WO2016042281A1

WO2016042281A1 - Multiphase stream conditioning

Info

Publication number: WO2016042281A1
Application number: PCT/GB2014/052794
Authority: WO
Inventors: Pablo Matias DUPUY; Oddbjørn Rekaa NILSSEN
Original assignee: Statoil Petroleum ASA
Current assignee: Equinor Energy AS
Priority date: 2014-09-15
Filing date: 2014-09-15
Publication date: 2016-03-24
Anticipated expiration: 2017-03-15

Abstract

The invention is concerned with conditioning a multi-phase stream, and in particular a gas stream with low liquid content. Specifically, the invention provides a method of conditioning a multi-phase flow of gas with entrained liquid droplets in a process stream section where a multi¬ phase measurement sensitive to the hold-up is to be made, comprising heating a surface within the section to a temperature such that droplets of liquid contacting the heated surface are subject to the Leidenfrost effect. This prevents the heated surface from being wetted by any droplets of liquid which may come into contact with it, so that liquid does not accumulate on the surface, but remains in the gas flow. As a result, accurate measurements of the overall density of the gas, and thus the amount of liquid entrained in the gas, can be made.

Description

MULTIPHASE STREAM CONDITIONING

The invention relates to a method for conditioning a multi-phase stream, and in particular a gas stream with low liquid content.

It is often necessary to measure the liquid content of a process stream. In particular, during the extraction of natural gas, it is important to measure the amount of liquid chemicals, oil, condensed hydrocarbons and/or water in the gas, to meet quality specifications. The presence of too much liquid in the gas can cause serious damage to equipment intended to operate on gas (such as compressors). It is particularly important to measure the liquid content of gas emerging from a gas scrubber or other device for separating liquid from the gas, to monitor the performance of the scrubber.

A known apparatus for measuring fluid density, and specifically the density of a natural gas stream, is disclosed in WO 201 1/039534. The apparatus includes a source of beta particles, and a detector capable of detecting the beta particles.

The fluid (which in practice is a gas with a small amount of entrained liquid) passes between the source and the detector, and the amount of beta particles detected by the detector varies with the density of the fluid. Since the density depends to some degree on the amount of liquid entrained in the gas, a variation in the amount of beta particles detected by the detector can be correlated with a variation in the amount of liquid entrained in the gas.

There are several known apparatus for the measurement of liquid content in a gas based on a measurement of the hold-up, that is, the liquid to gas area fraction at a given cross-section or representative volume.

A problem which can arise with devices of this type when used in gas streams with very low liquid content (around 1 part per million by volume) is the formation of liquid film on measuring surfaces (as a result of droplet deposition or condensation). As liquid film will move at a lower velocity than the gas, a measurement of the hold-up (area ratio) will not be representative of the liquid to gas flow ratio. For example, if a liquid film were to form on the detector of WO

201 1/039534, the film would have a considerable effect on beta particles passing through it. Further, if that liquid were to move (say) 100 times slower than the gas flow, it could be interpreted as 100 times the amount of liquid moving at the gas velocity. Other types of density or phase measuring device are also prone to errors arising from the presence of liquid films, and a liquid film with a thickness of only a few micrometres can produce errors up to two orders of magnitude on instrument reading.

It is possible to form water-repellent coatings on surfaces. If such a coating were to be formed on the surface of a detector as described above, it would prevent water in the gas stream from being deposited or condensing on the coated surfaces. However, as mentioned above, a natural gas stream can also include condensed hydrocarbons, and a water-repellent coating will not necessarily prevent the condensed hydrocarbons from forming a film. In addition, the uncertainty due to aging or service fouling is difficult to account for.

It is also possible to calibrate the relationship between area ratio and flux ratio, or use some empirical terms or models. However, such a solution is less attractive due to the need for calibration and increased uncertainty of fluid composition, and general reduced flexibility.

The invention has been made in view of the above problems, and aims to condition the multiphase stream to prevent formation of a liquid film, to allow more accurate measurement of the liquid content of a gas stream.

According to a first aspect of the present invention, there is provided a method of conditioning a multi-phase flow of gas with entrained liquid droplets in a process stream section where a multi-phase measurement sensitive to the hold-up is to be made, comprising heating a surface within the section to a temperature such that droplets of liquid contacting the heated surface are subject to the

Leidenfrost effect.

As will be explained in detail below, the Leidenfrost effect prevents the heated surface from being wetted by any droplets of liquid which may come into contact with it. As a result, liquid does not accumulate on the surface, but remains in the gas flow, and so accurate measurements of the overall density of the gas, and thus the amount of liquid entrained in the gas, can be made.

In a preferred form, the surface is heated by means of a heater located beneath the surface. This arrangement allows the surface to be heated without affecting the flow within the pipe.

In an alternative arrangement, the surface is heated by means of a hot gas which is directed towards or along the surface. As well as heating the surface, the hot gas will also tend to blow any liquid away from the surface. The hot gas can be supplied from outside the pipe; however, it is preferred for a part of the multi-phase flow of gas is dried and heated to produce the hot gas. This avoids any risk of the gas in the pipe being contaminated.

