GB2527794A

GB2527794A - Permittivity measurement of layers

Info

Publication number: GB2527794A
Application number: GB1411807.9A
Authority: GB
Inventors: Jan Kocbach; Kjetil Folgerã
Original assignee: Tecom Analytical Systems; TeCom AS
Current assignee: Tecom Analytical Systems; TeCom AS
Priority date: 2014-07-02
Filing date: 2014-07-02
Publication date: 2016-01-06
Anticipated expiration: 2034-07-02
Also published as: GB201411807D0; GB2527794B

Abstract

A system and method that permits measuring properties and thickness of layers, particularly layers close to a pipeline wall, and more particularly fluids flowing inside a pipe is provided. The system comprises a sensor operating in a non-radiating mode in a first frequency range and operating in a partly radiating mode in a second frequency range, where the non-radiating mode is defined as when reflections from a first layer interface are dominating the reflection coefficient, and where the radiating mode is defined as when reflections from a second interface of the layer interfere with reflections from a first interface. The sensor may be a coaxial probe or circular waveguide and may use a short pulse radar technique signal or a swept frequency signal for the second frequency range. The first frequency range may be used to determine permittivity and the second frequency range may be used to determine layer thickness value.

Description

Background of the Invention

Technical Field

The invention relates to measurement in general and more specifically a system and a method for characterising layers close to a pipeline wall.

Background Art

From prior art one should refer to NOl 9971024 regarding device for measurement of coefficient of reflection for high frequency waves in fluids as well as method for determining water contents in multi phase pipe flow using said device.

The problem is that this invention does not assist in determining fluid layer thickness.

One should also refer to JPH5-1 13323 regarding measurement method for film thickness.

This document does not provide any means for determining the property of the film or layer that is measured and the disclosed method requires information about the layer property in order to calculate the thickness.

One should also refer to US2O1 0064820 regarding measurement of multiple-phase fluid in a pipe. However this document discloses a method that requires two technologies for measurement and cannot be used for thicknesses greater than the sensitivity or penetration depth of the largest probe.

Also one should refer to US61 98293 disclosing method and apparatus for thickness measurement using microwaves. However the method disclosed requires prior knowledge of the permittivity of the material to be measured.

Disclosure of the Invention

Problems to be Solved by the Invention Therefore, a main objective of the present invention is to provide a system and method that permits measuring properties and thickness of layers and fluid films, particularly layers and fluid films close to a pipeline wall, and more particularly fluids flowing inside a pipe. The fluid film may be either stationary or flowing, and the fluid behind the layer or film may be either stationary of flowing. Possible layers include deposits of wax, scale, asphaltenes or hydrates.

Means for Solving the Problems The objective is achieved according to the invention by an apparatus for characterising layers as defined in the preamble of claim 1, having the features of the characterising portion of claim 1.

A number of non-exhaustive embodiments, variants or alternatives of the invention are defined by the dependent claims.

The present invention attains the above-described objective by a probe for combined perm ittivity and thickness estimation.

The permittivity in a first frequency range is estimated by using the probe in a first frequency range.

The permittivity in a second frequency range is estimated from the permittivity in the first frequency range in combination with application specific knowledge of the permittivity spectra characteristics.

The thickness of the layer is found by analysis of the response at the second frequencies frequency range in combination with the estimated permittivity value in the second frequency range and application knowledge.

Effects of the Invention The technical difference over N019971024 is the use of a second frequency range for thickness estimation. This allows for determining a thickness estimate of layers close to a pipeline wall.

These effects provide in turn several further advantageous effects: * it makes it possible to determine a thickness estimate of layers in a fluid flowing in a pipeline * it makes it possible to use for layers thicker than the sensitivity or penetration depth of the largest probe * using a single probe reduces the complexities, cost and number of intrusions in the pipe 3*

Brief Description of the Drawings

The above and further features of the invention are set forth with particularity ri the appended claims and together with advantages thereofwdl become clearer from considerabon of the following detailed descriphon of an Iexemplawl embothment of the invention given with reference to the accompanying drawnigs The invention wILl be further described below in connection wth exemplary embothmerits which are schematically shown in the drawings. wherein Fig 1 shows an embodtment of a probe mounted in pipetine Fig 2a shows probe measuring a Layer of thickness d and permittivity 4 backed by a material/fluid with permittivity 4 Fig 2b shows simulated reflection coefficient (absolute value) for some layer thicknesses Fig 3 shows an example of permittivity as a function of frequency for water oil and a 40% oil-inwater emulsion Fig 4 shows calculated apparent permittivity for layers of water Fig 5a shows the absolute value of the measured reflection coefficient for various layers O Figure Sb shows the absolute value of as a function of frequency for LIt) 20 various layers o Fig 5c shows inverse Fourier transform of the functional relationship R1g) Fig Sd shows inverse Founer transform normahzed to maximum peak value Fig 5e shows time corresponding to a peak in the inverse Fourier signal versus layer thickness Fig Sf shows estimated thickness versus reference thickness

