Field of the Invention

The invention is concerned with mass flow monitoring and, in particular, the monitoring of flows of materials in pipelines.

Background to the Invention

Tomographic techniques based on both Electrical Resistance Tomography (ERT) and Electrical Capacitance Tomography (ECT) have become established methods for analysis & visualizing the flows of liquid and particulate (pneumatically & hydraulically conveyed) materials in pipelines. Examples of these techniques are described in Patent Document No.

WO2016/038391 A1.

Application areas include the assessment of multi-phase flows, commonly found in many industrial processes, such as oil and gas production and processing, along with many other petrochemical and chemical processing steps. Another significant area of interest is in industrial processing is the hydro transport of particulate matter in the form of a slurry. This application covers the minerals processing area, oil sand processing and the dredging industries. In the measurement of such slurry flows, the amount of material transported by the‘hydro-suspension’ is of great interest. The use of Coriolis meters are often employed to measure‘mass flow’ rates in industrial process control applications, however, for the very large pipeline cross-sections used in hydro transport applications, Coriolis technology becomes prohibitively expensive, if not technically impracticable. Consequently, the flow velocity and density of the flow are typically measured separately and the mass flow inferred from the two meters. In this case, the velocity of the flow stream can be determined via the use of standard flow meters such as magnetic flow meters, but to determine the amount or mass of solids transported, the effective density of the flow stream has to be measured using another technique, including nuclear (gamma) density meters. A range of other technologies have been developed for the measurement of flow stream density, including a physical‘weight’ based concept [A Non-Nuclear Density Meter & Mass Flow System for Dredging Slurries” Proceedings from WEDA XXXII Technical Conference, 2012], approaches based on

acoustic/vibrational [Density Meter offers Non-nuclear Measurement Option” Engineering and Mining Journal 12 ^th Oct 2015] and ultrasonic probing

[“Rhosonics launches slurry density meter” Mining Magazine, 3 ^rd June 2016] and electrical tomography based detection. Of these approaches, electrical tomography is particularly complementary in its nature to magmeter configurations, as they both involve the use of electrodes to probe and detect induced voltages in the flow. A distributed magflow technique is also disclosed in WO2011128656A1.

For the measurement of flow rate, magnetic flow meters are often selected because they are obstruction-less, cost-effective for aggressive slurries and provide highly accurate volumetric flow measurement. Through multiple suppliers, a range of liner materials, electrode options and line sizes accommodate a wide range of applications.

Additionally, magnetic flow meters are beneficial as they are effective for both very low and high volume flow rates and are relatively immune to changes in process variables. In a magnetic flow meter, the operation is based on Faraday's law of electromagnetic induction, and the voltage generated between two electrodes placed diametrically across a flow stream of diameter D, is given by:

Dί/ = BCVD where Ail is the voltage difference between the pair of electrodes, B is the effective magnetic flux density (Wb), C is the sensitivity (dimensionless), V is the mean velocity in the pipe (m.s ^'1 ) and D is diameter of the pipe (m).

‘Mag’ flowmeters have the advantage that measures volume rate

independent of the effect of viscosity, density, turbulence, or suspended

material of the flow. WO2014028450 discloses the application of a rotating magnetic field across a fluid flowing in a pipe, tank, cell or vessel to determine a flow analysis

based at least partly on the signaling received

Statements of the Invention According to a first aspect of the present invention there is provided a mass flow monitoring system, the system comprising: one or more processors; a magnetic source operable to selectively provide a magnetic field through a section of flow conduit to be monitored; and, an electric source comprising a plurality of electrodes arranged around the circumference of the section of said flow conduit to provide conductive paths therein; wherein the one or more processors is operable to select an electrode-pair, from the plurality of electrodes; and, wherein, in a flow density mode, the processor is operable to apply an electrical signal across the electrode-pair and measure the responsive electrical signal across one or more other electrode-pairs; and in a flow rate mode, the processor is operable to cause the magnetic field and the plurality of conductor paths to be angularly displaceable relative to each other, and to measure the responsive electrical signal of a selected electrode-pair; and compute the measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals to determine the mass flow through said section of conduit.

The one or more processor advantageously selects the one or more electrode-pair from the plurality of electrodes in a predetermined sequence.

In the flow rate mode, the responsive electrical signal is advantageously measured from the electrode-pair orthogonal to the direction of the magnetic field.

The magnetic field source advantageously comprises a plurality of coils arranged around the plurality of electrodes and the direction of the magnetic field is

advantageously rotatable to provide angular displacement relative to the plurality of conductor paths, in Wangle by adjustment of the field intensity in the coils, as

Hi=Hocos(^-tk^, where Ho is the maximum amplitude of magnetic field, fi is the magnitude of excitations applied to the number of coils, d is the minimum rotation angle and k is step of rotation.

