Going with the flow

Flow measurement isn’t always foremost in industry debates, but the role it plays – and getting it right – is key. Elaine Maslin found out more.

NEL’s flow measurement facility.  Image from NEL.

As the oil and gas industry moves into harsher, more complex environments, so too do metering technologies and the requirements on them.

Building, testing and calibrating wet gas and multiphase meters, creating meters for high-viscosity oils, and using data now available through new generation devices for diagnostics and more intelligent calibration regimes, are some key themes for the flow measurement sector.

NEL’s 32nd International North Sea Flow Measurement Workshop brought together industry experts to discuss some of these issues and more.

The event, held in Scotland late last year, also saw the launch of the Flow Measurement Institute, which is looking to bring together academia and industry in flow measurement and also be a means to raise the profile in the UK of measurement and the value it brings to the UK.

Phil Mark, director of sales and marketing at NEL, outlined some of the key themes at the event. One is the debate over using substitute versus live fluids when testing or calibrating multiphase or wet gas meters. NEL supports the use of substitute fluids because of the improved uncertainty in calibration calculations, compared with that achievable with live fluids. Making sure that testing and calibration facilities can replicate as closely as possible both operating pressures and temperatures is also high on the agenda, be it for wet gas, multiphase or single phase.

“One of the big drivers for industry and vendors, is trying to find solutions for low cost multiphase flowmeters,” Mark says. “At the moment they are very high cost, at several hundred thousand of dollars (each), which potentially limits the number of applications. If a multiphase meter could be made less expensive, whilst delivering the same or similar performance, it would open up a bigger market.

“Other challenges facing the industry include the effect of increasing oil viscosity on the performance of both single phase and multiphase meters, and the effect of gas entrainment on the performance of ostensibly single phase meters. The use of diagnostics parameters, available from the more modern technologies, gives you access to much more than a straight-forward flow signal.”

NEL, based near Glasgow, Scotland, is looking at how changes in diagnostic parameters, can be used to monitor the performance of the flow meter, in order to move to condition-based monitoring and calibration, instead of time-based monitoring. The organization, which is the custodian of the UK’s National Flow Measurements Standards, is also looking at how the data gathered from such devices can be interpreted in a rigorous and reliable fashion.

Last year, NEL invested in the development of its high pressure wet gas facility, now the UK’s only independent commercial test facility that can test meters to flow rates over 2000cu m/hr under wet gas conditions and is capable of using oil and water components simultaneously.

NEL’s new high pressure wet facility near Glasgow. Image from NEL.

The event also heard from a number of suppliers and operators:

BP

Bill Pearson and Samir Ismayilov, from BP Azerbaijan-Georgia-Turkey – Azerbaijan, presented the development and field trials of SONAR Meter Technology.

SD Sonar 10in. Expro flow meter. Image from BP.

The project was a collaboration between BP AGT Region and Expro Meters, part of Expro Group. BP required retrospective, non-intrusive flow measurement in a number of “allocation” metering applications. For two applications, a clamp-on SONAR flow meter was trialled, including field cabling, transmitter, mounting clamp, transducer module, and cover assembly. The ActiveSONAR meter was EX ‘d’ Zone 1 certified.

The results indicated that clamp-on SONAR technology, as a non-intrusive retrofit meter technology, was at least comparable with other type of technology, such as ultrasound or differential pressure producer meters, and in some cases it was likely to exceed the performance of installed meters.

“As with any type of black box technology, assurance around meter performance is implicit from key parameters or meter diagnostics,” the presenters said. “As is current industry practice, with e.g. ultrasonic meter technology, it is possible to benchmark key meter performance parameters during factory calibration / function tests and use these to establish a baseline for performance monitoring / verification in the field.”

Pearson also described the prototyping of a SONAR technology meter diagnostic interface from which measurement stakeholders can verify meter performance. In the event that meter performance becomes sub-optimal diagnostics expert support can be sought and, where necessary, the clamp-on meter components can be readily changed under permit to work without the need for shutdown and dependence on isolations for workforce safety.

SD Sonar 10in. Expro flow meter. Image from BP.

The trial sites included a crude oil reception line at the Sangachal Terminal, Azerbaijan, where offshore oil and gas is processed prior to export. ActiveSONAR was also used as a secondary measurement tool to verify an existing legacy ex-test separator gas measurement device as well providing additional measurements during critical well testing operations.

GE

GE’s Anusha Rammohan and Baskaran Genesan, GE Global Research – USA, presented their insights into multiphase flow through ultrasound doppler.

Multiphase flow could be fully characterized and measured if the velocities and phase fractions of all the components in the flow were available through sensor measurements, they say. For a three-phase flow, for example, this involves measuring six independent parameters. “Since this is an almost impractical expectation, in reality only a subset of these measurements is available to a flow meter,” the pair say. “For example, not all component velocities can be directly measured. The burden then falls on the flow meter algorithm to fill in the missing pieces with an appropriate model.”

