GB2382140A

GB2382140A - Underwater leak detection using acoustic backscatter

Info

Publication number: GB2382140A
Application number: GB0127755A
Authority: GB
Inventors: Christopher Teal
Original assignee: Individual
Current assignee: Individual
Priority date: 2001-11-20
Filing date: 2001-11-20
Publication date: 2003-05-21
Anticipated expiration: 2021-11-20
Also published as: GB2382140B; GB0127755D0

Abstract

A method of leakage detection at an underwater location involving the steps of generating an acoustic or other signal pulse at a first location; directing the generated signal into a test region, at a known position relative to the first location, in which test region there could exist an interface L between fluids of varying density and particle or backscatter load; detecting reflected acoustic or other signal pulses at a second location, at a known position relative to the first location, back scattered from an interface as aforesaid existing in the test region; and generating an output indicative of the existence of the or each interface and its location relative to the first and/or second location.

Description

LEAK DETECTION This invention relates to leak detection.

The ability to be able to detect leaks from sub sea oil and gas installations such as pipelines, valves, risers, etc. , has been a requirement since the early years of offshore hydrocarbon exploration. Apart from a short term need to meet safety and operational requirements there are significant middle and long term environmental and economic aspects to be considered.

The environmental aspect is a major concern and increasingly regulatory authorities around the world have been becoming less and less tolerant of potential releases of polluting material into the marine environment.

I Sub-sea networks constructed to transport hydrocarbons (oil or gas), whether new and at the testing stage or existing installations, require regular inspection and maintenance for which detection and location of leaks is essential.

Two main methods of detection have been used to date with mixed success.

FLUORESCENCE.

Fluorescent dye is mixed with fluids present in a pipeline. Leakage results in fluorescent fluid being released into the sea and its presence serves to provide for detection, and to some extent the size, of the leak.

The majority of sub-sea inspections are conducted using a Remotely Operated Vehicle ('ROV') that equipped with an array of inspection equipment and sensors. li is known for an ROV to carry an underwater'black light' (Ultra Violet light). As the ROV scans the installation any leaked fluorescent material is exposed to the UV light which serves to excite the fluorescent dye to produce an emitted light at a longer wavelength. This emitted light would be detected by eye (diver or camera). These days it is more common for the UV light and the fluorescence sensors to be housed together in a single unit. The light source and emission sensors are'tuned'with optical filters in order to optimise the fluorescent properties of the dye thus increasing sensitivity and

reducing the effects of ambient light and spurious events. The data is transmitted from the ROV to a surface ship by way of an umbilical cable where the data is monitored and recorded on a computer.

This method has been successful on a number of occasions, however, it does have limitations, namely spatial coverage and sediment load.

SPATIAL COVERAGE.

A fluorometer needs to pass into the dye to detect it. Unless several such sensors are used in an array, detection, especially for small leaks in high water flows, can be a hit or miss process. Detecting the precise location of a leak (such as at a flange or coupling) can also be difficult in the case of larger leaks and/or in low water flow conditions as the dye forms a cloud. The sensor detects the presence of a leak but without a lot of ROV manipulation, and some luck, identifying the location of the leak can be elusive.

SEDIMENT LOAD.

Fluorometers are influenced by turbidity and the presence of suspended sediment.

Absorbtion of light presents a problem to the extent that complete blocking can occur.

Even when only some light is absorbed, the ephemeral nature of the transportation of suspended sediments causes superimposed'noise'on the sensor output signal. This highly variable noise makes it difficult or impossible to distinguish between data from the detection of fluorescence and the signal generated from light absorbtion.

ACOUSTIC HYDROPHONES These can be used to detect noise characteristic of a leak. Again there are a number of disadvantages to be allowed for. Each leak will have it's own acoustic signature which will emit sound at amplitude levels and frequencies depending on the shape and size of the break, the pipeline pressure at the break, the ambient flow conditions, etc. It is therefore difficult to pre-tune to a noise as can be done with fluorescence.

