The present application claims priority to GB 0919386.3 filed on 5 Nov. 2009, which is hereby incorporated by reference in its entirety for any and all purposes.
Aspects of the invention relate to systems and methods of detecting the presence of slugs of one phase in a conduit carrying a multiphase flow.
The use of steam as a heating medium in industrial processes is very widespread. Most process and heating steam systems use saturated steam which is a two phase fluid comprising vapour as the first phase and condensate as the second phase.
Loss of heat in a steam system causes the steam to condense into droplets. These droplets generally form along the inner surfaces of the pipes along which the steam flows. The droplets coalesce into a film of condensate along the inside of the pipe which is swept along by the flow. The film of condensate accumulates in wells or recesses in the pipe from where it is picked up by the flow as a slug of condensate and carried along the pipe. As the slugs travel along the pipe they collide with components such as valves, connectors and flow meters, disposed within the pipe causing what is commonly referred to as “water hammer” or “water impact”. Steam systems are designed to accommodate saturated steam flow and are unsuitable for withstanding high energy impacts caused by slugs of condensate. The presence of slugs in a flow of saturated steam is therefore potentially damaging to components of the steam system and may lead to unexpected failure.
EP0593164 relates to monitoring steam flow rate in steam systems. U.S. Pat. No. 5,031,466 and US20050229716 disclose methods concerned with measuring the amplitude or frequency of fluctuations in the output signal to determine the proportion of a particular phase in a multiphase flow. Slugs of flow create a sudden change of phase in the multiphase flow which is seen as a direct positive spike in the output signal. The methods defined in US20050229716 and U.S. Pat. No. 5,031,466 are thus unsuitable for detecting slugs of flow in a multiphase flow since the spike in the output signal is either filtered from the signal or incorporated as one of many fluctuations. A consequence of this is that potentially damaging slugs of flow go undetected.
Therefore, there is a need in the art for improved systems and methods for monitoring and detecting slugs of one phase in a multiphase flow.
According to aspects of the invention there is provided a method of detecting the presence of a slug of one phase in a conduit carrying a multiphase flow. In one embodiment the method may comprise using a flow meter to provide an output signal representing mass flow rate of the multiphase flow; determining the rate of change of the output signal; and generating an alert in response to a predetermined profile of the output signal, when the profile includes a rate of change of the output signal falling outside a predetermined range. In certain embodiments, a slug is a discrete quantity of heavier or lighter phase fluid entrained in the flow of multiphase flow.
The profile may comprise a pulse, in which the rising portion of the pulse has a rate of increase exceeding a predetermined value. A profile comprising a pulse a rate of increase greater than a predetermined value improves detection of slugs and reduces the possibility of falsely detecting a slug in the flow. The falling portion of the pulse may have a rate of decrease exceeding a predetermined value. In certain embodiments, this may provide further certainty that the rate of increase first detected is caused by a slug. The falling portion of the pulse may extend below a value of the output signal representing the mean flow rate before the pulse.
As used herein, a flow meter may be any kind of flow meter which provides an output signal that can be analysed to determine its rate of change. Exemplary flow meters may include, but are not limited to: variable area flow meters such as that disclosed in EP 0593164, and/or target flow meters wherein the output signal may represent the displacement of, or a reaction force, exerted by a component of the flow meter. In other embodiments, the flow meter may be an orifice type flow meter, the output signal representing the pressure difference across the orifice.
The multiphase flow may be saturated steam. The multiphase flow may comprise any fluid which is entrained by a gas, such as water entrained by air.
Aspects of the embodiments may be provided in a non-transitory computer-readable medium having computer-executable instructions to perform one or more of the process steps described herein. According to another aspect of the invention there is provided a system comprising a flow meter in communication with a computer-readable medium having computer-executable instructions that when executed by a processor perform a method processing the output signal of the flow meter. Exemplary methods so as to carry out the one or more processes as described or defined herein.