Flow models typically perform well, as long as the flow conditions are within the ranges for which they have been built, but their extrapolative capabilities are questionable making their performance and accuracy unreliable and unpredictable in the field, say Rammophan and Genesan.

Ultrasound Doppler measurements under slug flow conditions from testing at NEL.  Image from GE.

The ultrasound Doppler technique is used to measure velocities based on the scattered signals from small particles or bubbles in the flow. Scatterers could be gas bubbles or oil or water bubbles in the liquid mixture /emulsion.

“The technique works based on the relationship between the measured Doppler shift in the signal frequency and the scatterer velocity,” say the presenters. “There is ample evidence in literature on the ultrasound Doppler method being used to measurement velocity in flows with a very small amount of gas (low gas volume fractions). Under such conditions, the flow exhibits a predominantly bubbly flow regime, which is conducive for the Doppler measurement with the small gas bubbles acting as the scatterers.”

However, GE has looked at the application of the ultrasound Doppler technique to a much wider range of three phase flow conditions, with varying water cut and gas volume fractions.

“The ability of the ultrasound Doppler technique to measure the velocity and strength of small scatterers in the flow makes it a unique measurement that provides invaluable insights into multiphase flow,” say Rammophan and Genesan. “Analysis of the Doppler data, based on an understanding of the flow physics, showed that the scattered signal originating from the small bubbles in the flow contain easily extractable information about the liquid velocity.”

Ulstrasound Doppler measuring negative velocities in churn flow indicating instantaneous reverse flows. Image from GE.

The analysis was verified through extensive experimental data collected at NEL and SwRI (Southwest Research Institute), as well as GE’s in-house experimental facility, demonstrating that liquid velocity and in turn liquid flow rate can be estimated using the Doppler technique with high accuracy.

“Unlike traditional approaches which use a slip correlation to derive the liquid velocity from other more easily measureable quantities such as the mixture velocity, the Doppler measurement can provide liquid velocity information without the need for complex slip models or any other additional pieces of information,” says Rammophan and Genesan.

“This adds to the robustness and reliability in the flow rate measurement. Moreover, the proposed technique also resolves the measured velocity both temporally and spatially thus creating an information rich picture of the flow variations. This information in conjunction with other sensor information can provide critical parameters that can be used as inputs to flow models, thus increasing the overall accuracy of a meter.”

Modeling

Uncertainties around modeling was put under the spotlight by Phillip Stockton, from Accord Energy Solutions Ltd – UK. He believes more can be done to calculate uncertainties in simulation factors more robustly.

“The main purpose of simulation models within hydrocarbon allocation systems is to provide information regarding how hydrocarbons are behaving in a process plant,” he says.

“Allocation algorithms often include factors generated by these models. In calculating the uncertainty in the quantities allocated to each party in an allocation system, the uncertainty in the factors supplied from a simulation has to be accounted for. The uncertainty in the measured quantities is often known with a good degree of confidence but the available data on what the uncertainty of, for example, a shrinkage factor is, is not known and usually arbitrarily assumed to be a value of say ±5%.

“Simulation factor uncertainties can be calculated more rigorously. The uncertainty of a factor generated by a simulation model very much depends on the parameter in question. For example the uncertainty in a shrinkage factor for a dead oil will be lower than that for a lively condensate.”

Stockton has been considering the sources of uncertainty within the models and attempts to demystify the black box reputation of these models.

He suggests there are a number of methods that can be used to calculate simulation factor uncertainties, ranging from Monte Carlo methods to more simplistic short-cut methods.

Monte Carlo uncertainties. Image from Accord.

FORCE

Denmark’s FORCE Technology has been building what it says will be the world’s largest calibration loop for calibration of natural gas meters.

The new system has been designed using the company’s in-house technology, which sees natural gas circulated in a closed loop, Jesper Busk, department manager at FORCE Technology told the NEL event.

FORCE Technology had been using the technology since 2004, and established a prototype system, which circulates up to 10,000cu m/hr natural gas at a pressure of up to 50 bar.

Market demand led the company to design larger facilities. The new system comprises a closed loop, able to work up to 65 bar and a flow of no less than 32,000cu m/hr - and of 41,000cu m/hr at lower pressures.

At the heart of the system is two parallel connected high-pressure blowers, each producing 22,000- 41,000cu m/hr and circulating the gas in the loop at a variable pressure from 3-65 bar. The high-pressure blowers are driven by two 900 kW engines, which makes it possible to calibrate metres with diameters of up to 750mm.

The system provides traceability to the new high pressure calibration system minimizing the calibration inaccuracy to the absolute minimum. It was internationally approved in 2013, and saw FORCE Technology join the European cooperation on Harmonization of The European Natural Gas Cubic Metre (EUREGA). The system was due to be ready for operation by the middle of 2014.