July 2019, Vol. 246, No. 7
2019 P&GJ Metering & Measurement
Looking to Lessen Fluctuating Flow Problems in Measurement
By Chris Mills, Engineer, TÜV SÜD National Engineering Laboratory, United Kingdom
Good measurement practices suggest flow measurements should occur without instabilities or pulsations in the flow. Even minor inaccuracies in the measurement of high-value products can lead to significant financial exposure over extended periods of time.
Given the high product value and throughput of most hydrocarbon production processes, a great deal of importance is placed on the accuracy of these measurements.
For oil flow measurement, turbine, Coriolis and ultrasonic flow meters are the most commonly used flow meters. However, as oil production declines, these measurement technologies are experiencing substantial changes in the stability of the flow, which could have serious consequences on accurate flow measurement.
Unstable flow can occur as a result of several factors: physical causes such as increased water content, decreased production and pressure fluctuations, or technological causes, including the type of pump being operated system or incorrectly sized test separators.
Each of these can have a significant effect on flow stability, with frequency fluctuations of a range of magnitudes occurring. This happens when the content of water within the oil stream has been increasing in the final export oil streams from oil platforms, with the water content varying significantly across time. This can cause flow stability issues and manifest as sizeable low-frequency fluctuations in the flow rate.
As export meters often measure the flow into common export pipelines, this creates another issue because unstable flow rate from one source adversely affects the flow stability of other sources.
This could occur via low-frequency fluctuations distorting the co-mingled flow, which would subsequently distort the flow rate at source. Such a situation would be significantly magnified in a situation where there is an export pipeline with multiple input lines. The fluctuating flow of one source could alter the stability of the other inputs and therefore the co-mingled flow.
The stability of flow must be considered to ensure that accurate measurements are achieved during both the export measurement and the meter calibration.
However, while it appears many flow meters are operating in the industry under fluctuating flow, each is still calibrated in a laboratory under steady-state flow conditions. Despite fluctuating flow being unacceptable for accurate measurement, due to declining oil production, aging assets and tightening budgets, this is becoming increasingly common and must be addressed.
Unsteady Flow
Although it has been identified as a potentially substantial issue, the effect of unsteady flow on different measurement technologies has not yet been fully investigated or examined.
This is in part because research and calibration flow measurement facilities calibrate flow meters in ideal, steady flow conditions. As noted previously, achieving fluctuating flow for research purposes has never been required or desirable from a measurement perspective.
Furthermore, the effect fluctuating flow has on reference measurements would require a total mass or volume comparison. That would eliminate flow measurement facilities using pipe-provers, compact-provers or master-meter systems. The most suitable reference system would require a weigh bridge system to determine total mass or volume passed.
There is a significant knowledge gap currently: Very little data exists on the performance of flow meters in fluctuating flows. Investigating the performance of turbine and differential pressure devices, along with ultrasonic and Coriolis flow meters could be extremely important for industry.
Differential pressure devices, such as orifice plates, Venturi devices and cone meters all use the Bernoulli equation to determine the flow rate. Since the pressure of the flow would be strongly influenced by fluctuations, then it is highly likely this could introduce significant flow measurement errors.
As a result, we are proposing research that would provide conclusive data on the effect of fluctuating flow on differential pressure technologies. However, determining the performance in both low-frequency and high-frequency fluctuations requires considerable research investment in terms of both cost and time.
One option would be to complete a large-scale test program to evaluate a range of flow meters under non-steady state conditions. This would allow a better understanding of the issues and provide quantitative evidence of flow measurement performance in unsteady flow.
At our U.K. National Standards flow facility, this would manifest itself as a range of flow fluctuations of up to 10 Hz, being created to enable a wide range of flow measurement technologies, such as turbine, Coriolis and ultrasonic flow meters, to be evaluated under non-ideal conditions. This could cover cyclical, asymmetric and random pulsations.
In the laboratory environment, the advanced flow meter diagnostics from these technologies could be trended and analysed with respect to unsteady flow.
For example, ultrasonic flow meters offer diagnostics that can provide measurements of parameters, including speed of sound, signal-to-noise ratio, flow velocity, alarm and failure indicator, and velocity sampling interval. It may also be possible to use the advanced flow meter diagnostics to indicate when fluctuating flow is occurring as a qualitative measurement.
Ideally, it may be possible to correct adverse effects on the measurement that may arise from fluctuating flow. While it is not possible to pre-empt the results of any such test program, this is certainly an area that would be of great interest to industry.
Due to the lack of traceable data currently available to industry, it is not possible to specify what technology might deliver the solution to the problem of fluctuating flow. Only by completing a thorough evaluation program, will such a determination be made possible.
However, once this goal is achieved, some measurement techniques that are prone to mismeasurement due to fluctuating flow could be replaced, delivering greater confidence in the measurement of fluctuating flow and consequently greater confidence in financial transactions across industry.
Author: Chris Mills is a consultant engineer at TÜV SÜD National Engineering Laboratory, a provider of technical consultancy, research, testing and program management services.
Comments