RU2780566C1 - Method for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline and a device for its implementation - Google Patents

Abstract

FIELD: gas industry.

SUBSTANCE: method for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline is proposed. On the outer surface of the pipe, a set of sensors is installed in series, capable of detecting vortex disturbances in the flow and generating signals about them. The hydrodynamic component of pulsating pressure extracted from turbulent vortices is used as a useful signal, the signal is then filtered, first removing the long-wavelength acoustic components of pressure pulsations of turbulent vortices, then filtering out the high-frequency component of the signal. The rest of the signal is processed. Then the speed of convection of vortex disturbances inside the flow is determined. Next, the volumetric flow rate of the liquid is calculated from the speed of convection of vortex disturbances. As sensors, elastic deformation sensors are used, made with the possibility of restructuring them to filter waves of different lengths. The signals taken from the sensors are amplified before filtering. After filtering, the signal containing the values ​​of pressure fluctuations of turbulent eddies in the convective region is subjected to frequency-wave processing using the k-ƒ curve. Also proposed is a device for non-invasive determination of the volumetric flow rate of liquid and gas in the pipeline.

EFFECT: increase in the accuracy of determining the volumetric flow rate of liquid and gas in the pipeline is achieved.

8 cl, 4 dwg

Description

The group of inventions relates to means for measuring the volumetric flow rate of liquid and gaseous media and can be widely used in measuring technology in the oil and gas industry.

The relevance of the group of inventions lies in the creation of a new non-invasive method for measuring volume flow, implemented in the form of an overhead flow meter with high-quality performance characteristics.

In this regard, the quality performance is understood as the following:

- providing a non-invasive method for measuring volume flow without narrowing the flow path;

- measurement of flow parameters according to its physical characteristics without creating artificial disturbances;

- the ability to measure the flow of liquid and gas in a wide range;

- the possibility of reinstallation on different sections of the pipeline without stopping the process;

- a wide range of measured costs.

At the moment, there are many devices designed to measure the flow of liquid and gaseous media, but they all have their own advantages and disadvantages. Among the most common flowmeters, there are two main drawbacks - this is a decrease in the flow area of the channel and the need for installation in the pipeline by tie-in. For example, differential pressure membrane flowmeters significantly reduce the flow area of the pipeline, vortex flowmeters are unstable at low speeds, ultrasonic flowmeters are highly dependent on acoustic wave reflection and calibration, Coriolis flowmeters are affected by vibrations, etc. All these factors entail economic losses and reduced safety.

Current trends in instrumentation are focused on energy efficiency and cost reduction. At present, non-invasive methods for measuring the flow rate of liquid and gaseous media in pipelines are widely used. This is the next round in the development of measuring instruments. One of such devices is a clamp-on frequency-wave flow meter, which determines the flow rate from the pressure fluctuations of turbulent eddies in the flow.

The device is designed to measure the volume flow of liquid, gaseous and multi-phase media flowing in the pipeline, with the possibility of reinstallation in different sections of the pipeline without stopping the process. This saves a lot of resources, improves the safety of the system, since there is no need to work with high pressure, and improves the quality of production, since there is no loss of product. It has a number of advantages compared to clamp-on ultrasonic flowmeters, since it does not require calibration when installed on a pipeline, it measures the physical parameter created by the flow, it does not depend on the speed of sound in the flow and pipeline, and there are no re-reflections of the radiated wave.

The proposed measurement method is based on Taylor's “turbulence freezing” hypothesis, which says that if the value of the pulsation velocity u’(t) is much less than the flow velocity U, then the spatial parameters of turbulence x can be determined from the temporal t (I) values.

In other words, the convection velocity of turbulent eddies U _c does not depend on the pulsation component u'(t) and is equal to the average flow velocity. Thus, knowing the vortex pulsation frequency - ƒ and the wave number - k, one can determine the convection velocity (II).

As a signal source, vortices moving in the boundary layer or near it, in a turbulent flow, are used, which can be defined as an instantaneous pressure value (III):

where P(x) - static pressure, p(x,t) - centered function describing random in space and time pressure fluctuations, which in turn can be represented as a sum (IV) of hydrodynamic p(x,t) _ha , acoustic p(x,t) _as well as other components of the pulsation pressure p(x,t) _etc .

