阅读说明：本技术 基于三角电极电容传感器的金属颗粒流速测量装置及方法 (Metal particle flow velocity measuring device and method based on triangular electrode capacitance sensor ) 是由 高鹤明 闵莹星 龙航 仲苏苏 常琦 于 2019-08-22 设计创作，主要内容包括：本发明公开了一种基于三角电极电容传感器的金属颗粒流速测量装置,包括依次电性连接的用于检测的三角电极差分式电容传感器、信号采集电路和计算机,三角电极差分式电容传感器包括绝缘管道,绝缘管道上设有上游电极对和下游电极对,上游电极对的上方、上游电极对与下游电极对之间、下游电极对下方均设有屏蔽电极,绝缘管道外侧套接有金属屏蔽罩,本发明还公开了基于三角电极电容传感器的金属颗粒流速测量方法,通过计算机对传感器的输出信号进行分析计算得到金属颗粒的速度。本发明的传感器装置具有灵敏场分别更均匀,结构简单,成本低,反应速度快,灵活性高,灵敏度高的优点,提高管道内金属颗粒流动速度的测量精确度。(The invention discloses a metal particle flow velocity measuring device based on a triangular electrode capacitance sensor, which comprises a triangular electrode differential capacitance sensor, a signal acquisition circuit and a computer which are electrically connected in sequence and used for detection, wherein the triangular electrode differential capacitance sensor comprises an insulating pipeline, an upstream electrode pair and a downstream electrode pair are arranged on the insulating pipeline, shielding electrodes are arranged above the upstream electrode pair, between the upstream electrode pair and the downstream electrode pair and below the downstream electrode pair, and a metal shielding cover is sleeved outside the insulating pipeline. The sensor device has the advantages of more uniform sensitive fields, simple structure, low cost, high reaction speed, high flexibility and high sensitivity, and improves the measurement accuracy of the flowing speed of metal particles in the pipeline.)

1. The metal particle flow velocity measuring device based on the triangular electrode capacitance sensor is characterized by comprising a triangular electrode differential capacitance sensor (1), a signal acquisition circuit (2) and a computer (3) which are electrically connected in sequence and used for detection;

The triangular electrode differential capacitive sensor (1) comprises an insulating pipeline (4), an upstream electrode pair and a downstream electrode pair are arranged on the insulating pipeline (4), shielding electrodes (8) are arranged above the upstream electrode pair, between the upstream electrode pair and the downstream electrode pair and below the downstream electrode pair, and a metal shielding cover (5) is sleeved on the outer side of the insulating pipeline (4).

2. The triangular-electrode capacitive-sensor-based metal particle flow rate measuring device of claim 1, wherein: the upstream electrode pair and the downstream electrode pair surround the insulating pipeline (4), the upstream electrode pair and the downstream electrode pair are respectively composed of a triangular excitation electrode (6) and a triangular detection electrode (7), and the oblique sides of the excitation electrode (6) and the detection electrode (7) are oppositely arranged.

3. The triangular-electrode capacitive-sensor-based metal particle flow rate measuring device of claim 2, wherein: the axial distance between the inclined edges of the excitation electrode (6) and the detection electrode (7) along the insulating pipeline (4) is 1-3 mm, one right-angle edge is perpendicular to the axis of the insulating pipeline (4) and has the length equal to the circumference of the outer wall of the insulating pipeline (4), and the other right-angle edge is parallel to the axis of the insulating pipeline (4) and has the length of 10-20 mm.

4. the triangular-electrode capacitive-sensor-based metal particle flow rate measuring device of claim 2, wherein: the excitation electrode (6), the detection electrode (7) and the shielding electrode (8) are all made of purple copper foils and are all embedded in the insulating pipeline (4).

5. The triangular-electrode capacitive-sensor-based metal particle flow rate measuring device of claim 1, wherein: the length of the metal shielding cover (5) is 1.5-2 times of the distance from the shielding electrode (8) above the upstream electrode pair to the shielding electrode (8) below the downstream electrode pair, and insulating materials are filled between the insulating pipeline (4) and the metal shielding cover (5).

6. The triangular-electrode capacitive-sensor-based metal particle flow rate measuring device of claim 1, wherein: the signal acquisition circuit (2) is a capacitance digital conversion circuit based on a PCAP01 chip and an interface circuit thereof, is connected with an upstream electrode pair, a downstream electrode pair and a shielding electrode (8) through a single-core shielding cable, and transmits acquired signals to the computer (3).