Other details and features will also be described in the sections that follow. This summary is not intended to identify critical or essential features of the inventions claimed herein, but instead merely summarizes certain features and variations thereof.
Some features herein are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements, in which:
In accordance with various aspects of the embodiments, methods, computer-readable media, and apparatuses are disclosed that relate to the detection of slugs.
The exemplary variable area flow meter shown in
A spring 16 extends axially within the conduit 2 between the plug 2 and a pressure plate 18. The flow measurement system shown in
Wires 26 extend from the gauges 24 and are fed, via a metal tube 28, to the exterior of the conduit 2. The wires 26 are sealed within the tube 28 to prevent leakage of the medium within the capsule 20. The wires 26 connect the strain gauges with a signal processor 32 which, in turn, is in communication with a memory 34. The position of the capsule 20 may be selected by means of a stop 30 which provides an abutment against which the capsule 20 is pressed by the action of the spring 16.
The cone shaped plug 12 may have a particular configuration which is designed such that there is a linear relationship between the flow rate within the conduit 2 and the strain measurements, which simplifies calibration of the device.
In use, fluid may enter the inlet 4 of the pipe 2 and flows freely within the pipe to the orifice plate 8. A variable opening is provided between the external surface of the cone shaped plug 12 and the opening 10 in the orifice plate 8. Depending upon the rate of flow of fluid within the pipe, a force will be exerted on the plug 12 by the fluid at the inlet side of the pipe. This force causes the plug 12 to slide axially along the shaft 14 to increase the size of the variable opening. Movement of the plug 12 is resisted by the spring 16, and the spring 16, in turn, exerts a force on the diaphragm 18 causing the diaphragm 18 to deflect. The capsule 20 remains stationary during operation of the system. Deflection of the diaphragm 18 may be detected by the strain gauges 24 which may provide a strain output signal according to the load applied to them. These signals pass via the wires 26 through the tube 28 to the signal processor 32. Those skilled in the art will appreciate that the signals may be wirelessly transmitted. The signal processor 32 communicates with the memory 34 (which may be any tangible, non-transitory computer-readable medium) for storing data outputted by the signal processor 32. The strain signals obtained can be directly related to the flow rate within the conduit 2. For steady flow, the plug 12 will reach an equilibrium position, for which a certain force is exerted on the pressure plate 18. In operation, the output signal generated during normal flow of saturated steam is approximately constant, or will vary gradually in response to changes in the operating conditions. However, small fluctuations in the flow rate create noise in the output signal. The noise is filtered from the output signal to leave an output signal which represents the mean flow rate.
Various illustrative methods of detecting the presence of a slug of condensate in the flow may involve using the signal processor to compare the output signal against a predetermined profile of the output signal stored in the memory 34. The predetermined profile is a pulsed profile comprising a rising portion having a rate of increase which is greater than a predetermined value and a falling portion having a rate of decrease greater than a predetermined value. The falling portion of the pulse extends below a value which corresponds to the mean flow rate before the pulse.
The method of comparing the output signal with the predetermined profile in order to determine the presence of a slug may be explained with reference to
Between t1 and t2 the output signal is approximately constant and represents the mean flow rate of the flow under normal operating conditions. The signal processor 32 may be configured to store the value of the mean flow rate at t1 in the memory 34 for later comparison.
Between t2 and t3 the output signal increases rapidly in response to the arrival of the front of slugs at the plug 12. The signal processor compares the rate of increase in the output signal with the predetermined level stored in the memory 34. The predetermined level corresponds to that which would be expected to occur when a slug of higher phase flow passes over the plug 12 and would typically be greater than that which could be expected to be generated by saturated steam during normal operation. The predetermined level may be determined empirically or theoretically. If the rate of increase of the output signal exceeds the predetermined level, the signal processor generates an alarm or else logs a positive detection of a slug in the system memory 34. The alarm indication can be by way of a local warning display at the flow meters, such as warning light or sounder, alarm terminals or a signal to system management software via a communication port.