According to J. Taylor's hypothesis, a useful signal source is the hydrodynamic component p(x, t) _where of the pulsating pressure, since vortex convection occurs precisely in this region of the signal. However, simply reading the signal at a point is not enough, since in order to calculate the convection velocity, it is necessary to determine the parameters of the converting vortex, such as the vortex pulsation frequency - ƒ and its wave number - k. For this, spatial spectral processing with a correlation function is used.

The closest in technical essence to the claimed are the technical solutions presented in paragraph US No. 8346491 according to class. G01F 1/00, 1/20, 7/00, 22/00, s. February 22, 2008, op. 01/01/2013 "Ultrasonic flow meter to ensure product identification" and selected as prototypes.

In a well-known patent, 2 variants of the measurement method and two variants of devices implementing them are presented.

Of the two options, the closest to the claimed ones are the technical solutions presented in the first option and characterized by the following formulas (the translation is given in the appendix to the application):

1st option

1. A device for identifying one or more liquid products flowing inside a pipe, including:

- a flow meter having a plurality of sensors capable of detecting eddy disturbances in the flow with liquid products and acoustic waves propagating through the product flow, and generating signals indicative of eddy disturbances and acoustic waves, and mounted on a pipe and;

- a processing unit capable of determining the speed of sound and volumetric flow of one or more liquid products, using signals from a flow meter, in which the processing unit includes a database having sound velocity data for a predetermined group of products and in which the next processing unit is able to identify the type of each product inside the pipe, giving the temperature and pressure of the product inside the pipe.

2. The apparatus of claim. 1, wherein the database contains data on the speed of sound as a function of the temperature and pressure of the products inside the pipe.

3. The device according to claim 2, in which the database is a table.

4. The apparatus of claim 1, wherein the processing unit is capable of determining the volumetric flow rate by determining the rate of convection of vortex disturbances within the flow of one or more liquid products, and wherein the rate of convection is determined by characterizing the convection of a ridge (protrusion) representing the vortex disturbances within the pipe.

5. The apparatus of claim 1, wherein the processing unit is capable of determining the volumetric flow rate by determining the rate of convection of eddy disturbances within the flow of one or more liquid products, and wherein the rate of convection is determined by cross-correlation instead of pressure changes.

(clauses 6-11 on the device refer to the 2nd option)

12. A method for identifying one or more products of a fluid stream flowing inside a pipe, including:

- providing a flow meter having a plurality of sensors capable of detecting eddy flow disturbances with liquid products and acoustic waves propagating through the product flow, and a processing unit having a database containing sound velocity data for a predetermined group of products, as a function of temperature and pressure;

- determining the speed of sound for each of the one or more liquid products, using a κ-ω curve, based on the first signals from the flow meter, represented by acoustic waves propagating through the flow;

- determining the speed of convection of vortex disturbances within the flow, based on the second signals from the flow meter, represented by vortex disturbances from one or more liquid products;

- identification of the type of each product, using a certain speed of sound for a given temperature and pressure of the products inside the pipe.

13. The method of claim 12, further comprising the step of determining the volumetric flow rate of one or more products within the pipe.

14. The method of claim 13, wherein the database is a lookup table.

15. The method according to claim 12, in which the speed of the convention of vortex disturbances within the flow is determined using the curve (graph) k-ω.

16. The method of claim 12 wherein the convention rate is determined by cross-correlation instead of pressure variations.

From the formula of the known device and method, the following can be seen:

1) as a source of a useful signal they use:

- signals received from vortex disturbances to determine the volume flow;

- signals received from acoustic waves to distinguish and identify the composition of a multiphase flow.

2) in this case, the volume flow is measured by determining the velocity of vortex disturbances within the flow or by characterizing a convection ridge (protrusion) representing vortex disturbances inside the pipe or by cross-correlation instead of pressure changes.