7. The triangular-electrode capacitive-sensor-based metal particle flow rate measuring device of claim 2, wherein: the excitation electrode (6) is connected with the high-level end of the signal acquisition circuit (2), the detection electrode (7) is connected with the low-level end of the signal acquisition circuit (2), and the shielding electrode (8) and the metal shielding cover (5) are connected in series and then grounded.

8. The metal particle flow velocity measuring method based on the triangular electrode capacitance sensor is characterized by comprising the following steps: the device for measuring the flow velocity of metal particles based on the triangular electrode capacitive sensor as claimed in claim 1, wherein the measuring method is characterized in that when the metal particles move in the insulated pipeline (4), the metal particles pass through a sensitive space of the triangular electrode differential capacitive sensor (1), the sensitive space is a space between the uppermost shielding electrode (8) and the lowermost shielding electrode (8), and two groups of capacitance signals delta C containing components of reaction particle flow information are generated₁(t) and Δ C₂(t), the capacitance signal is sampled by a signalcollecting by a collecting circuit (2), transferring the capacitance signal to a computer (3) by the signal collecting circuit (2) through conversion, carrying out correlation operation on an input electric signal by computer software based on a cross-correlation velocity measurement principle, analyzing two paths of output signals and a differential signal frequency spectrum respectively based on a spatial filtering velocity measurement principle, and extracting the transit time tau of the two paths of signals_mAnd equivalent peak frequency f of the signals before and after the difference_m1，f_m2，f_mdThen, the movement speed information of the metal particles is preliminarily obtained, and then the four speeds v obtained based on different principles are subjected to the feature layer data fusion algorithm based on the weighted average method_c，v_m1，v_m2，v_mdFurther processing is performed to obtain a measurement result with higher accuracy.

9. The method for measuring the flow velocity of the metal particles based on the triangular electrode capacitive sensor as claimed in claim 8, wherein the process of processing the collected signals in the computer is specifically as follows:

Step 1, programming two groups of collected signals delta C₁(t) and Δ C₂(t) performing cross-correlation operation, and calculating correlation function R of two-way capacitance output signals of the sensor according to a correlation theorem_xy(τ), expressed as:

In the formula,. DELTA.C₁(t) and Δ C₂(t) two output signals of the sensor are obtained by determining a correlation curve R_xythe time corresponding to the peak of (τ), i.e. the transit time τ_mThe velocity v of the metal particles can be obtained_c：

v_c＝λ/τ_m＝(w₁+w₂+2d)/τ_m (2)

Wherein λ is the upstream electrode pair C₁Electrode pair C with downstream₂axial spacing therebetween, w₁to measure the electrode width, w₂For shielding electrode width, d is the excitation constituting the capacitanceThe spacing of the electrodes and the detection electrode from the shield electrode;

Step 2, in order to better obtain a capacitance change rule caused when metal particles pass through a sensitive space of the sensor, the axial sensitivity distribution of the triangular electrode differential capacitance sensor is defined by the following equation:

In the formula: z is the total axial length of the sensitive space of the sensor, S_i(z) is the sensitivity of the sensor when the metal particles are at different axial positions z, C_i(z) is the two output signals of the sensor when the particle is at the axial position z, C_εlAnd C_εhThe capacitance value of an empty tube and the capacitance value of a full tube corresponding to the sensor when the sensor is filled with air and metal particles, mu is a correction factor related to the particle size of the metal particles and is defined as the ratio of the sensitive space volume of the sensor to the volume of the metal particles, and a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve;

Step 3, when the particles with the determined positions only have velocity components in the z-axis direction, the velocity components in other directions are zero, the spatial weight function only has periodicity in the z-axis direction, and the spatial weight function is the same in the z-axis direction, the output capacitance signal deltaC of the capacitance sensor_i(t) is expressed as:

ΔC_i(t)＝k₀∫∫∫ρ(x,y,z+v_zt)·s_i(x,y,z)dxdydz,i＝1,2 (4)

Wherein k is₀Is a constant related to the dielectric properties and geometric characteristics of the metal particles, the output signal Δ C_i(t) is the spatial particle distribution function ρ (x, y, z, t) and the spatial sensitivity function s_iConvolution integral of (x, y, z);