Pipeline operations tend to be an inherently noisy process. ROVs and other underwater vehicles have a number of mechanical, electrical and hydraulic systems on

board. These systems are normally in constant use producing noises that are continuously varying as the vehicle is in operation. The highly variable vehicle noises are superimposed on acoustic signals generated by a leak. Because of this it is extremely difficult to distinguish, or even recognise, the presence of a leak.

ACOUSTIC BACKSCATTER For many years techniques involving acoustic back-scatter ('ABS') have been used as a means of measuring the concentrations of suspended sediments over ranges of up to 600m or so when operating in the frequency band between 1200kHz and 75kHz.

However, in more recent years systems utilising ABS at high frequencies have been developed that can profile a column of water over lengths of several metres in fine detail. These systems generate high frequency acoustic pulses of the order of 1 to

6mHz. These pulses are focussed into a beam and transmitted from a dual mode receive/transmit transducer into the water. Particles, such as sediments, bubbles, plankton, scatter some of the acoustic signals with a proportion being received back at the transducer. These received signals are proportional to the amount of acoustic scattering, which in turn, is proportional to the concentration and particle size of the scatterers. By time gating the signal return from each transmitted pulse, data from cells along the beam path can be individually recorded, i. e. , the system produces a profile of data from known points along the beam. The number of cells or pockets of water along the direction of the beam that scattered acoustic signals are received back at the transducer can be 240 or more over distances of several metres.

According to a first aspect of the present invention there is provided a method leakage detection involving the steps of: generating an acoustic or other signal pulse at a first location; directing the generated signal into a test region, at a known position relative to the first location, in which test region there could exist an interface between fluids of varying density and particle or backscatter load;

detecting reflected acoustic or other signal pulses at a second location, at a known position relative to the first location, back scattered from an interface existing in the test region; and generating an output indicative of the existence of the or each interface and its location relative to the first and/or second location or to some other datum.

According to a further version of the first aspect of the present invention the detecting step involves detecting a reduced backscatter at a location when the pulse passes from the ambient fluid (such as sea water) into a leaked fluid where backscatter particles are at a lower concentration.

According to yet a further version of the first aspect of the present invention the detecting step involves detecting an increased backscatter at a location when the pulse passes from the ambient fluid (such as seawater) into a leaked fluid where backscatter particles are at a higher concentration.

Typically when an acoustic or other signal pulse impinges on a boundary between a layer bounding a first fluid and an adjacent layer of a second fluid differing in density from the first (for example juxtaposed layers of fresh and saline water) a detectable back scatter is readily detected.

It is believed that the identification of back-scattering to detect leakage could also be undertaken by other forms of generated signal besides acoustic. Typically a light signal could be used of an appropriate frequency or range of frequencies. It is also envisaged that other forms of electromagnetic emission could be used in environments and/or conditions in which the resulting back-scattering provides for an optimised detection opportunity. Given use of a light signal then phase change detection could be used to identify the occurrence of back-scattering.

Fluids within the sub-sea pipelines, whether liquid or gaseous, will normally exist at a different density to the surrounding seawater. Should the pipeline system have a leak then the acoustic back-scatter system will'see'the density gradient boundary layer between the surrounding seawater and the pipeline fluid.

As an additional detection element and included in the invention is the recognition of the backscatter changes in the volume of fluid containing leaked fluid. The leaked fluid will either have little or no acoustic backscatter or, in certain circumstances, have a high backscatter load. It is unlikely that the leaked fluid and the ambient fluid would have matching acoustic characteristics. It is this property and the ability to detect leaks that is also included in the invention.

In addition, a hard pipeline surface will provide a strong'bottom'signal thus assisting in location the exact position of the leak from, for example, a coupling or flange.

Another benefit over the fluorescent dye system mentioned earlier is the ability of an ABS system to penetrate through high suspended sediment loads producing backscatter signals at discrete locations along the acoustic path. When the beam travels from a high sediment load in saline water to a lower density medium almost free of suspended sediment a low signal return will be'seen'thus indication a leak. In such conditions known fluorometers would fail to sense an emitted dye.