More specifically, in view of the drawings and description, the following can be said. The known device contains a set of strain gauge sensors installed on the pipeline in series, through a certain distance. Each pair of sensors is connected by an adder and is a spatial (wave) filter that removes long-wavelength acoustic components of pressure fluctuations of turbulent eddies. The resulting filtered signal is fed to a bandpass (frequency) filter, which cuts off the high-frequency component of the signal. Thus, a part of the signal remains, containing the parameters of pulsations of turbulent vortices in the convective region of the frequency-wave spectrum, which enters the correlation processing module, where the data between the two processed signals are compared, and the time delay is calculated, which corresponds to the time of passage of the converting vortex in the flow. Since the distance between the spatial filters is known in advance, it is possible to calculate the vortex convection velocity in the boundary layer. With the aid of the transfer coefficient, the vortex transfer rate is converted into a flow velocity. Multiplying the velocity by the cross-sectional area of the pipeline, we obtain the volume flow.

This device has a number of disadvantages: a narrow range of spatial filtering, since the distance between the sensors is strictly defined, while the vortices moved by the flow have a different scale, and, accordingly, different pulsation frequency and wave number.

The main problem of the known method is the low measurement accuracy, since the vortex convection rate is calculated for a specific wave number and frequency, and this is not enough to accurately determine the convection rate.

Also in the specified patent No. 8346491 it is said about the use of a frequency-wave method for measuring the speed of convection of vortices, which determines the front of the convective maximum (crest) in a given range of frequencies and wave numbers. For this, data collection for a certain period of time and the fast Fourier transform (FFT) are used, after which the data is fed to the "Array Processing Node". At the output of the “Array Processing Node”, a frequency-wave spectrum or k-ω graph is obtained.

The main problem of the known method is the low accuracy of determining the convective front during signal processing, to improve it, an optimization function is used. However, there is a risk of obtaining values that do not correspond to the convective region of the vortex field. The description of the invention does not indicate how it is possible to isolate the hydrodynamic component of pressure fluctuations for subsequent frequency-wave processing, there is only a mention of a preliminary fast Fourier transform. As a result, the measurement results will not correspond to the true average flow rate.

In addition, in the well-known patent under consideration, these methods are presented as part of a system for measuring the flow rate of multiphase flows with phase separation, and there is no data separately for measuring the volumetric flow rate of a single-phase flow without the use of pressure and temperature sensors.

At the same time, the method for determining the volumetric flow rate is very complicated due to the use of the fast Fourier transform, the need to process acoustic signals, and the device itself is also very difficult, because requires the use of acoustic pressure sensors, a database of sound speed versus temperature and pressure.

The task is to improve the accuracy of determining the volumetric flow rate of liquid and gas in the pipeline while simplifying the technical means used for this.

The problem is solved by:

- in a method for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline, which consists in the fact that a set of sensors is installed sequentially on the outer surface of the pipe, capable of detecting vortex disturbances in the flow and creating signals about them, which are then filtered, first removing the long-wave acoustic components of pressure pulsations turbulent vortices, then filtering the high-frequency component of the signal, the hydrodynamic component of pulsating pressure extracted from turbulent vortices is used as a useful signal, which is then processed and the speed of convection of vortex disturbances inside the flow is determined using the k-ƒ graph, then the volumetric liquid flow rate, according to the invention, as sensors, elastic deformation sensors are used, made with the possibility of restructuring them for filtering waves of different lengths, the signals taken from the sensors are amplified before f filtering, after spatial filtering, the signal is subjected to frequency filtering, after which the filtered signal containing the values of pressure fluctuations of turbulent eddies in the convective region is subjected to frequency-wave processing in the k-ƒ module.

In this method, after filtering, the signal can be subjected to two-dimensional correlation processing in time and space, comparing the data between the first and subsequent processed signals and calculating the delay time between them:

where W - two-dimensional correlation function between the first signal and subsequent ones, N - number of samples, Μ - number of sensors, p ₁ - signal in the first filter, p ₂ - signal in subsequent filters, n - rows, m - columns, τ - time shift signal, q=1, 2, …, N-1, x - distance between sensors, t - time.

Then, two-dimensional frequency-wave processing is performed according to the dependence (VI):

where S is the frequency-wave spectrum, k is the wavenumber, ƒ is the frequency.

In addition, in the method, after filtering, the signal can be subjected to frequency-wave processing according to the dependence:

where S - frequency-wave spectrum, k - wave number, ƒ - frequency, N - number of samples, Μ - number of sensors, p - set of signals (two-dimensional matrix) from sensors after filtering, n - rows, m - columns, τ - signal time shift, x - distance between sensors, t - time.