The magnetic field source advantageously comprises a ferromagnetic stator.

The one or more processor is advantageously operable to compute the measured flow rate mode responsive electric signals using a velocity field weight matrix and a voltage difference vector.

The velocity distribution vector is preferably derived from:

Dϋ = 5CVD where AU is the voltage difference vector between electrode-pairs, B is the effective magnetic flux density, C is the sensitivity weight matrix, V is the local velocity vector and D is the distance vector between relevant measurement electrode-pairs.

The electric source advantageously comprises a second plurality of electrodes arranged around the circumference of the section of said flow conduit to provide conductive paths therein, the second plurality of electrodes being electrically insulated from the first plurality of electrodes.

In the flow density mode, the processor advantageously operable to apply a high frequency modulated excitation signal, and, in the flow rate mode, the processor is advantageously operable to apply DC or low frequency modulated excitation signal.

The measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are advantageously frequency or time multiplexed.

The measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are advantageously demodulated by one of analogue frequency filtering, digital phase and frequency sensitive demodulation.

According to a second aspect of the present invention there is provided a method of monitoring mass flow, the method comprising, providing: one or more processors; a magnetic source operable to selectively provide a magnetic field through a section of flow conduit to be monitored; and, an electric source comprising a plurality of electrodes arranged around the circumference of the section of said flow conduit to provide conductive paths therein; selecting an electrode-pair, from the plurality of electrodes; in a flow density mode, applying an electrical signal across the electrodepair and measuring the responsive electrical signal across one or more other electrode-pairs; in a flow rate mode, causing the magnetic field and the plurality of conductor paths to be angularly displaceable relative to each other, and measuring the responsive electrical signal of a selected electrode-pair; and computing the measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals to determine the mass flow through said section of conduit.

The one or more electrode-pairs are advantageously selected from the plurality of electrodes in a predetermined sequence.

In the flow rate mode, the responsive electrical signal is advantageously measured from the electrode-pair orthogonal to the direction of the magnetic field.

The magnetic field source advantageously comprises a plurality of coils arranged around the plurality of electrodes and the direction of the magnetic field is

advantageously rotatable, to provide angular displacement relative to the plurality of conductor paths, in kd angle by adjustment of the field intensity in the coils, as

Hi=H(fios( -f S), where H ₀ is the maximum amplitude of magnetic field, fi is the magnitude of excitations applied to the number of coils, d is the minimum rotation angle and k is step of rotation. The magnetic field source is advantageously provided with a ferromagnetic stator.

Computing the measured flow rate mode responsive electric signals advantageously comprises using a velocity field weight matrix and a voltage difference vector.

The velocity distribution vector is advantageously derived from:

Dϋ = BCVO where AU is the voltage difference vector between electrode-pairs, B is the effective magnetic flux density, C is the sensitivity weight matrix, V is the local velocity vector and D is the distance vector between relevant measurement electrode-pairs.

The electric source advantageously comprises a second plurality of electrodes arranged around the circumference of the section of said flow conduit to provide conductive paths therein, the second plurality of electrodes being electrically insulated from the first plurality of electrodes.

In the flow density mode, a high frequency modulated excitation signal is

advantageously applied, and, in the flow rate mode, a DC or low frequency modulated excitation signal is advantageously applied.

The measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are advantageously frequency or time multiplexed.

The measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are advantageously demodulated by one of analogue frequency filtering, digital phase and frequency sensitive demodulation.

The present invention combines the magneto-electrical sensing of flow

velocity using Faraday’s induction law, and the determination of flow density via tomographic sensing. This combined measurement allows the mass flow rate to be determined. The systems can be co-located in the same

measurement plane, allowing the length of the measurement unit to be no

longer than a conventional magmeter unit.

Both approaches utilize electrodes in contact with the flow stream, but the

signals generated by the two systems are electrically separated by using

either i) modulation frequency separation techniques (frequency multiplexing), ii) time interleaving (time multiplexing) iii) spatially separating the

measurements such that their respective fields do not overlap, whilst still

benefiting from manufacturing efficiencies.

Figure 1 illustrates the concept: Here, a series of electrodes that contact the flowing fluid are placed around the circumference of the pipeline. The

electrodes are connected to tomography processing unit, that sequentially applies a set of probing voltages /currents to pairs of electrodes and detects the peripheral voltages generated across the other electrode pairs as per the tomographic processing as defined in reference [1-3]. This processing is activated for a period of time T _tomo . T _tomo could be as short as 100ms or less, or more typically in the range of 250ms to 1 second.

The system then switches to the magflow determination. The H field is activated via a set of electromagnet coils. This field could be a static field, or more commonly, an AC driven field, or‘switched’ +/- DC field. The induced voltage generated across the primary pair of electrodes used for velocity sensing (the diametrically opposing pair, perpendicular to the magnetic field direction, H, as designed by e _mag and e _mag This measurement is conducted for an interval T mag _, where T _mag is in the range of as short as 100ms or less, or more typically in the range of 250ms to 1 second.