Normally, acoustic beams in these frequencies are of the order on 50mm2 or less with a spread of no more than 10 thus the sensing area is very small. To overcome this limitation in apparatus for undertaking the present invention it is intended to use a 'fan'beam transducer that spreads the beam by 200 in one plane as is used in surveying side-scan systems. By angling the beam to be normal (that is to say at right angle) to the pipeline a beam width of about 0.35m is present at the pipeline. By attaching several transducers to ROV arms utilising the invention and arranging them in an arc facing towards the pipeline then it is possible to get complete acoustic coverage and thus detect leaks at any point in the sub-sea system under any ambient water flow conditions.

An exemplary embodiment of the invention will now be described with reference to the accompanying drawing of a leak detection operation of which: Figure 1 is a diagrammatic view of a pipeline being scanned; and Figure 2 is a close up of a part of Figure 1 whilst scanning is being undertaken.

Figure 1 shows a pipeline 11, having a joining flange 12, located in an underwater location. A boat, not shown, is undertaking an integrity survey of the pipeline 11 and using a sensor head 13 comprising an array A of acoustic transducers 14,15, 16,17, 18 mounted on a carrier 20. Each of the transducers 14-18 comprises an acoustic wave pulse emitter generator and a microphone adapted to receive output pulses reflected back towards the transducer. Typically transducer 16 comprises a pulse emitter 16E and a receiver 16M contained in a waterproof housing 16 H. The The head 13 is suspended from the boat (alternatively it can be mounted on an ROV, towed vehicle, sledge, AUV (autonomous underwater vehicle) or diver). The mounting provides for movement and positioning of the array A to be closely controlled so that it can be accurately aligned relative to the pipeline 11 in order to survey a test volume V including the flange 12 to establish whether or not a leak exists.

A video camera 21 is also included in the sensor had 13 to provide for ready visual inspection of test volume V or adjacent matter.

The transducers 14-18 are each coupled by a cable to an inboard control unit 22 (Figure 2) by means of which each transducer can be operated and the results displayed on a VDU 23.

With no leak existing in the vicinity of the flange 12 or locally on the pipeline 11 an acoustic scan by way of transducers 14-18 will not result in back scattered acoustic waves being detected by a microphone in one of the transducers.

In the event a leak L does exist from flange 12 in the pipeline 11 (as shown in Figure 2) then fluid from within the pipeline 11 (an oil or other fluids, liquid or gaseous) leaks into the water surrounding the pipeline 11 resulting in one or more interfaces between leaked fluid and water which reflect acoustic pulses emitted from one or more of the transducers to be back scattered towards the transducer array where the reflected pulses will be detected. The configuration and operating characteristics of the array A provides for and results in a detected leak signal being displayed unambiguously on VDU 23 to enable a decision to be made as to maintenance work to be undertaken. The

relative positions of transducer array, flange and pipeline are readily established and maintained by conventional positioning equipment incorporated in the system.

VDU 24 provides for the display of the picture obtained by means of the camera 21.

Claims

1 A method of leakage detection at an underwater location involving the steps of: generating an acoustic or other signal pulse at a first location; directing the generated signal into a test region, at a known position relative to the first location, in which test region there could exist an interface between fluids of varying density and particle or backscatter load; detecting reflected acoustic or other signal pulses at a second location, at a known position relative to the first location, back scattered from an interface as aforesaid existing in the test region; and generating an output indicative of the existence of the or each interface and its location relative to the first and/or second location or to some other datum.

2 A method as claimed in Claim 1 wherein the detecting step involves detecting a reduced backscatter at a location when the pulse passes from the ambient fluid (such as sea water) into a leaked fluid where backscatter particles are at a lower concentration.

3 A method as claimed in Claim 1 or Claim 2 wherein the detecting step involves detecting an increased backscatter at a location when the pulse passes from the ambient fluid (such as seawater) into a leaked fluid where backscatter particles are at a higher concentration.

4 A method of leakage detection at an underwater location an acoustic or other signal pulse is caused to impinge on a boundary between a layer bounding a

first fluid and an adjacent layer of a second fluid differing in density from the first (for example juxtaposed layers of fresh and saline water) an detecting detectable back scatter from the boundary.

A method of leakage detection as hereinbefore described with reference to the accompanying drawings.