- in a device for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline, including a flow meter having a set of sensors installed on the outer side of the pipeline at an equal distance, capable of detecting vortex perturbations of the flow through a solid wall with liquid products and creating signals indicating them, associated with the meter flow processing unit containing signal filtering elements and connected to the filtered signal processing module k-ƒ, connected to the flow computer, according to the invention, elastic deformation sensors are used as sensors, configured to rebuild them to filter waves of different lengths, connected by outputs with signal amplifiers connected to the inputs of the processing unit.

At the same time, in the device, the outputs of each pair of adjacent sensors can be connected to the inputs of two signal amplifiers connected to the inputs of the adder to form the primary wave (spatial) filter specified by the assembly, the output of each of which is connected to the processing module through a band-pass filter.

In addition, in the device, elastic strain sensors can be made with a predetermined frequency-wave signal bandwidth configured to detect pressure fluctuations of turbulent eddies in the convective region, and their outputs are connected to the inputs of the filtered signal processing module k-ƒ.

And also in the k-ƒ device, the module may contain blocks for processing the filtered signal, while the first block is used to perform correlation signal processing, the second - for frequency-wave processing, in which the speed of convection of turbulent vortices in the flow is determined, and the third block is designed to calculate volume flow using a transfer ratio.

In addition, in the k-ƒ device, the module may contain blocks for processing the filtered signal, while the first block is used to perform frequency-wave processing, in which the speed of convection of turbulent eddies in the flow is determined, and the second block is designed to calculate the volume flow using the transfer coefficient .

Application in the claimed method to detect vortex disturbances in the flow and create signals about them of elastic deformation sensors, made with the possibility of rebuilding them for filtering waves of different lengths, in combination with amplifying the signals taken from the sensors before filtering, and frequency-wave processing in k-ƒ The module of filtered signals containing the values of pressure fluctuations of turbulent vortices in the convective region makes it possible to increase the accuracy of determining the convention velocity, and, consequently, the determination of the volumetric flow rate with a very simple processing technique.

In the claimed device, the use of elastic strain sensors as sensors in the flow meter, which can be reconfigured to filter waves of various lengths, connected by outputs to signal amplifiers connected to the inputs of the processing unit, together provides an increase in the accuracy of determining the convection velocity, and hence determining volumetric flow with a very simple scheme.

The technical result is an increase in the accuracy of determining the volumetric flow rate of liquid and gas in the pipeline while simplifying the means used for this.

The claimed method and device for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline are novel in comparison with the prototype, differing from it in such essential features as:

- in the method: the use of elastic deformation sensors made with the possibility of restructuring them for filtering waves of different lengths, amplifying the signals taken from the sensors before filtering, frequency-wave processing in the k-ƒ module after filtering signals containing the values of pressure fluctuations of turbulent eddies in the convective region , providing in the aggregate the achievement of a given result;

- in the device: use as sensors in the flow meter of elastic deformation sensors, made with the possibility of restructuring them to filter waves of different lengths, connected by outputs to signal amplifiers connected to the inputs of the processing unit, which together ensure the achievement of the desired result.

Although by themselves, such operations as spatial and band-pass filtering, frequency-wave processing are known in science and the given field of technology, and the elements of devices that implement them, however, their joint use in measuring volumetric flow with an increase in the accuracy of this measurement and simplification of means of implementation for the applicant unknown, so he believes that the claimed technical solutions meet the criterion of "inventive step".

The inventive method and device for determining the volumetric flow rate of liquid and gas in a pipeline can be widely used in measuring technology in the oil and gas industry and therefore meet the criterion of "industrial applicability".

The inventions are illustrated by drawings, where they are presented on:

- fig. 1 is a schematic diagram of the operation of the device based on the frequency-wave method for determining the speed of vortex convection;

- fig. 2 - frequency-wave spectrum using preliminary correlation;

- fig. 3 - frequency-wave spectrum without the use of preliminary correlation;

fig. 4 is a schematic diagram of the operation of the device based on the frequency-wave method for determining the speed of vortex convection.