The flow velocity determined by the processing unit is fed, along with the density determined via the tomographic processor to a‘totalizer’ unit that calculates the product of these input quantities to produce a measure of mass flow rate.

For a Ttomo and T _mas the ranges described above, this would produce a total measurement time of ~ 200mS or less, or 500mS to 2 S. If the induced voltage across a set of electrodes is nulled out by driving a certain current back through the electrode pair so as to produce a voltage - AU, then this will become a form of tomographic‘excitation’ signal, and the other electrode pairs in the system will 'see' the influence of this to a degree determined by the local conductivities in the flowing media. The flow rate can be determined by monitoring -AU at the null point as a measure of the magnetically induced flow voltage. If this is repeated for different pairs of electrodes in differing places, including on a different radial axis through the cross sectional plane of the pipe/conduit, the result will be many signals

generated across the other electrode pairs that can be input to a tomographic processing system. The signals will be a composite of mag-induced and ERT induced, but these may be separated.

Detailed description of the Invention

Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which Figure 1 is a schematic of a mass flow monitoring system according to the present invention;

Figures 2a and 2b are schematics of cross sections through the electrode

arrangement of the system of Figure 1, in a flow density mode;

Figures 3a and 3b are schematics of cross sections through the electrode

arrangement of the system of Figure 1 , in a flow rate mode;

Figure 4 is a schematic of a cross section through the magnetic field source arrangement of the system of Figure 1 ; and

Figure 5 is a schematic of a cross section through the magnetic field source and electrode arrangement of the system of Figure 1. Referring to Figures 1-5, a mass flow monitoring system 10 comprises an electric source comprising a sensor arrangement 12 having a plurality of electrodes 14i-14i2 disposed in a spaced apart arrangement around the circumference of a conduit 16 such that the electrodes 14 14 _x are exposed to flow of material through the conduit 16. Although, the sensor arrangement is illustrated by example having twelve electrodes r x, other numbers of electrodes may be used, as appropriate, such as, for example, sixteen electrodes are illustrated in Figure 5.

The electric source also comprises a flow density monitoring processor 18 having a controller, a current source (not shown) and a voltage monitor (not shown). The controller controls the current source and the voltage monitor. The current source is electrically connected to each of the electrodes 14 14 _x to selectively apply an energizing electric signal to selected electrode pairs and thereby provide a conductive paths between electrode pairs extending across the conduit 16 through which material flows. The voltage monitor is also connected to each of the electrodes 14i-14 _x to monitor the responsive electrical signal of selected electrode pairs.

The system 10 further comprises a magnetic source 20 operable to provide a uniform parallel magnetic field through the conduit 16.The magnetic source 20 comprises magnetic coils 22 wound on a stator 24 disposed around the external surface of the conduit 16 (see Figure 5) to, in use, provide a magnetic field H. The magnetic source 20 further comprises a flow rate monitoring processor 26. The flow rate monitoring processor 26 is electrically connected to an electrode-pair. For example, in the embodiment of Figure 1 the flow rate monitoring processor 26 is electrically connected across the electrode-pair 14 ₄ - 14i ₀ to monitor the responsive electrical signal between that electrode pair. The magnetic source 20 is operable to cause the magnetic field H and the plurality of conductor paths to be angularly displaceable relative to each other. For example, in use, the magnetic field H is rotated relative to the conductive paths of selected electrode-pairs in Wangle by adjustment of the field intensity in the coils 22, as Hj=H(fios(<fc-tkS), where H ₀ is the maximum amplitude of magnetic field, is the magnitude of excitations applied to the number of coils, d is the minimum rotation angle and k is step of rotation. In use, as material flows through the conduit 26, with the system 10 in a flow rate mode, the magnetic source 20 provides a uniform magnetic field H through the conduit 16 and rotates the magnetic field H relative to the plurality of electrodes 14i-14 _x . As the magnetic field H rotates, the flow rate monitoring processor 26 sequentially selects the electrode-pair orthogonally orientated relative to the direction of the magnetic field H and sequentially measures the responsive electrical signal of each selected orthogonally orientated electrode-pair to provide a dynamic mean velocity profile representation of the material flowing through the conduit 16.

The flow rate monitoring processor 26 computes the flow rate of the material flowing through the conduit 16 using a velocity field weight matrix and a voltage difference vector.

With the system 10 in a flow density mode, the flow density monitoring processor 18 sequentially selects an energized electrode-pair from the plurality of electrodes 14i- 14 _x . The current source energizes the selected energized electrode-pair and the voltage monitor measures the responsive electric signal of one or more of other selected responsive electrode-pairs. The flow rate monitoring processor 18 then processes the measured responsive electric signals, using known tomography processing, to determine a flow density profile of the material flowing through the conduit 16.The system 10 rapidly switches between the flow rate mode and the flow density mode to monitor the flow rate and the flow density of the material flowing through the conduit 16. The measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are frequency or time multiplexed.

The measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are demodulated by one of analogue frequency filtering, digital phase and frequency sensitive demodulation. The demodulated measured flow density mode responsive electric signals and the measured flow rate mode responsive electric signals are computed to provide the mass flow of material flowing through the conduit 16.

The system 10 described may be implemented in the following alternative

embodiments, which are presented as examples and not an exhaustive list of potential utility:

Simultaneous determination of flow velocity induced voltage and all the tomographic processing based on frequency multiplexing concepts.

Measurement of the Faraday induced flow voltage across multiple pairs of electrodes: This provides additional flow data that provides more information about the flow, particularly in stratified flows. This may be implemented in;

A horizontally structured measurement mode, (see Figure 2), or

In a radially-rotational fashion, (see Figure 3). In each of these cases, the tomography processing system could be adapted to separate out the tomographic signals from those generated by a modulated magnetic field.

Use of tomographic algorithms for deciphering the composite signal generated across various electrode sets by a single set of processing electronics.

Advantages of the use of the present invention include: Mass flow can be determined using the same flowmeter‘footprint’ as required for a standard Magmeter

The distribution of solids in the flow can be visualized to allow determination

of flow distributions, particularly to allow a the assessment of standing beds Once the flow distribution is determined, this information can be used to

determine the relative velocities (slip) of each phase

On a similar basis, oil / water flow regimes may be determined

The relative flows (slip) of oil and water can also be determined In the case of hydraulic transport of slurries, the operation of pumps may be optimised by determining the minimum velocities that all solids are mobilised, thus reducing fuel consumption and minimising wear on pump components

and piping

With the tomography measurements It is possible to determine the fill height of a partially filled pipe with this information improve the performance of the magnetic flow measurements

The combination of an image of flow regime with velocity is expected to bring new insights and process benefits

Further aspects of the present invention include: EM generation method

According to Magnetic field H stated in above and Figures 2 and 3, the field can be generated via a pair of coils or a number of coils which are arranged around the peripheral of the sensing section, generating uniformed parallel magnetic field and producing a profile of mean water or conductive fluid velocities from pairs of electrodes arranged in orthogonal to the magnetic field (see Figure 2b).

The coils can be arranged in either adjacent or partially overlapped to generate more homogeneous magnetic field. The magnitudes of electrical excitations applied to the number of coils are in the form of cosine of f, as indicated in Figure 3b.

The magnetic field direction as stated in a and b and above relevant sections can be rotated in kS angle by adjusting the field intensity on coils, as Hi=Hocos(<fr-tkS), where Ho is the maximum amplitude of magnetic field, d is the minimum rotation angle and k is step of rotation.

A stator made from ferromagnetic materials is applied to intensify the magnetic field intensity (see Figure 4).

EMF method: A number of mean velocity profiles can be obtained from pairs of electrodes arranged orthogonal to the rotated magnetic field.

A velocity field weigh matrix can be analytically or numerically calculated and therefore, the velocity spatial distribution (V) can be tomographic-inversed computed with the weigh matrix and the voltage difference vector (see below equation). The weight matrix is computed from a homogeneous distribution of permeability and conductivity of flow mixture.

The weight matrix can also be derived from a nonhomogeneous distribution of permeability and conductivity of flow mixture, updated with disperse phase distribution obtained from ERT. The velocity distribution vector can be simply derived with either the back projection or the iterative methods.

Dϋ = BCVD where AU is the voltage difference vector between pairs of electrodes, B is the effective magnetic flux density, C is the sensitivity weight matrix, V is the local velocity vector and D is the distance vector presented between relevant measurement pairs of electrodes.

Electrode device: a. According to the electrode system stated in the previous section and Figure 1 , a second ring of electrodes may be installed, which is adjacent to the first electrode system within a certain distance.

b. The disperse phase velocity spatial distribution can be derived with the crosscorrelation method.

Measurement system and device:

Modulation frequency separation techniques (frequency multiplexing), time interleaving (time multiplexing) stated in the previous sections, a dual frequency excitation is applied, where a DC or low frequency excitation is applied for EMF and a high frequency is applied for ERT.

According to above a, the excitations can be applied in either time interleave or simultaneous formats.

Demodulation can be made in either analogue frequency filtering method or digital phase or frequency sensitive demodulation.

Both the disperse phase and water phase flow rates can be obtained using the mean velocity profiles of water and disperse phase and the concentration profiles derived from ERT.

The flow rates of each phase can also be derived from the spatial distributions of water and disperse phase velocity and concentration.