Detailed Description

Various aspects of the disclosure are described more fully hereinafter with reference to the accompanying drawings This disclosure may, however, be embodied in many different forms and should not be construed as limited to any specific structure or function presented throughout this chsciosure Rather, these aspects are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art Based on the teachings herein one skilled in the art should appreciate that the scope of the disclosure is intended to cover any aspect of the disclosure disclosed herein, whether implemented independently of or combined with any other aspect of the disclosure For example, an apparatus may be implemented or a method may be practiced using any number of the aspects set forth herein In addition, the scope of the disclosure is intended to cover such an apparatus or method which is practiced using other structure, functionality, or structure and functionality in addition to or other than the various aspects of the disclosure set forth herein. It should be understood that any aspect of the disclosure disclosed herein may be embodied by one or more elements of a claim.

The invention will be further described in connection with exemplary embodiments which are schematically shown in the drawings, wherein Fig. 1 shows a sketch of a sensor system for measuring the permittivity close to the inner wall of a pipeline. The system operates by measuring the complex reflection coefficient from an open-ended coaxial probe which is in contact with the material under examination. The permittivity at a given frequency of the material close to the probe can be calculated from the measured amplitude and phase of the reflection coefficient at the given frequency, as the reflection coefficient depends on the fringe field response of the probe. The perniittivity is a frequency dependent material parameter, and the frequency spectrum of the perrnittivity is determined by sweeping the reflection coefficient over the frequency range of interest. The electromagnetic field of an open-ended coaxial probe when operated in non-radiating mode decays exponentially into the material under test, such that the reflection coefficient only senses materials close to the probe interface. The sensitivity depth of a probe is approximately equal to the inner radius of the outer conductor of the coaxial probe.

Principles forming the basis of the invention An alternative method for measuring the thickness of material layers is the short-pulse radar method (aka ground penetrating radar GPR). Related technologies may include near-field radar. Short-pulse radar is the electromagnetic analogue to sonic and ultrasonic pulse-echo methods. In this method a short pulse is transmitted from an antenna and is reflected at interfaces between materials with difference in permittivity. As a first order approximation, the transit time of the reflected pulse is at = L. where c is the speed of light and d is the thickness of the layer. Note that the complex permittivity of the material must be known in order to calculate the thickness from the transit time measurements. These applications typically use high-efficiency antennas such as horn antennas.

The invention consists of using a single probe for combined permittivity and thickness estimation. The permittivity is estimated by using the probe in a low frequency range (the ordinary frequency range of the probe), whereas thickness information is found by analysis of the response at higher frequencies in combination with application knowledge. The principles of the invention can be explained as follows: At high frequencies the probe starts to partly radiate electromagnetic signals, and will start to behave like a shod monopole antenna. These radiated signals penetrate much deeper than the fringe field that normally is used for permittivity measurements. As illustrated in figure 2(a), the radiated signals may be reflected from any impedance mismatches (e.g. layers of materials with different permittivity).

By analysing these reflections it is possible to calculate the thickness of the layer.

There are several potential methods for doing this calculation: 1. The short pu/se radar technique This method was described in last section, but has some drawbacks when combined with the open-ended coaxial probe. Preferable, a swept frequency signal is applied to the probe when measuring permittivity in the regular frequency range. The short pulse radar technique operates in the time-domain, and a complex electronic unit operating in both time and frequency domain is therefore required if the two techniques shall be combined.

2. Swept frequency analysis In this method a swept frequency signal is applied in the same manner as used for ordinary permittivity measurement with the open ended probe -but in a higher frequency range. The reflected signals from the material-backlayer interface will then interfere with the reflections from the probe-material boundary. This interference may be constructive or destructive depending on the actual frequency, thickness and layer permittivity and is seen as ringing in the measured reflection coefficient (see figure 2(b)). In previous work this ringing was considered to be problematic when calculating the permittivity of the layer. In contrast the present invention discloses a method and corresponding algorithms that show that it is possible to calculate the layer thickness by analysing the frequency of the ringing. This can be done either directly in the frequency domain or by inverse Fourier transformation into the time domain.

