Instrument Society of America Oil and Gas Conference
Maracaibo, Venezuela
22-25 September 1998

Implementation of new technology metering
by

A.W. Jamieson, Shell U.K. Exploration and Production

Summary

Flow measurement in the Oil and Gas industry is undergoing a revolution, and this is particularly evident in the North Sea. Novel techniques such as wet gas metering, ultrasonic metering are being applied for gas, while clamp-on ultrasonic metering and radar tank gauging are being tried out for oil. Multiphase metering is making slow but steady progress, and meters are now being installed that really give added value to field developments. Within operating companies, technical expertise in these techniques is limited, and is even more limited within manufacturers and engineering design houses. The major challenge is the building up of expertise across the industry to allow these techniques to be implemented successfully and to realise the large savings in capital and operational costs that are made possible through the application of new metering techniques.

Introduction

Over the last decade the efforts to drive down the costs of producing oil and gas has resulted in the development of new metering techniques that are now beginning to be applied. A decade ago high accuracy oil metering was done using PD meters or turbine meters with meter provers. High accuracy gas metering was done with orifice meters or turbine meters. The only multiphase meters we had were test separators or first stage production separators. Subsea metering was none existent. The situation for current producing fields is not that much different. Although the new techniques are more or less available, operating companies are justifiably wary of applying them. Experience with the new techniques remains almost completely with the developers of the individual techniques and their immediate sponsors in the oil companies. Operational experience is scarce. Obviously we have a chicken and egg situation: one cannot gain experience with a new technique until one has committed to use it, and project teams do not want to commit to a new technique without the availability of operational experience.

Almost all of Shell Expro's projects are keen to apply new metering technology as it is evident that large capital and operational savings can be made. My advice is straightforward: if there is no clear benefit from applying a new metering technique, don't. There are too many examples of new technology being applied as a 'nice to have'. Unfortunately, when it is switched on, doesn't work, and operational staff aren't really interested in it working, it isn't at all a 'nice to have'. Any confidence that there was in the new technique is lost, and it takes a very long time to restore that confidence. However, when a metering technique is developed to the stage where it is practical to implement it and there is a clear benefit of applying it in a project, I promote it vigorously.

The decision to apply a new metering technology should not be taken lightly, as if it cannot be made to work satisfactorily there may be no other option but to replace it with something else. There needs to be a large contingency budget for the things that will almost inevitably go wrong. This may appear to give a negative view of applying new technology. However, where we have applied new techniques and have kept a tight watch on the design, installation and commissioning, we have had fewer problems than with conventional systems, where everything is supposed to be fully understood.

In the rest of the paper I will first take a look into the future to 2010 to see what techniques are most likely to be in use. Then I'll discuss these techniques in more detail. Finally, I'll look at what needs to be put in place to allow the projection to 2010 to come true.

A look into the future

In 2010 there will be a much wider range of metering equipment in use than there is today. This simply reflects the fact that there will be a wider range of types of facilities for producing oil and gas. Thus we will have the old conventional facilities, simplified unmanned facilities, floating production systems, subsea production systems all coexisting together and feeding into multi-user pipelines. With advances in well technology we may even have subsurface metering on multilateral wells draining different accumulations. Added to this there will be much more automation than there is today.

There will be a number of the conventional turbine meter plus meter prover stations for oil export metering. Similarly for gas export there will be a number of conventional orifice metering stations. These systems have defined the accuracies the oil and gas metering industry accepts as reasonable. There is no drive to demand higher accuracy, but there is a strong drive to maintain these accuracies for critical systems but at a lower cost. However, every metering system will be looked at in its own context for cost effectiveness, with accuracy and cost being traded against overall financial exposure. The range of options can extend from no metering at all to full fiscal quality metering. If today’s conventional systems continue to provide cost effective solutions for the applications in which they are installed, they should continue to be used.

For most applications new techniques will be preferred. It is clear that in 2010 ultrasonic metering will be preferred for high accuracy gas metering. Ultrasonic meters, either clamp-on or multi-path appear to be the likely choice for crude oil transfer from floating production systems, but in both cases turbine meters are strong competitors. With facilities near the end of their useful lives there will be special requirements. Deferring abandonment as long as possible is worth a great deal to the industry, and having means to meter produced fluids to be sure of optimum performance is essential. By 2010 multiphase metering in its widest meaning will be widely used and it should be clear which of the techniques currently under development are most cost effective. Metering unprocessed streams will be common, with a strong preference for meters to be installed in individual flowlines. The best systems will be capable of near fiscal accuracy, and it will be practical to consider installing such high accuracy systems subsea.

The savings produced by using new metering techniques, especially multiphase metering, are large. Eliminating bulk separation facilities saves $15 - 20 Million per facility; a subsea test line costs typically $25 - 40 Million. In a company such as Shell Expro such savings over a five year period can easily be $150 Million.