Output signal Δ C by programming_i(t) Fourier transform, squaring the obtained spectral amplitude to obtain the power spectral density function P of the measured output signal_ΔCi(f) Expressed as:

in the formula, S_ΔCi(f) For the sensor output signal Δ C_i(t) Fourier transform, m is a constant related to dielectric properties, geometry, and spatial position of the particles, and time frequency f is expressed as space frequency f_zAnd velocity v_zThe spatial sensitivity distribution rule of the sensor is analyzed by using a finite element analysis method, and the axial distribution s of the spatial sensitivity function at the determined radial position can be known_i(z) Power spectral density function P of the capacitive output signal, which can be fitted with a Gaussian function_ΔCi(f) Is shown as

Wherein a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve, and F (-) represents Fourier transform;

Step 4, aiming at the limitation of the existing frequency domain characteristic parameter extraction method, introducing equivalent peak frequency f_mconcept defined as the magnitude P of the power spectral density function_d(f_z) With corresponding frequency f_zThe result of the weighted sum is divided by the quotient of the sum of the amplitudes, for two output signals ac which are not differentiated_i(t) equivalent peak frequency f obtained by power spectrum analysis_mi：

The corrected result can reduce the influence of frequency domain characteristic parameter extraction on the speed measurement result of the space filtering method to a certain extent, and in practical application, a dimensionless proportionality coefficient k and the speed v of metal particles are introduced due to the influence of particle distribution, size, speed distribution, fluid uniformity and stability_miis defined as:

v_mi＝k·λf_mi＝k·(w₁+w₂+2d)f_mi,i＝1,2 (8)

In the formula (I), the compound is shown in the specification,lambda is upstream electrode pair C of triangular electrode differential capacitance sensor₁electrode pair C with downstream₂The dimensionless proportionality coefficient k is experimentally determined, v_m1and v_m2Is to output signals Delta C of two paths of upstream and downstream₁(t) and Δ C₂(t) carrying out power spectrum analysis to obtain the flow velocity of the metal particles;

step 5, two paths of output signals delta C of the triangular electrode differential capacitance sensor_i(t) performing a difference process to obtain a sensitivity function s of the difference output signal_d(z) is represented by:

In the formula, lambda is the axial interval between the upstream and downstream electrode pairs, and a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve;

Step 6, differential output signal Delta C is output through programming_d(t) Fourier transforming, squaring the obtained frequency spectrum amplitude to obtain power spectral density function P of the differential capacitance output signal of the measurement output signal_d(f_z) Expressed as:

As can be seen from the above equation, the difference processing can reduce the influence of the baseband frequency of the dc component in the output signal on the accuracy of extracting the frequency domain characteristic parameters;

Step 7, for the differential output signal Delta C_d(t) equivalent peak frequency f obtained by power spectrum analysis_mdi.e. by

introducing a dimensionless scaling factor k, the velocity v of the metal particles_mdIs defined as

v_md＝k·λf_md＝k·(w₁+w₂+2d)f_md (12)

Wherein λ is the upstream electrode pair C₁Electrode pair C with downstream₂The axial interval between the two components, the dimensionless proportionality coefficient k is determined by experiments, and the coefficient k is calibrated by using a phase Doppler velocimeter;

Step 8, combining the programming with a feature layer data fusion algorithm based on a weighted average method to four speeds v_c，v_m1，v_m2，v_mdFurther processing is performed to obtain a measurement result with higher accuracy.

Technical Field

the invention belongs to the technical field of analysis and measurement control, and relates to a metal particle flow velocity measuring device based on a triangular electrode capacitance sensor and a metal particle flow velocity measuring method based on the triangular electrode capacitance sensor.

