GB2302408A

GB2302408A - Capacitance measurement

Info

Publication number: GB2302408A
Application number: GB9612642A
Authority: GB
Inventors: Wuqiang Yang
Original assignee: PROCESS TOMOGRAPHY Ltd
Current assignee: PROCESS TOMOGRAPHY Ltd
Priority date: 1995-06-16
Filing date: 1996-06-17
Publication date: 1997-01-15
Anticipated expiration: 2016-06-17
Also published as: GB2302408B; GB9612642D0

Abstract

A method and apparatus for measuring the capacitance between combinations of electrodes in a multi-electrode array such as a capacitance tomography system for investigating multi-phase flows comprises the steps of applying excitation signals to each electrode and deriving measurement signals representative of inter-electrode capacitances from the same electrode in each case. Each electrode to which an excitation signal is applied is connected to one input of a first operational amplifier 24,25, to the other input of which an excitation signal source v i ,-v i is connected. The measurement signal is derived from the output of a second operational amplifier 27,28, one input of which is connected to the output of the first operational amplifier and the other input of which is connected to the excitation signal source.

Description

CAPACITANCE MEASXMENT The present invention relates to a method and apparatus for measuring the capacitance between combinations of two or more electrodes.

Electrical capacitance tomography (ECT) sensors are well known and are used to produce a visual representation of a physical process, for example a visual representation of the distribution of a two component fluid flow in an oil pipeline. ECT systems produce cross-sectional images of a flow based on variations in the permittivity of a material flowing through or located within a sensing assembly. Conventional sensing assemblies consist of an array of electrodes mounted either inside or outside an insulating pipe. For example in one known system twelve equally spaced electrodes are distributed around the circumference of a pipe.

In a conventional ECT system, only one of the electrodes in the array is used in turn as an excitation electrode, the other electrodes acting as detection or idle electrodes. The number of independent measurements that can be made with such an arrangement is N(N1)/2, where N is the number of electrodes. Thus in the case of a 12-electrode ECT sensor excited in this manner it is only possible to produce 66 independent measurements.

Conventional ECT systems exhibit relatively low detection sensitivity at the centre of the electrode array which corresponds to the centre of the vessel within which measurements are being made. Generally all of the electrodes other than the single excitation electrode are held at a fixed potential and as a result field lines fan out from the excited electrode. The field density at the centre of the array is relatively weak and hence the sensitivity in this region is low. The signal to noise ratio and hence image quality in the central area could be improved by changing the excitation arrangement to produce a modified electric field inside the sensing array.This could be achieved for example by exciting half of the electrodes with an appropriate potential and by using the remaining electrodes as detection electrodes with each detection electrode being maintained at an appropriate complimentary potential to that of the opposite excitation electrodes.

It must be appreciated that if the cross section of a pipe is represented by a square grid of only 32x32 pixels, there will be typically 812 pixels within the pipe. To determine the value of each of these image pixels to a resolution of one part in 256 means that in effect approximations to 207,872 solutions must be obtained from the less than 100 measurements.

The solution of the image calculation problem is therefore highly undetermined. If more measurements can be obtained, a more accurate solution for the image will be obtained.

A number of methods can be used to measure the inter-electrode capacitances.

However, all current methods use electronic or other switches to select and excite the electrodes. Great care must be taken to minimise the effects of stray capacitance to earth on the measurement of the inter-electrode capacitances since the capacitances to be measured are themselves very small. Measurement errors can arise because of problems associated with leakage through the switches.

It is an object of the present invention to obviate or mitigate the problems outlined above.

According to the present invention there is provided a method for measuring the capacitance between two or more electrodes, wherein an excitation signal is applied to at least one of the electrodes, and a measurement signal representative of the capacitance is derived from the electrode to which the excitation signal is applied.

The invention also provides an apparatus for measuring the capacitance between two or more electrodes, comprising means for applying an excitation signal to at least one of the electrodes, and means for deriving a measurement signal representative of the capacitance from the electrode to which the excitation signal is applied.