From this brief look at the world in which the metering technology we are developing today will be used, we now look in more detail at the specific areas of technology.

Oil

The use of floating production systems has led to the need for cheaper oil export metering. Clamp on ultrasonic meters allow metering on the large diameter discharge lines from the floating production system to the shuttle tankers. This means that higher flow rates are possible, reducing loading times and the likelihood of having to disconnect in bad weather. Using duplicate meters overall accuracies are claimed to be of the order of 0.5 to 1%, but there still remains the problem of how to verify the performance of these meters on crude oil at very high throughputs. Large diameter multipath ultrasonic meters are also suitable for this application, and offer higher accuracy and better stability. For lines of about 20 inches diameter and larger there are very few facilities that can provide a calibration using crude oil, and a water calibration is not ideal.

Large diameter, high quality turbine meters could also be used for this application. They are very competitive in price with ultrasonic meters, and if used in pairs for checking each other will give the same indication of discrepancies as ultrasonic meters. Indeed, some meter types have been used for several decades and their performance is well established. Manufacturers of large turbines meters do not appear to want to compete in what is likely to be a fairly large market, and hence ultrasonic meters are getting an opportunity to dominate this market.

Radar tank gauging on the oil tanks of storage tankers is also being applied, but the movement of the vessel when on station degrades the overall accuracy significantly from that which can be achieved in calm conditions in harbour. It is likely to be too difficult to improve the corrections for movement sufficiently for this approach to become fully acceptable.

For oil metering in smaller line sizes Coriolis meters are becoming common. In Shell Expro we still have concerns over the use of thin walled vibrating pipes in main oil flowlines, especially for those above 4" diameter. However, better quality control in manufacture and a wider choice of materials means that we are using more of these meters. From a metering point of view we are pleased with the performance, but there are still significant issues to be resolved in mounting Coriolis meters to overcome errors due to vibration. This still seems to be somewhat of a black art.

Gas

High accuracy (dry) gas metering in gas producing companies will probably be done in the future using ultrasonic meters, for export and for allocation. Shell Expro has installed three Instromet 5-path Q-Sonic meters at St Fergus for fiscal metering of the gas supplied to the power station at Peterhead operated by Scottish Hydro Electric. This was the first application in the UK where ultrasonic meters were used as the primary fiscal measurement, and probably in the world. Where fields need to increase the capacity of the metering stations, ultrasonic meters are first choice. Our new developments are using ultrasonic meters in their base cases. It is unlikely that Expro will install any more conventional orifice metering stations for fiscal duty. Indeed, on one current project where orifice metering will be installed, a thick plate flow conditioner will be used to reduce the overall length of the metering system.

The advantages of an ultrasonic metering station over an orifice metering station are lower overall cost of about 30 % and size of 50%. Maintenance is significantly reduced and the inbuilt diagnostics tell when things are beginning to go wrong. Currently ultrasonic meters need to be calibrated at a high quality test facility. We have seen a sensitivity of the meter calibration to the surface finish of the meter interior of the order of 0.5% from clean to dusty. The reasons for such shifts are not at all clear, but we suspect that they do not only occur with ultrasonic meters, but affect all gas meters. The shifts observed are usually considerably smaller than the +/- 0.6% which is the basic uncertainty for an orifice plate discharge coefficient.

For gas transportation companies the situation is somewhat different. In continental Europe gas transporting companies have long experience of using gas turbine meters at large export stations and as sales meters to large users. They will continue to use these, and are increasingly using two meters of two different types in series to give warning of discrepancies. Thus ultrasonic meters are being used as back up meters for turbine meters.

Ultrasonic meters are also very suitable for production metering from separators. We have also installed a Daniels 4-path ultrasonic meter on Schooner on the test separator gas outlet. The near fiscal metering on Schooner is by wet gas venturis; the ultrasonic meter is used, among other things, to provide a check on the venturi meters calibration. Metering of wet gas will be dealt with under multiphase metering.

Multiphase

Multiphase metering is slowly coming of age. Development of multiphase meters began seriously in the early 1980s, thus we have somewhat in excess of 15 years effort, and a investment across the industry of at least $80 Million. Much of this investment has gone into setting up an infrastructure in which multiphase metering has become recognised as a viable technique. Multiphase meters are now commercially available and can be installed topsides to give clear added value to field developments. For subsea costs are much higher and installation more difficult. Nevertheless it is now evident that for Shell Expro, the largest benefits from multiphase metering will come when subsea meters can be used to allocate third party production.

In 1995 a forum consisting of Shell, BP, Statoil, Hydro and Saga presented a paper at the North Sea Flow Metering Workshop entitled "Oil companies needs in multiphase flow metering". This paper is still quite up to date, the main requirements on accuracy being 5 - 10% relative uncertainty for liquid and gas flow rates, and 2% absolute uncertainty for watercut.