Background

The collection and transportation of metal particles generated in the cutting process in the mechanical manufacturing industry are realized in a common pipeline type conveying mode in a centralized chip removal system. In the numerical control machine tool and multi-machine automatic production, the flowing state of the metal particles in the pipeline can be known by detecting the speed of the metal particles in the pipeline in real time, the pipeline blockage in the conveying process is effectively prevented, the normal operation of automatic processing circulation is ensured, and the unmanned treatment of cuttings is realized. The capacitance sensor can be applied to realize the detection of the metal particles in the pipeline conveying process of the metal particles, and the information of the size, the flow speed, the volume concentration, the mass flow and the like of the particles can be extracted from capacitance signals by combining a corresponding signal analysis method and a measurement principle. At present, domestic and foreign researches mainly focus on detection of micron-sized metal particles in an engine lubricating oil system, and can be divided into an electrical method, an acoustic method, a magnetic method, an optical method and the like according to different detection principles. The capacitance method is widely applied to the measurement of the flow parameters of the non-metal particles, and has less research in the field of metal particle detection because of the characteristics of low cost, quick response, non-invasive type, wide application range, good safety and the like. In the case of a capacitive sensor structure, a capacitive sensor based on a ring electrode, a spiral electrode or a matrix electrode is used to measure speed, and a capacitive tomography (ECT) based on an arc array capacitive sensor is used to measure concentration. The existing capacitive sensor has the problem of uneven distribution of sensitive fields, the measurement result is greatly influenced by a soft field effect, and the problem still exists in the application of the sensor to the measurement of the speed of metal particles.

Disclosure of Invention

The invention aims to provide a metal particle flow velocity measuring device based on a triangular electrode capacitance sensor, which solves the problems of uneven distribution of a sensitive field and large influence of a soft field effect on a measuring result in the prior art, and expands a measuring object of a capacitance method to the field of metal particle velocity measurement.

The technical scheme adopted by the invention is that the metal particle flow velocity measuring device based on the triangular electrode capacitance sensor comprises a triangular electrode differential capacitance sensor, a signal acquisition circuit and a computer which are electrically connected in sequence and used for detection;

The triangular electrode differential capacitive sensor comprises an insulating pipeline, an upstream electrode pair and a downstream electrode pair are arranged on the insulating pipeline, shielding electrodes are arranged above the upstream electrode pair, between the upstream electrode pair and the downstream electrode pair and below the downstream electrode pair, and a metal shielding cover is sleeved on the outer side of the insulating pipeline.

The invention is also characterized in that:

The upstream electrode pair and the downstream electrode pair surround the insulating pipeline, both the upstream electrode pair and the downstream electrode pair consist of a triangular excitation electrode and a triangular detection electrode, and the inclined edges of the excitation electrode and the detection electrode are oppositely arranged.

The axial distance between the inclined edges of the exciting electrode and the detecting electrode along the insulating pipeline is 1-3 mm, one right-angle edge is perpendicular to the axis of the insulating pipeline and has the length equal to the circumference of the outer wall of the insulating pipeline, and the other right-angle edge is parallel to the axis of the insulating pipeline and has the length of 10-20 mm.

The exciting electrode, the detecting electrode and the shielding electrode are all made of purple copper foils and are all embedded in the insulating pipeline.

The length of the metal shielding cover is 1.5-2 times of the distance from the shielding electrode above the upstream electrode pair to the shielding electrode below the downstream electrode pair, and insulating materials are filled between the insulating pipeline and the metal shielding cover.

the signal acquisition circuit is a capacitance digital conversion circuit based on a PCAP01 chip and an interface circuit thereof, is connected with an upstream electrode pair, a downstream electrode pair and a shielding electrode through a single-core shielding cable, and transmits acquired signals to a computer.

The exciting electrode is connected with the high-level end of the signal acquisition circuit, the detecting electrode is connected with the low-level end of the signal acquisition circuit, and the shielding electrode and the metal shielding cover are connected in series and then grounded.

The invention also aims to provide a metal particle flow velocity measuring method based on the triangular electrode capacitance sensor.

the invention adopts another technical scheme that a metal particle flow velocity measuring method based on a triangular electrode capacitance sensor utilizes a metal particle flow velocity measuring device based on the triangular electrode capacitance sensor, and the measuring method is specifically that when metal particles move in an insulating pipeline, the metal particles pass through a sensitive space of a triangular electrode differential capacitance sensor, the sensitive space refers to a space from an uppermost shielding electrode to a lowermost shielding electrode, and two groups of capacitance signals delta C containing components of reaction particle flow information are generated₁(t) and Δ C₂(t), the capacitance signal is collected by the signal collecting circuit, the capacitance signal is transmitted to the computer by the signal collecting circuit through conversion, the input electrical signal is subjected to correlation operation by computer software based on the cross-correlation velocity measurement principle, two paths of output signals and the frequency spectrum of the differential signal are respectively analyzed based on the spatial filtering velocity measurement principle, and the transition time tau of the two paths of signals is extracted_mAnd equivalent peak frequency f of the signals before and after the difference_m1，f_m2，f_mdThen, the movement speed information of the metal particles is preliminarily obtained, and then the four speeds v obtained based on different principles are subjected to the feature layer data fusion algorithm based on the weighted average method_c，v_m1，v_m2，v_mdFurther processing is performed to obtain a measurement result with higher accuracy.