The present invention enables the number of measurements to be taken for 9 given number of electrodes to be substantially increased, thereby enabling greater measurement accuracy. Furthermore the present invention also enables the pattern of fields between the electrodes to be varied as between one measurement and another which allows an increase in accuracy and sensitivity. Given that each electrode can be used as a source of a measurement signal as well as a point to which an excitation signal can be applied, it is no longer necessary to operate most of the electrodes simply as passive detection electrodes. Hence the number of measurements which can be made for a given electrode configuration can be substantially increased.

Preferably excitation signals are applied to each electrode of an array and measurement signals are derived from each of those electrodes.

Preferably a guard electrode is associated with each capacitance electrode, the signal applied to each capacitance electrode also being applied to the associated guard electrode.

In one circuit arrangement, each electrode to which an excitation signal is applied is connected to one input of a first operational amplifier to the other input of which an excitation signal source is connected, and the or each measurement signal is derived from the output of the second operational amplifier one input of which is connected to the output of the first operational amplifier and the other input of which is connected to the excitation signal source.

The one input of the first operational amplifier is preferably directly connected to the electrode to which the excitation signal is applied to avoid any risk of spurious signal injections.

In a circuit arrangement incorporating operational amplifiers as described above a guard ring may be provided which is connected to the excitation signal source.

It will be appreciated that the invention eliminates the need for switches and hence avoids problems with switch leakage as referred to above.

An array of equally spaced electrodes may be provided with each electrode in the array being connected to a respective circuit capable of applying an excitation signal and deriving a measurement signal from the electrode to which it is connected. A series of measurement signals may be derived from each of the electrodes with the excitation signals applied to the electrodes being varied during the series to alter the pattern and the direction of the field between the electrodes.

Embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 illustrates schematically the structure of the conventional 12-electrode ECT sensor; Figure 2 shows a simplified circuit diagram of a known ECT system which is used to monitor the capacitance between any two of the electrodes shown in figure 1; Figure 3 illustrates a circuit which can be used to put into effect the method in accordance with the present invention; Figure 4 illustrates a circuit similar to that of figure 3 but incorporating guard electrodes to prevent stray capacitance affecting the system; Figure 5 illustrates an equivalent circuit to the structure shown in figure 4; Figure 6 illustrates the overall structure of an ECT 12-electrode sensor utilising circuits of the type illustrated in figures 3 to 5;; Figure 7 illustrates one arrangement for varying the excitation signals to a circuit of the type illustrated in figure 6; and Figures 8, 9, 10 and 11 illustrate alternative electrode connections to indicate some of the wide range of electric field distributions which can be established in accordance with the invention inside a sensor of the general structure shown in figure 1.

Referring to figure 1, twelve electrodes indicated by reference numerals 1 to 12 are arranged around the outside of a pipe 13 the interior 14 of which will in use carry for example a 2-component fluid flow. In the illustrated arrangement, one of the electrodes is connected to an excitation source and each of the other electrodes is connected to a detector. Figure 2 illustrates part of the resultant circuit in schematic form. Referring to figure 2, a terminal 15 is connected to a signal source and a terminal 16 is connected to a measurement circuit. A capacitor 17 represents the capacitance between two of the electrodes shown in figure 1. For example, the upper half of the capacitor 17 could represent the electrode 3 and the lower could represent the electrode 9.The upper electrode of the capacitor is initially selected as an excitation electrode by closing switch 18 and opening switch 19 and the other electrode is connected to the detector by closing switch 20 and opening switch 21. This connection can be reversed by appropriate control of the switches 18 to 21. This known arrangement involving electronic switches can be vulnerable to measurement errors because of problems associated with leakage, including charge injection effects and spurious coupling capacitance.

As the measured capacitance in an ECT system, and especially the capacitance change which must be accurately detected, is very small, of the order of I0-1'F, and the switch equivalent capacitance is relatively large, of the order of 10. 12F, any leakage of charge through the switch causes measurement inaccuracies.

Referring now to figure 3, this shows a circuit which can be used to implement the present invention. The illustrated circuit is AC-based, that is it relies upon applying an alternating signal, for example a sinusoidal signal. Terminals 22 and 23 are connected to sine-wave sources with the same amplitudes but in anti-phase, and these terminals are connected to the positive terminals of two operational amplifiers 24 and 25.

