Join OEdigital on Facebook Join OEdigital on LinkedIn Join OEdigital on Twitter
 

Improving flow measurement accuracy

Written by  Neil Bowman, NEL Wednesday, 01 November 2017 00:00

Changing production regimes over the life of a field can prove to be a flow measurement challenge. NEL's Neil Bowman looks at the challenges.

A turbine meter. Image from NEL.

As fields mature, so do their production characteristics. In the North Sea, for example, the maturity of the basin means there is more produced water here than on average globally.

In 2015, some 69% of well stream fluids were produced water, according to the Oil & Gas UK Environmental Report 2016. In 2015, the UKCS generated 2.4 tonnes of produced water per tonne of hydrocarbon. In Norway, this figure is about 2.3 tonnes.

For flow measurement, this can be challenging. Flow measurement is a vital aspect of hydrocarbon production, providing the means for well testing, process monitoring and production optimization, as well as the basis for fiscal and custody transfer measurement of hydrocarbons. Accuracy is paramount as even small measurement errors can amount to significant revenue loss.

Flow meters, like most types of process equipment have fundamental performance limitations making them susceptible to changing field conditions that occur as a reservoir matures. Over time, reservoirs that once produced almost exclusively hydrocarbons can gradually transition to a state where they produce almost exclusively water – in excess of 95% in some cases.

These large shifts in production rates impose a heavy burden on production equipment, which typically have a limited operating envelope and are increasingly pushed to perform under conditions they were not originally designed for.

This impacts flow measurement in a number of ways. Once of the most significant impacts comes from the diminished hydrocarbon production rate forcing meters to operate below their operating or calibrated range.

The range over which a flow meter can effectively operate, known as ‘Turndown,’ is the ratio of the maximum to minimum flow rate over which the flow meter can perform. This is sometimes also referred to as ‘Rangeability.’ Different types of flow meter typically have different turndown ranges depending upon their fundamental principle of operation.

Let’s consider the humble turbine flow meter as an example. Turbine flow meters, if implemented correctly, can be extremely accurate, and are frequently used in fiscal applications where accuracy is vital. However, like most flow meters, they have a limited turndown and care must be taken in selecting the correct meter or range of meters for a given application. The graph shows a typical calibration curve for a turbine flow meter.

Turbine Flowmeter Calibration. Graph from NEL.

Since the flow meter returns a pulse signal, which is proportional to the flow rate, the flow meter is calibrated in terms of the number of pulses per unit volume (K-Factor) of fluid, which flows through the flow meter. Intuitively, we might expect that the relationship between K-Factor and flow rate should be linear, however in reality we can see that this is not the case. The mechanics of the meter mean that there is a limited range of which this relationship is true. Out with these ranges, the behavior of the meter becomes significantly non-linear and more complex, generally resulting in lower accuracy. Flow meters are typically calibrated over this linear and ‘predictable’ flow range.

All meters exhibit this limitation in one way or another, however, depending upon the principle of operation, some meters have wider turndown than others. Differential pressure meters for instance, such as orifice plates or venturis, have a very low turndown of around 4:1, whereas some modern types such as ultrasonic meters can operate at turndown ranges of 200:1 or higher.

As reservoirs mature, dropping hydrocarbon rates typically force meters to operate at the bottom of their operating range, if not below, whereas produced water systems tend to operate at the high end of their operating range. In some cases, it is possible to characterize the behavior of a device below its defined operating range and thereby extend its capability – so long as the behavior of the device does not become significantly non-linear. In some cases, the only solution will be to either replace the flow meter or redesign the flow measurement system.

In addition to the operating range of the meter, another significant factor is phase contamination. As separation systems are pushed to operate at the extremes of their performance envelope, phase contamination becomes a serious risk. This can result in a mismeasurement of the hydrocarbon produced, but depending upon the meter type, can seriously affect its performance. Electromagnetic water meters for instance are affected by the presence of oil since oil is not conductive, and transit-time ultrasonic meters are seriously affected by the presence of small amounts of a secondary phase. Great care must be taken in understanding the performance limitations of the meter and the conditions in which it operates.

In addition to this, there are a number of other problems that can arise from changing field conditions. These include material erosion caused by increased sand loading, distorted, swirling or asymmetric flow profiles caused by upstream process equipment, or pipework resulting from system modifications and changes in fluid physical properties to name just a few. All of these factors must be monitored as they can have a detrimental effect on meter performance.

Since every application is unique, the solution to this problem is not always straight forward. A number of factors should be considered before selecting a course of action. In particular, accuracy requirements in addition to other relevant technical factors such as fluid properties, ideal turndown, contamination and operating conditions should be reviewed, and economic and human factors should be taken into account. In addition to which, manufacturers and regulators should be consulted and appropriate standards observed. 

About the Author

Neil Bowman is a project engineer at NEL, a provider of technical consultancy, research, testing and program management services. He has a Master’s degree (MEng) in Aero-Mechanical Engineering from the University of Strathclyde and is a Chartered Engineer and a Member of the Institution of Mechanical Engineers. 

Read 1181 times