The inventive method for non-invasive determination of the volumetric flow rate of liquid and gas in the pipeline is as follows.

On the outer surface of the pipe, a set of elastic deformation sensors is installed in series, capable of detecting vortex disturbances in the flow and creating signals about them, configured to rebuild them for spatial filtering of waves of different lengths. In this case, the hydrodynamic component of the pulsating pressure extracted from turbulent vortices is used as a useful signal. These signals are amplified and then filtered, first removing the long-wavelength acoustic components of the pressure fluctuations of turbulent eddies, then filtering out the high-frequency component of the signal. The rest of the signal, containing the values of pressure fluctuations of turbulent vortices in the convective region, is subjected to frequency-wave processing using the k-ƒ curve. Then the speed of convection of vortex disturbances inside the flow is determined, and then the volumetric flow rate of the liquid is calculated from the speed of convection of vortex disturbances.

In this case, after filtering, the signal can be subjected to two-dimensional correlation processing, followed by two-dimensional frequency-wave processing according to the above dependences in formulas V, VI, or only frequency-wave processing according to dependence VII.

The device for determining the volumetric flow rate of liquid and gas in the pipeline is made as follows (Fig. 1).

On the outer wall 1 of the pipeline for recording the pressure pulsations of the vortices 2 in the flow inside the pipe, sensors 3 of elastic deformations are installed, for example, piezofilm, optical with a Braga grating, piezoceramic, etc. The outputs of the sensors 3 are connected by means of a cable 4 to the inputs of the amplifiers 5. The outputs of the amplifiers 5 by means of cables 6 are connected to the inputs of adders 7, which receive signals from a pair of such sensors 3. This assembly 8 is a primary wave (spatial) filter that removes the long-wave component of the signal. The filtered signal from the wave (spatial) filter 8 is fed through the cable 9 to the bandpass (frequency) filter 10, where the high-frequency part of the spectrum containing the acoustic components of the signal is cut off. As a result, at the output of filter 10, the signal contains the values of pressure fluctuations of turbulent eddies p(x,t) _where in the convective region, which are transmitted via cable 11 to module 12 of the frequency-wave signal processing k-ƒ of module 12, in which there is a memory element 14, where the k-ƒ curve is stored. The data from the remaining filters 10 is also transmitted via cables 11 to the k-ƒ module 12, which can be implemented in two variations: with or without a correlation processing unit 13 15.

More specifically, the processing is performed as follows. In the first case, in block 13, the correlation processing of the signal in time and space according to the formula (V) is performed, after which two-dimensional frequency-wave processing is performed according to the dependence (VI) in block 14:

where W - two-dimensional correlation function between the first signal and subsequent ones, N - number of samples, Μ - number of sensors, p ₁ - signal in the first filter, p ₂ - signal in subsequent filters, n - rows, m - columns, τ - time shift signal, q=1, 2, …, N-1, x - distance between sensors, t - time.

where S is the frequency-wave spectrum, k is the wavenumber, ƒ is the frequency.

In the second case, in block 15, frequency-wave processing according to dependence (VII) is sufficient.

where S - frequency-wave spectrum, k - wave number, ƒ - frequency, N - number of samples, Μ - number of sensors, p - set of signals (two-dimensional matrix) from sensors after filtering, n - rows, m - columns, τ - signal time shift, x - distance between sensors, t - time.

Based on the results of processing, two-dimensional graphs of the frequency-wave spectra are constructed: for block 13 with correlation, it is shown in Fig. 2, for block 15 without correlation is presented in figure 3. As can be seen from the results, the front of convective maxima 18 corresponds to a flow velocity of ~5 m/s, which was set at the inlet to the pipeline. Each point lying on the top of the front determines the speed of the vortex convection of its own scale. Thus, it is possible to determine the speed of convection of the vortex field in the flow, and not of an individual wave. To do this, a linear approximation of the values of the front of convective maxima is performed and the angle of inclination φ of the front line 18 is determined, the values of which are transmitted via cable 16 to module 17, where the vortex convection velocity U _c can be calculated. Knowing the speed of convection, it is possible to calculate the average flow rate through the transfer coefficient γ, and from the average speed to calculate the volume flow Q (VIII).