An additional element of the invention is placing a liquid film breaker upstream of the measurement section. The film breaker (which may be an entrainment disk) is a device which acts to ensure that most of the annular flow (at the outer edge of the process stream section) is entrained into the gas flow. This will facilitate the heating discussed above, as it is easier to prevent deposition of a few droplets than it is to evaporate the entire annular liquid film.

As mentioned above, the presence of a film of liquid on the detector can have a very significant effect on the accuracy of the measurements made. The Leidenfrost effect prevents a film of liquid from forming on the detector, and thus allows accurate measurements to be made.

A preferred embodiment of the invention will now be described by way of example only and with reference to the accompanying drawings, in which:

Figure 1 is a schematic cross-sectional view of a part of a natural gas pipe with a detector, incorporating means according to the preferred embodiment.

As mentioned above, Figure 1 shows a part of a pipe 10 for carrying a natural gas stream. The natural gas stream will include some liquid, in the form of water and/or condensed hydrocarbons.

The pipe is provided with a detector 12 for measuring the density of the gas stream flowing past, and this density can be used to determine the liquid content of the gas flow in a known manner. In the current embodiment, the detector 12 is arranged downstream of a venturi 14.

In order to avoid the formation of liquid films on the surfaces of the detector where such films would cause problems, heating means are provided to raise the temperature of these surfaces. More specifically, the surfaces are heated to a sufficiently high temperature for the Leidenfrost effect to occur.

The Leidenfrost effect occurs when a drop of liquid comes into contact with a surface whose temperature is well above the boiling point of the liquid. At the initial contact of the drop with the hot surface, the bottom part of the drop boils immediately. This produces a layer of gas between the hot surface and the droplet, and this layer of gas supports the rest of the droplet, preventing further direct contact between the drop and the hot surface. Since the gas will normally have a much lower thermal conductivity then the liquid, further heat transfer between the hot surface and the drop is reduced, so the drop remains suspended above the hot surface and does not come into further contact with it.

Thus, as long as the surfaces are heated to above the critical temperature for the Leidenfrost effect to occur (hereinafter "the Leidenfrost temperature"), liquid will not form a film on the surfaces, but will instead form discrete droplets. The Leidenfrost temperature for water (or aqueous phase) depends on a number of factors (such as the nature of the surface); however, the Leidenfrost temperature for water is greater than the Leidenfrost temperature for the condensed

hydrocarbons likely to be found in a natural gas stream, and so it is only necessary to heat the surfaces to the Leidenfrost temperature for water (or aqueous phase) to ensure that the liquids in the natural gas stream form discrete droplets rather than a film. Other hydrocarbon mixtures can have a different Leidenfrost temperature, but this can easily be found theoretically or through a routine lab-scale test.

Since the surfaces are exposed to the flowing gas stream, the motion of the gas will tend to push the droplets off the surfaces, thus keeping the surfaces entirely liquid-free and ensuring accuracy of the density measurements made by the detector.

Two different ways of heating the surface are shown in Figure 1. In practice, either or both methods can be used.

Firstly, a heating element 20 can be located underneath the surfaces 22 where condensation is to be avoided, in order to increase the temperature of these surfaces. Figure 1 shows a heating element 20 disposed in the wall of the pipe.

Secondly, it is possible to draw off a part of the gas flow upstream of the detector. Some of this drawn-off gas is dried using a meshpad 30 or similar device, passed through a heater 32 where it is heated to above the Leidenfrost

temperature, and then injected back into the pipe so that it flows along the surface 22 to be heated generally parallel thereto. The remainder of the drawn-off gas is reintroduced into the pipe downstream of the detector 12. In an alternative form, the drier device may be located inside the process pipe, so that the drawn-off gas is already dried. The heated gas will heat the surface 22 to raise its temperature to above the Leidenfrost temperature, and will also have the effect of helping to blow droplets away from the surface.

The Leidenfrost effect prevents a film of liquid from being formed on the surfaces; instead, liquid will form droplets that do not come into contact with the surface, and the movement of the gas stream will tend to sweep the droplets from the surface. Thus, the surfaces will remain free of liquid, so that the accuracy of the flow density measurements can be improved.

Further, since no liquid is deposited on the surfaces, the liquid will remain dispersed in the gas. As a result, the average velocity of the liquid phase will be very close to the average velocity of the gas phase, and so the slip ratio (the ratio between the velocity of the gas phase and the velocity of the liquid phase) will be close to unity (that is, a "no slip" condition). The "no-slip" condition reduces the need to calibrate multi-phase meters such as the detector described in WO

201 1/039534. Trustable (and calibration-free) multiphase metering is particularly important for subsea applications and monitoring of subsea processes.

Claims

1. A method of conditioning a multi-phase flow of gas with liquid in a process stream section where a multiphase measurement sensitive to hold-up is to be made, comprising heating a surface within the section to a temperature such that droplets of liquid contacting the heated surface are subject to the Leidenfrost effect.

2. A method as claimed in claim 1 , wherein the surface is heated by means of a heater located beneath the surface.

3. A method as claimed in claim 1 , wherein the surface is heated by means of a hot gas which is directed towards or along the surface.

4. A method as claimed in claim 3, wherein a part of the multi-phase flow of gas is dried and heated to produce the hot gas.

5. A method as claimed in any preceding claim, wherein a liquid film breaker, such as an entrainment device, is installed upstream of the process stream section to facilitate the conditioning.