The measured response will depend on the propagation constant multiplied by the layer thickness (yd), where the propagation constant depend on the square root of the complex permittivity. For the fundamental frequency this can be described as:

-(2)

Equations including higher order modes are given in for instance equation 21-23 in (Folgerø et all 996). Thus, if the layer thickness d shall be calculated using equation (2) or similar, the permittivity of the material must be known. Note that since the permittivity changes with frequency, the permittivity must be known in the high frequency range in order to calculate the layer thickness by this method. This can be measured by a separate sensor (with small sensitivity depth), but in our invention we combine low frequency permittivity measurements with application knowledge to estimate the high-frequency permittivity. This ensures that both permittivity and layer thickness can be measured with a single sensor.

As an example, a wet-gas flow of oil, water and gas is considered. In this case the liquid layer consists of a water-in-oil or oil-in-water emulsion. The permittivity of the liquid layer is given by Bruggeman's equation (here shown for oil-continuous flow): -= -(3) --) As the permittivity of water and oil changes with frequency the resulting mixture permittivity will also change with frequency. In a typical application, the water content and the conductivity of the water can be calculated from permittivity measurements in the MHz range (cyan area in Figure 3) by using an appropriate open-ended probe.

The mixture's permittivity in the GHz range (green area in Figure 3) can then be estimated using the apriori known frequency dependency of water and oil. An open-ended probe with sensitivity range in the MHz range will typically be radiating in the GHz range, and thus introduce ripples in the measured reflection coefficients.

Analysis of these ripples together with knowledge of the mixture's permittivity will then give information about the layer thickness.

In summary:

(a) The open-ended probe is used for permittivity measurement in its low (ordinary) frequency range (b) The permittivity at higher frequencies is estimated from the measurements in the low (ordinary) frequency range by using application knowledge and adequate models (c) The probe is used as an antenna in its high frequency range to calculate the thickness*propagation constant, and the thickness can then be calculated as the propagation constant is known from (b) Best Modes of Carrying Out the Invention This section describes of possible embodiments of the invention according to Fig. 1 and 2.

1. Measurements in the low (ordinary) frequency range A sensor made of stainless steel and Peek as dielectric material will be used as example. The outer radius of the coaxial probe is 10mm. The apparent permittivity of a material under examination can be calculated using the methods explained in (Folgerø et al, 1996), e.g. the bilinear calibration procedure * _______

I -(4) r, p

Here A, B and r are calibration coefficients that characterize the probe. These coefficients are calculated from reference measurements of known standards (in this case known fluids such as e.g. air, saline water, and ethanol). These reference measurements can be done one-time and offline, such that a set of calibration coefficients can be reused when the probe is installed in an application.

Other models than the bilinear calibration procedure can be used to calculate the permittivity (see Folgerø 1996).

The calculation models that assume infinite thick layers can be used to calculate the apparent permittivity as long as any radiation from the probe is not interfering with the main reflection. In the following discussion we will refer to this frequency range as frequency range 1. Fig. 4 shows the permittivity of different layers of water as a function of frequency using the bilinear calibration procedure. It is seen that the permittivity is estimated well in the frequency range 100 MHz -1 GHz. At lower frequencies the response is distorted by noise. At higher frequencies (>-1 GHz), reflection from the water-air interface will give ripples in the calculated permittivity.

The black curve shows the permittivity of a very thick water layer (>20 mm). Thus, from these measurements we can calculate the apparent permittivity at frequencies below 1 GHz.

Typically, frequency range 1 is from 100 MHzto 1 GHz, but frequencies down to 100 kHz and up to 10 GHz may also be applicable depending on application and sensor geometry. Frequency range 2 is typically between 1 GHz and 20 GHz, but frequencies down to 100 MHz and up to 60 GHz may be applicable depending on application and sensor geometry. In some embodiments, frequency range 1 may overlap frequency range 2. In other embodiments, frequency range 1 may include frequency range 2.

2. Estimation of permittivity at extended frequencies The frequency range where reflections from the layer-backing material interfere with the first reflection from the probe-layer interface is defined as frequency range 2 (above a couple of GHz in Feil! Fant ikke referansekilden.). The lower frequency of frequency range 2 will depend on probe geometry, layer and backing material -thickness, and layer and backing material permittivity. Typically, the lower limiting frequency of frequency range 2 is between 1 and 10 GHz. When measuring on for instance wet-gas and slugging multiphase flow, a liquid layer backed by gas will be present in front of the probe. Consider the following special cases: a) The layer thickness is larger than the sensitivity depth of the probe In this case the permittivity will be described by the Bruggeman equation (or similar relationships). The permittivity of oil and water is known apriori. Water is well characterized, and the permittivity can be calculated if the conductivity and temperature is known (see e.g.). The permittivity of oil is known to typically be in the range 2-2.5 which in many cases is accurately enough.