The invention is also characterized in that:

The process of processing the acquired signals in the computer specifically comprises the following steps:

Step 1, programming two groups of collected signals delta C₁(t) and Δ C₂(t) performing cross-correlation operation, and calculating correlation function R of two-way capacitance output signals of the sensor according to a correlation theorem_xy(τ), expressed as:

In the formula,. DELTA.C₁(t) and Δ C₂(t) two output signals of the sensor are obtained by determining a correlation curve R_xyThe time corresponding to the peak of (τ), i.e. the transitionThe time of flight tau_mthe velocity v of the metal particles can be obtained_c：

v_c＝λ/τ_m＝(w₁+w₂+2d)/τ_m (2)

wherein λ is the upstream electrode pair C₁electrode pair C with downstream₂Axial spacing therebetween, w₁To measure the electrode width, w₂The width of the shielding electrode is d, and the distance between the excitation electrode and the detection electrode which form the capacitor and the shielding electrode is d;

Step 2, in order to better obtain a capacitance change rule caused when metal particles pass through a sensitive space of the sensor, the axial sensitivity distribution of the triangular electrode differential capacitance sensor is defined by the following equation:

In the formula: z is the total axial length of the sensitive space of the sensor, S_i(z) is the sensitivity of the sensor when the metal particles are at different axial positions z, C_i(z) is the two output signals of the sensor when the particle is at the axial position z, C_εlAnd C_εhThe capacitance value of an empty tube and the capacitance value of a full tube corresponding to the sensor when the sensor is filled with air and metal particles, mu is a correction factor related to the particle size of the metal particles and is defined as the ratio of the sensitive space volume of the sensor to the volume of the metal particles, and a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve;

Step 3, when the particles with the determined positions only have velocity components in the z-axis direction, the velocity components in other directions are zero, the spatial weight function only has periodicity in the z-axis direction, and the spatial weight function is the same in the z-axis direction, the output capacitance signal deltaC of the capacitance sensor_i(t) is represented by

ΔC_i(t)＝k₀∫∫∫ρ(x,y,z+v_zt)·s_i(x,y,z)dxdydz,i＝1,2 (4)

Wherein k is₀is a constant related to the dielectric properties and geometric characteristics of the metal particles, the output signal Δ C_i(t) is the spatial particle distribution function ρ (x,y, z, t) and a spatial sensitivity function s_iConvolution integral of (x, y, z);

output signal Δ C by programming_i(t) Fourier transform, squaring the obtained spectral amplitude to obtain the power spectral density function P of the measured output signal_ΔCi(f) Expressed as:

In the formula, S_ΔCi(f) For the sensor output signal Δ C_i(t) Fourier transform, m is a constant related to dielectric properties, geometry, and spatial position of the particles, and time frequency f is expressed as space frequency f_zAnd velocity v_zThe spatial sensitivity distribution rule of the sensor is analyzed by using a finite element analysis method, and the axial distribution s of the spatial sensitivity function at the determined radial position can be known_i(z) Power spectral density function P of the capacitive output signal, which can be fitted with a Gaussian function_ΔCi(f) Is shown as

Wherein a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve, and F (-) represents Fourier transform;

Step 4, aiming at the limitation of the existing frequency domain characteristic parameter extraction method, introducing equivalent peak frequency f_mConcept defined as the magnitude P of the power spectral density function_d(f_z) With corresponding frequency f_zthe result of the weighted sum is divided by the quotient of the sum of the amplitudes, for two output signals ac which are not differentiated_i(t) equivalent peak frequency f obtained by power spectrum analysis_mi：

The corrected result can reduce frequency domain characteristic parameters to a certain extent and extract the speed measurement result of the space filtering methodInfluence, in practical application, due to the influence of particle distribution and size, velocity distribution, fluid uniformity and stability, a dimensionless proportionality coefficient k is introduced, and the velocity v of the metal particles_miis defined as:

v_mi＝k·λf_mi＝k·(w₁+w₂+2d)f_mi,i＝1,2 (8)

