in the slug-flow regime.

The operator station for the MCF system, says Kongsberg, is normally located in an adjacent control room. The local data-processing feature of the MCF system, however, also permits transfer of measurement data from remote locations, via modem or satellite communication, to a centrally located control room.

NEW LABORATORY

In connection with developing the MCF series of multiphase flow meters, Kongsberg established an advanced multiphase flow laboratory at Kongsberg, Norway.

The laboratory operates with water, air, and oil (Shell Ondina 15) in any proportion at pressures up to 7 barg.

The flow line is a stainless steel piping system complete with reference flow meters for each phase as well as separation and pumping systems. Flow meters for a variety of pipe diameters can be tested. Advanced instrumentation is installed for a(-curate reference flow measurements.

Copyright 1995 Oil & Gas Journal. All Rights Reserved.