Multiphase meters are not yet at the stage where they are meeting these target accuracies. Thus they should only be installed where there are means of checking their performance. As with any new technology it is important with the early installations not to lose the confidence of the operators. We must be able to stand very firmly by what the meter is telling us, this means being very critical in the assessment of any test data and not becoming too attached to a particular technique. The range of possible multiphase meter applications is enormous. It is unreasonable to expect that one type of meter will be able to satisfy all applications. As an operator we also want to have a choice of manufacturer to ensure that prices are not over-inflated.

Shell Expro are pursuing multiphase metering very actively. The following gives the status at present.

Operational meters:

Schooner - venturi wet gas metering (From October 1996)

Anasuria - Fluenta multiphase meter for well testing (from October 1996)

Auk, Gannet, Southern Field Unit - Tracers (from about mid 1995)

Gannet - 2 MFI multiphase meters for internal allocation (from July 1997)

Auk - ESMER for well testing (from December 1997)

Applications:

Every project is now considering multiphase metering at the conceptual design stage. We are looking very hard for a suitable subsea application.

Meter development:

We support Shell Group work in developing multiphase meters. We have also been supporting the development of the ESMER pattern recognition multiphase meter, originally at Imperial College, but now with PSL (Petroleum Software Ltd) and Daniel Europe. This promises to give a versatile and relatively cheap multiphase meter.

In Shell Expro we assess the suitability of a multiphase meter for a particular application, matching the meter performance envelope to the well or field production profile, as described in the above mentioned paper. We also translate the results of tests or manufacturers' data into a format that we want to use rather than the numerous formats used by manufacturers to try to enhance the apparent performance of their meters.

From evaluation of test data to date and our limited operational experience the following points can be made:

Accurate measurement of water cut at all gas volume fractions range is most important, and particularly at the higher water cuts. The multiphase meters that are commercially available are fairly reasonable two phase meters and are approaching the target accuracies for liquid and gas. They still do not provide good enough watercut measurement.

There is a need to be able to calibrate multiphase meters in situ. It may be possible to extend the present tracer techniques, but at present it is difficult to accept the validity of calibrations made in laboratory test loops at conditions very different from those in the field.

However, whenever we have compared multiphase meters with current conventional separator measurements, there have always been significant problems in doing so.

This leads on to the need for test facilities that can more closely simulate field conditions. Obviously such facilities will be expensive, but only a small fraction of the potential savings multiphase metering can achieve.

Needs to achieve the above

As I stated at the beginning of the paper, a revolution is taking place in metering. There is concern within Expro and by our co-venturers that where new metering technology is being applied and there is no recognised pertinent standard, there is a risk that meter selection and meter station facilities may be inappropriate for the required duty.

Almost all of our projects are applying novel, cost saving techniques in metering. To recap, examples are clamp on ultrasonic meters for export from floating production systems, wet gas metering using venturis, fiscal ultrasonic meters, thick plate flow conditioners for gas, Coriolis meters, possible wet gas ultrasonic, then all the multiphase applications. There are no standards existing for these techniques, although there are committees of interested parties producing early drafts. It is a widely held view across the industry that it is too soon to be drawing up definitive standards in these areas. Shell Expro, in common with other operating companies, is seeking to use industry wide standards where possible, and reduce the number of internal standards to a minimum.

Within Shell Expro we consider that the best approach is that followed by the United Kingdom DTI, and which is embodied in their latest guidelines for metering systems. Each application is considered in the light of its particular circumstances, and appropriate techniques agreed between the relevant operators, manufacturers and the DTI as conforming with good oil field practice. This puts the emphasis on getting a good functional specification agreed by all parties. The basis for that specification can be standards, special test results, expert opinion from appropriate resources, and so on. There need be no risk that meter selection and meter station facilities may be inappropriate if care is taken at this stage. Indeed, this approach is fully auditable and forces one to consider all the relevant issues. We think this is a much more practical approach than trying to draw up new standards that cannot yet accurately reflect real operating requirements.

Given the variation in requirements across applications and the wide range of techniques, it is essential that facilities designers, engineering contractors and meter vendors have an open dialogue with the users of the metering information, for example reservoir and petroleum engineers or the commercial departments responsible for the sales contracts. Further, a metering system specification must take into account the expected changes in flow rate over the expected service life. Consultants, contractors and their clients must realise, in turn, the substantial investment of time and effort needed by the vendor to tender a proper quotation for a system using new technology, especially if it is for a multiphase metering system. Co-operation will be necessary to allow vendors, intermediaries and end users to properly assess and qualify the metering system.

Conclusions

The new metering techniques that have been in development over the last decade are beginning to be applied. In the UK this has facilitated by the view shared by operators and the DTI that the metering techniques used on a particular application should be cost effective.

By the year 2010 a much wider range of metering techniques will be in use than at present. Multiphase metering especially will play a large role.

Government departments, operators, design contractors manufacturers and consultants will have to co-operate to ensure that expertise is built up to achieve the cost saving benefits possible with new metering technology.