An alternative implementation of the invention is possible (Fig. 4). If sensors 19 are used with a predetermined frequency-wave signal bandwidth configured to detect pressure fluctuations of turbulent vortices 2 in the convective region, filtering using wave 8 and band-pass filters 10 is not required and signals from sensors 19 are transmitted via wires 20 to amplifiers 21, and the amplified signal over the wires is 22 in the k-ƒ module 12.

In comparison with the prototypes, the inventive means for determining the volumetric flow rate of liquid and gas in the pipeline provide a more accurate determination of the volumetric flow rate and are simpler.

Claims (15)

1. A method for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline, which consists in the fact that a set of sensors is installed sequentially on the outer surface of the pipe, capable of detecting vortex disturbances in the flow and creating signals about them, using the hydrodynamic component extracted from turbulent vortices as a useful signal pulsating pressure, the signal is then filtered, first removing the long-wavelength acoustic components of the pressure pulsations of turbulent vortices, then filtering out the high-frequency component of the signal, the rest of the signal is processed, then the speed of convection of vortex disturbances inside the flow is determined, then the volumetric flow rate of the liquid is calculated from the speed of convection of vortex disturbances, which differs due to the fact that elastic deformation sensors are used as sensors, which can be reconfigured for filtering waves of different lengths, the signals taken from the sensors are amplified before filtering, after After filtering, the signal containing the values of pressure fluctuations of turbulent eddies in the convective region is subjected to frequency-wave processing using the k-ƒ curve. 2. The method according to claim 1, characterized in that, after filtering, the signal is subjected to two-dimensional correlation processing in time and space, comparing the data between the first and subsequent processed signals and calculating the delay time between them:

where W - two-dimensional correlation function between the first signal and subsequent ones, N - number of samples, Μ - number of sensors, p ₁ - signal in the first filter, p ₂ - signal in subsequent filters, n - rows, m - columns, τ - time shift signal, q = 1, 2, …, N-1, x - distance between sensors, t - time. Then, two-dimensional frequency-wave processing is performed according to the dependence (VI):

where S is the frequency-wave spectrum, k is the wavenumber, ƒ is the frequency. 3. The method according to p. 1, characterized in that after filtering the signal is subjected to frequency-wave processing according to the dependence:

where S - frequency-wave spectrum, k - wave number, ƒ - frequency, N - number of samples, Μ - number of sensors, p - set of signals (two-dimensional matrix) from sensors after filtering, n - rows, m - columns, τ - signal time shift, x - distance between sensors, t - time. 4. A device for non-invasive determination of the volumetric flow rate of liquid and gas in a pipeline, including a flow meter having a set of sensors installed on the outer side of the pipeline at an equal distance, capable of detecting vortex perturbations of the flow through a solid wall with liquid products and creating signals indicating them, associated with the meter flow processing unit containing signal filtering elements and connected to the filtered signal processing module k-ƒ, containing a memory element with a k-ƒ curve and connected to the flow computer, characterized in that elastic deformation sensors made with the possibility of rebuilding them for filtering waves of different lengths, connected by outputs to signal amplifiers connected to the inputs of the processing unit. 5. The device according to claim 4, characterized in that the outputs of each pair of adjacent sensors are connected to the inputs of two signal amplifiers connected to the inputs of the adder to form the primary wave (spatial) filter specified by the assembly, the output of each of which is connected to the module through a bandpass filter processing. 6. The device according to claim 4, characterized in that the elastic strain sensors are made with a predetermined frequency-wave signal bandwidth configured to detect pressure fluctuations of turbulent eddies in the convective region, and their outputs are connected to the inputs of the k-ƒ module for processing filtered signals . 7. The device according to claim 4, characterized in that the k-ƒ module contains blocks for processing the filtered signal, while the first block is used to perform correlation processing of the signal, the second - for frequency-wave processing, in which the speed of convection of turbulent vortices in the flow is determined , and the third block is designed to calculate the volume flow using the transfer coefficient. 8. The device according to claim 4, characterized in that the k-ƒ module contains blocks for processing the filtered signal, while the first block is used to perform frequency-wave processing, in which the speed of convection of turbulent vortices in the flow is determined, and the second block is designed to calculate volume flow using a transfer ratio.

Images