The output signal remains approximately constant between t3 and t4 as the slugs passes through the orifice plate 8 and around the plug 12. During this period, the heavier phase slugs exerts a larger force on the plug 12 than would be exerted by saturated steam during normal operating conditions, owing to the greater density of the water forming the slug, compared to that of vapour. Consequently, the mass flow rate is significantly greater than the mean mass flow rate upstream or downstream of the slugs. The duration of time between t3 to t4 corresponds to the size of the slugs and its velocity along the pipe. The time lag between the increase and decrease in the amplitude of the signal may therefore be used to determine the size of the slugs.
Between t4 and t5 the output signal decreases rapidly. The rapid decrease in the output signal corresponds to the slug completing its passage through the orifice plate 8 and the flow over the plug 12 returning to normal flow of saturated steam. The signal processor 32 compares the rate of decrease of the output signal against the predetermined rate of decrease of the output signal stored in the memory 34. The predetermined rate of decrease corresponds to that which would be expected to be produced as a slug passes over the plug 12. The predetermined rate of decrease may be obtained empirically or theoretically. If the actual rate of decrease exceeds the predetermined level then the signal processor 32 confirms the positive detection of the slugs logged in the memory 34.
The rapid decrease in the output signal caused by the passage of the slug is followed by a continued decrease in the output signal below a level which represents the mean flow rate before the arrival of the slugs. This further decrease is a result of a recoil action following the rapid decrease in the flow rate, caused by the plug 12 rebounding past the mean mass flow rate position following passage of a slug.
The signal processor 32 may be configured to execute computer-executable instructions on a tangible, non-transitory computer-readable medium (such as memory 34), that when executed, compares the flow rate at time t5, when the flow rate is at a minimum, with the mean flow rate before the passage of the slug stored in the memory 34. If the output signal drops below the mean flow rate by a predetermined amount, then the signal processor 32 may execute computer-executable instructions on a tangible, non-transitory computer-readable medium, that when executed provides further confirmation of positive detection of a slug. The predetermined amount is that which would ordinarily be generated in the wake of a slug and may be obtained empirically or theoretically.
The decrease in the output signal is followed by a steady increase between t5 and t6 in which the output signal returns to a level representative of the mean flow rate.
It will be appreciated that the profile of the output signal, i.e. rapid increases followed by rapid reduction to a recoil level below the mean signal level can be compared in its entirety with a predetermined profile to detect the presence of a slug. Alternatively, individual features of the profile can be compared.
Although the presence has been described with reference to a flow of saturated steam, it will be appreciated that it can also be applied to other fluids, for example to detect the presence of slugs of air or other gas in liquid flows, such as water.
The presence of slugs of water in a steam system is a symptom of problems within the system. Consequently, the ability to detect the presence of such slugs can serve as an alert to the system operator that the system requires investigation. Also, the ability to detect the presence of slugs of water at an early stage enables remedial action to be taken to avoid damage to components and processes of the system.
In use, fluid enters the inlet 4 of the conduit 2 and flows along the conduit 2 towards the outlet 6. As the fluid washes over the target 8 it exerts a force on the target 8 which deflects the rod 10. The amount of deflection of the rod 10 is proportional to the force exerted on the target 8, and hence the flow rate through the conduit 2. The deflection of the rod 10 is detected by the strain gauges 18 which provide a strain output signal proportional to the strain. This signal is passed via the wires 20, or wirelessly, to the signal processor 32. The signal processor 32 communicates with a memory 34 for storing the signal outputted by the signal processor 32. The strain signals obtained can be directly related to the flow rate within the conduit 2. The method of detecting the presence of a slug of condensate in the flow may involve using the signal processor 32, such as executing computer-executable instructions on a tangible, non-transitory computer-readable medium (e.g., memory 34), that when executed, may compare the output signal from the signal processor 32 against a predetermined profile of the output signal as described above in relation to the variable area flow meter.
Number | Date | Country | Kind |
---|---|---|---|
0919386.3 | Nov 2009 | GB | national |