Although in this described example sinusoidal signals are applied to the electrodes, any appropriate waveform could be applied, for example a square wave. The gain-setting feedback components for the operational amplifiers are shown as capacitors Cf, although resistive elements will also be needed in practice to set an appropriate DC gain at a finite value. A capacitor 26 represents the capacitance Cx defined between for example electrodes 3 and 9 of figure 1. The left hand electrode of capacitor 26 is held at the potential applied to the terminal 22 and the right hand electrode of capacitor 26 is similarly held at the potential applied to terminal 23.The potential applied to terminal 22 is Vi = A sin cot and the potential applied to terminal 23 is the complementary potential -V*. The two electrodes of the capacitor are thus excited simultaneously. Considering the right hand side of the circuit shown in figure 3, the output of the amplifier 25 is given by:

Generally the unknown capacitance Cx will be much smaller than Cf, and therefore Vl is dominated by the standing value -Vj. This value is removed by adding it to its inverse value using a further operational amplifier 27 the output V2 of which is given by:

Therefore the output of the right hand side of the circuit of figure 3 is proportional to the unknown capacitance Cx.

The left hand side of the circuit of figure 3 incorporates an output amplifier 28 which produces an output voltage V3 given by:

Thus the outputs V2 and V3 are proportional to the capacitance between the two electrodes to which the circuits are connected. Although in the simplified case represented in figure 3 it is assumed that there are only two electrodes, where there are three or more electrodes each of which has connected to it a circuit of the type shown to the left or right of figure 3, the capacitance "seen" from one electrode will be different from the capacitance "seen" from other electrodes as there will be a capacitance defined between every pair of electrodes which are not held at the same voltage.Thus in a complex system signals will be extracted from the electrode which are not simple mirror images of each other as is the case in the circuit shown in figure 3. Thus the circuit of figure 3 enables excitation signals to be applied to and measurement signals be extracted from each of the electrodes of the complex system simultaneously.

An alternative arrangement can be provided which allows the detection electrodes to remain at ground potential. This can be achieved by offsetting the potentials of all the electrodes by the same voltage such that the detection electrodes are at ground potential. This is at the expense however of not being able to measure the capacitances between all the electrodes simultaneously.

The circuit illustrated in figure 3 will be vulnerable to stray-immune effects as it does not incorporate a guarding scheme. Figure 4 shows an arrangement corresponding to that of figure 3 but incorporating guard electrodes 29 and 30. The left hand guard electrode 29 is directly connected to the terminal 22 and the right hand guard electrode 30 is directly connected to the terminal 23. The guard electrodes are themselves located inside an earthed screen 31. Figure 5 schematically represents the equivalent circuit for the structure of figure 4. Capacitances Csi and Cs2 represent stray capacitances between the measurement electrode and the guard electrode, and capacitances Cs3 and Cs4 representing the stray capacitances between the guard electrodes and the earth screen. Since Csi and Cs2 are driven by the same sine-wave sources as those used to set the potentials on the electrodes of the unknown capacitor Cx, the potentials across Cs and C52 are zero. Therefore there are no currents through C51 and C52 and they do not affect the capacitance measurement. Furthermore, as Cs3 and C54 are directly driven by the sine-wave sources, they draw current from the sources only and hence do not affect the measurement of Cx. Therefore the illustrated circuit is strayimmune.

Figure 6 illustrates an ECT sensor based on the principles described with reference to figures 3, 4 and 5. Each of the twelve electrodes has its own associated guard electrode located radially outside the measurement electrode. Twelve independent measuring channels are connected to the electrode array, each of the channels corresponding to one of the two circuits shown in figures 3, 4 and 5. Each channel is connected to a respective measurement electrode and a respective driven guard electrode. The electric field inside the sensor depends on the excitation signals Vik (k=12...12) and can be changed arbitrarily by applying different potentials to the electrodes. Capacitance measurements between all other electrodes in the system are carried out continuously and simultaneously for each field configuration.