Otherwise, it is possible to measure the oil permittivity directly or to give a good estimate based for instance on hydrocarbon composition (Folgerø, 2012). As the permittivity of oil and water is known (see e.g. Peyman 2007), Bruggeman's equation can be applied to calculate the water-fraction of the liquid film from the measured permittivity in frequency range 1.

Next step is to apply Bruggeman's equation to estimate the permittivity in frequency range 2. As the water-fraction in the liquid layer is known from the calculations above, the permittivity in frequency range 2 is calculated by inserting the apriori known water and oil permittivities in frequency range 2 into the Bruggeman equation. Note that this permittivity in general will vary with frequency as the water permittivity will vary within frequency range 2.

b) The layer thickness is smaller than the sensitivity depth of the probe In this case Bruggeman's equation cannot be used directly to calculate the water fraction as the measured (apparent) permittivity differs from the layer permittivity. However, if the thickness of the layer is assumed to be known, the empirical model or more complex models such as the full-wave model described in (Folgerø et al, 1996) can be used to calculate the layer permittivity. The same method as described in a can then be used to calculate the permittivity of the layer in frequency range 2.

As the thickness of the layer is not known, we suggest to use an iteratively method to calculate the thickness and permittivity in this case. In a first iteration a layer thickness is assumed. The permittivity in frequency range I is then calculated as described, the permittivity in frequency range 2 is estimated as described in a, and then the thickness of the layer is estimated using the algorithm in step 3 (below). The estimated thickness from step 3 is then used as input thickness for next iteration. This is repeated until the estimated thickness in step 3 converges.

A similar case to consider is when a deposit is growing in front of the probe. The same approach as described above may be applied here to estimate the permittivities in frequency range 2, but the permittivity models used will differ from Bruggemans. A thorough description of various efficient permittivity models can be found in (Sihvola).

3. Thickness estimation in frequency range 2 One possible way to determine the thickness from swept frequency measurements is described in the following, exemplified in Feil! Fant ikke referansekilden.. First we define the following parameters: * Measured reflection coefficient for a layer with thickness d and permittivity s * Rcç). A functional relation of the measured reflection coefficient The thickness can then be found by the following steps: 1) Measure reflection coefficient of layer with thickness d and calculate the functional relationship R(J): (see example in Feil! Fant ikke referansekilden.a) 2) Calculate the inverse Fourier transformation of the functional relationship R(1*) (see example in Feil! Fant ikke referansekilden.b) 3) Identify the times t.r corresponding to peaks in the time-domain signal (see example in Feil! Fant ikke referansekilden.c) Calculate the layer thicknesses from the peak information either by theoretical models, or from empirical relationships (see example in Feil! Fant ikke referanse ki I den. d).

An alternative method for estimating the thickness is to analyse the ripples directly in the frequency domain.

Alternative Embodiments A number of variations on the above can be envisaged. For instance: * The invention can also be used with other sensor types than coaxial probes.

Examples of other sensor types applicable are open ended circular or rectangular waveguides, planar patch antenna sensors, helical sensors, and horn antennas.

* Reflection measurements done using time-domain methods (Time Domain Reflectometry).

* Applying a modulated signal (e.g. chirp) as input for the thickness estimation is also a possibility.

* Measurements in the far field (instead of near field) is possible * Several sensors or sensor systems can be attached to the pipeline to characterize the layer or fluid film at several positions * The system may be applied to other applications than flow inside a pipeline

Industrial Applicability

The invention according to the application finds use in measurements of multiphase flows in pipelines.

References K. Folgerø and T. Tjomsland "Permittivity measurement of thin liquid layers using open-ended coaxial probes" Measurement, Science & Technology, vol. 7, pp 1164-1173, 1996 K. Folgerø Coaxial sensors for broad-band complex permittivity measurements of petroleum fluids", Dr. Science. Dissertation, 1996 K Folgerø, A L Tomren, S Frøyen "Permittivity calculator. Method and tool for calculating the permittivity of oils from PVT data", 30th Int. North Sea Flow Measurement Workshop, St. Andrews, Oct2012 K Folgerø, J Kocbach, "Inline measuring apparatus and method", PCT/N02011/000134 K Haukalid, K Folgerø "Measurements of water conductivity in oil continuous emulsions", 10th Int Conf on Electromagnetic Interaction with Water and Moist Substances, Weimar, Germany, Sept 25-27, 2013 J. Hilland, "Simple sensor system for measuring the dielectric properties of saline solutions," Measurement Science and Technology, vol. 8, no. 8, pp. 901-910, 1997.