Figure 7 illustrates one circuit for varying the excitation signal applied to one electrode by one pair of operational amplifiers. A micro-processor 31 controls a digital to analogue (DAC) converter 32 which in turn controls operational amplifiers 33 and 34. A sine-wave source is connected to terminal 35, and thus the output of the amplifier 34 corresponds to the signal applied to for example electrode 23 of figure 3. The magnitude of that signal is a function of the output of the DAC 32 which in turn can be controlled by the micro-processor 31.

Figures 8 to 11 illustrate some of the excitation signal patterns which are possible using the circuitry of figure 6 and 7. Each pattern results in a different field distribution in the sensor. Figure 8 shows a pattern corresponding to a conventional ECT excitation arrangement where only one electrode at a time is set to be an excitation electrode and all of the other electrodes are held at the same zero potential and used for detection purposes only.

Such an excitation arrangement can produce 66 independent measurements given that there are 12-electrodes. Figure 9 shows how two electrodes at a time are in effect combined by applying to them the same excitation signal, all the other electrodes being held at zero potential. With such an arrangement 120 extra measurements can be obtained. Given that electrodes can be combined in groups of 3, 4...1 1, many more measurements can be obtained by appropriate selection of electrode groups. Figure 10 shows how a substantially parallel field can be set up inside the sensor by applying different potentials to pairs of electrodes.

Such an arrangement can produce 72 further measurements, that is 12 individual capacitance measurements for each of six rotations.

In the alternative arrangement described above in which the detection electrodes are at ground potential, all the illustrated potentials shown in Figure 10 would be increased by 15 volts when the bottom two electrodes are being used as detection electrodes such that both the detection electrodes would be at ground potential. Figure 11 shows how a quasi-parallel field can be set up inside the sensor by applying appropriate voltages to the electrodes on one side of the sensor whilst the remaining electrodes are held at zero potential. Such an arrangement can produce 72 extra measurements, that is 6 measurements for each of 12 rotations.

Thus it will be apparent from figures 8-11 that the present invention enables many more measurements to be made as compared with conventional ECT structures.

Claims

1. A method for measuring the capacitance between two or more electrodes, wherein an excitation signal is applied to at least one of the electrodes, and a measurement signal representative of the capacitance is derived from the electrode to which the excitation signal is applied.

2. A method according to claim 1, wherein excitation signals are applied to each electrode and measurement signals are derived from each electrode.

3. A method according to claim 1 or 2, wherein a guard electrode is associated with each capacitance electrode, the signal applied to each capacitance electrode also being applied to the associated guard electrode.

4. A method according to any preceding claim, wherein the or each electrode to which an excitation signal is applied is connected to one input of a first operational amplifier to the other input of which an excitation signal source is connected, and the or each measurement signal is derived from the output of a second operational amplifier one input of which is connected to the output of the first operational amplifier and the other input of which is connected to the excitation signal source.

5. A method according to claim 4, wherein the said one input of the first operational amplifier is directly connected to the electrodes to which the excitation signal is applied.

6. A method according to anyone of claims 4 to 5 as dependent on claim 3, wherein the guard ring is connected to the excitation signal source.

7. A method according to any preceding claim, comprising an array of three or more electrodes each of which is connected to a respective circuit capable of applying an excitation signal to and deriving a measurement signal from the electrode to which it is connected.

8. A method according to claim 7, wherein a series of capacitance measurement signals are derived from the electrode with the excitation signals applied to the electrode being varied during the series to alter the pattern and the direction of the field between the electrodes.

9. A method according to any preceeding claim, wherein the potential of each electrode from which a measurement signal is derived is held at ground potential.

10. An apparatus for measuring the capacitance between two or more electrodes, comprising means for applying an excitation signal to at least one of the electrodes, and means for deriving a measurement signal representative of the capacitance from the electrode to which the excitation signal is applied.

11. A method for measuring the capacitance between two or more electrodes substantially as hereinbefore described with reference to figures 3 to 11 of the accompanying drawings.

12. An apparatus for measuring the capacitance between two or more electrodes substantial as hereinbefore described with reference to figure 3 to 11 of the accompanying drawings.