0 lsaksen <<lnnretning for máling av refleksjonskoeffisienten til høgfrekvente elektromagnetiske bølgjer i vske, samt fremgangsmáte for a bestemme vassinnhold i fleirfasestraum ved bruk av innretninga>>, NO19971025 T. Jakobsen and K. Folgerø "Dielectric measurements of gas hydrate formation in water-in-oil emulsions using open-ended coaxial probes" Measurement, Science & Technology, vol. 8, pp 1006-1015, 1997 Baker-Jarvis J, Janezic M D, Domich P D and Geyer R G 1994 Analysis of an open-ended coaxial probe with lift-off for nondestructive testing IEEE Trans Instrum. Meas. 43711-18 Gerardo 0. Clemena "Short-Pulse Radar Methods" in "Handbook on Nondestructive Testing of Concrete," edited by V. M. Malhotra, Nicholas J. Carino A. Peyman, C. Gabriel, and E. H. Grant, "Complex permittivity of sodium chloride solutions at microwave frequencies," Bioelectromagnetics, vol. 28, no. 4, pp. 264-274, 2007.

A. H. Sihvola and Institution of Electrical Engineers, Electromagnetic Mixing Formulas and Applications. Institution of Electrical Engineers, 1999.

Claims

Claims 1. A system for characterising layers close to a pipeline wall, comprising: a sensor operating in a non-radiating mode in a first frequency range and operating in a partly radiating mode in a second frequency range, wherein the non-radiating mode being defined as when reflections from a first layer interface is dominating the reflection coefficient, wherein the radiating mode being defined as when reflections from a second interface of the layer interfere with reflections from a first interface.
2. The system according to claim 1, wherein the sensor is an open-ended coaxial probe.
3. The system according to claim 1, wherein the sensor is an open-ended circular waveguide.
4. The system according to claim 1, further comprising a signal generator for emitting a short pulse for use with short pulse radar technique.
5. The system according to claim 1, further comprising a signal generator for emitting a swept frequency signal generation.
6. The system according to claim 1, wherein the signal generator is arranged for emitting a short pulse for use with short pulse radar technique.
7. A method for characterising layers close to a pipeline wall, using a system according to claim 1 comprising: emitting a signal in a first frequency range, determining a complex reflection coefficient determining a value for permittivity, emitting a signal in a second frequency range, determining a complex reflection coefficient, and determining a value for layer thickness, characterized in that for the first frequency range the probe operates in a non-radiating mode, for the second frequency range the probe operates in an at least partly radiating mode, the permittivity in the second frequency range is determined from the permittivity in the first frequency range and the layer thickness is determined using a value for propagation constant determined from the value for permittivity in the first frequency range.
8. The method according to claim 7, wherein radiation from the probe is not interfering with the main reflection and calculation model assume infinite thick layers can be used to calculate the apparent permittivity.
9. The method according to claim 8, wherein the first frequency is in the range kHz to 10 GHz.
10. The method according to claim 8, wherein the first frequency is in the range lOMHztol GHz.
11. The method according to claim 7, wherein reflections from the layer-backing material interfere with the first reflection from the probe-layer interface.
12. The method according to claim 11, wherein layer thickness is larger than the sensitivity depth of the probe, wherein Bruggeman's equation is used to estimate the perm ittivity.
13. The method according to claim 11, wherein layer thickness is smaller than the sensitivity depth of the probe, wherein (describe method here) is used to estimate the permittivity.
14. The method according to claim 11, wherein the first frequency is in the range 1 GHztolOGHz.
15. The method according to claim 7, wherein the signal in the second frequency range is a short pulse, wherein the value for thickness is determined using short pulse radar technique with time of flight calculations.
16. The method according to claim 7, wherein the second frequency is a swept frequency, wherein the value for thickness is determined using swept frequency analysis.
17. The method according to claims 14 -16, wherein the value for layer thickness is determined using frequency of ringing for the reflection.
18. The method according to claims 14 -16, wherein the value for layer thickness is determined using the following steps: a: determine * r,: a measured reflection coefficient for a layer with thickness d and permittivity * R(1.. a functional relation of the measured reflection coefficient b: measure reflection coefficient of layer with thickness d and calculate the functional relationship R(l') (see example in Feil! Fant ikke referansekilden.a) c: calculate the inverse Fourier transformation of the functional relationship d: identify the times corresponding to peaks in the time-domain signal e: calculate the layer thicknesses from the peak information either by theoretical models, or from empirical relationships.