In the formula, lambda is the upstream electrode pair C of the triangular electrode differential capacitance sensor₁electrode pair C with downstream₂The dimensionless proportionality coefficient k is experimentally determined, v_m1And v_m2Is to output signals Delta C of two paths of upstream and downstream₁(t) and Δ C₂(t) carrying out power spectrum analysis to obtain the flow velocity of the metal particles;

Step 5, two paths of output signals delta C of the triangular electrode differential capacitance sensor_i(t) performing a difference process to obtain a sensitivity function s of the difference output signal_d(z) is represented by

In the formula, lambda is the axial interval between the upstream and downstream electrode pairs, and a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve;

Step 6, output signal Δ C is programmed_d(t) Fourier transforming, squaring the obtained frequency spectrum amplitude to obtain power spectral density function P of the differential capacitance output signal of the measurement output signal_d(f_z) Expressed as:

As can be seen from the above equation, the difference processing can reduce the influence of the baseband frequency of the dc component in the output signal on the accuracy of extracting the frequency domain characteristic parameters;

step 7, for the output signal Delta C_d(t) equivalent peak frequency f obtained by power spectrum analysis_mdi.e. by

introducing a dimensionless scaling factor k, the velocity v of the metal particles_mdIs defined as

v_md＝k·λf_md＝k·(w₁+w₂+2d)f_md (12)

wherein λ is the upstream electrode pair C₁Electrode pair C with downstream₂The axial interval between the two components, the dimensionless proportionality coefficient k is determined by experiments, and the coefficient k is calibrated by using a phase Doppler velocimeter;

Step 8, combining the programming with a feature layer data fusion algorithm based on a weighted average method to four speeds v_c，v_m1，v_m2，v_mdFurther processing is performed to obtain a measurement result with higher accuracy.

The invention has the beneficial effects that: the triangular electrode capacitance sensor has the advantages of more uniform sensitive fields, simple structure, low cost, high response speed, high flexibility, high sensitivity and the like, because of the differential arrangement of the sensor electrode array, a plurality of speeds can be obtained based on different speed measurement principles, the speed measurement error of a spatial filtering method can be reduced by carrying out differential processing on output signals and taking equivalent peak frequency as a characteristic parameter of power spectrum estimation, and the characteristic layer data fusion is carried out at a plurality of speeds to obtain higher measurement accuracy and repeatability, according to specific conditions, if the flow state of a measured particulate matter is stable and the sampling frequency of a system is higher, the speed data obtained by a cross-correlation algorithm is allocated with larger weight, otherwise, the speed data obtained by the spatial filtering algorithm is allocated with larger weight, and by utilizing the working characteristics of the triangular electrode capacitance sensor, the special triangular electrode shape can obtain more uniform sensitive field distribution, the 'soft field effect' of the capacitive sensor is obviously improved, the influence of particle distribution on the measurement result is effectively reduced, and the measurement of the flowing speed of metal particles in the pipeline is realized.

Drawings

FIG. 1 is a schematic structural diagram of a metal particle velocity measuring device based on a triangular electrode capacitance sensor according to the present invention;

FIG. 2 is a schematic structural diagram of a triangular electrode capacitance sensor in a metal particle velocity measuring apparatus based on the triangular electrode capacitance sensor of the present invention, (a) is a sectional view of the sensor, (b) is an expanded view of the sensor surface;

FIG. 3 is a schematic diagram of a typical streamline location for a triangular-electrode capacitive sensor of the present invention;

FIG. 4 is a graph of axial sensitivity profiles at different streamline locations for a triangular-electrode capacitive sensor of the present invention.

in the figure, 1 is a triangular electrode differential capacitance sensor, 2 is a signal acquisition circuit, 3 is a computer, 4 is an insulating pipeline, 5 is a metal shielding cover, 6 is an exciting electrode, 7 is a detection electrode, and 8 is a shielding electrode.

Detailed Description

The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

The invention relates to a metal particle flow velocity measuring device based on a triangular electrode capacitance sensor, which comprises a triangular electrode differential capacitance sensor 1, a signal acquisition circuit 2 and a computer 3 which are electrically connected in sequence and used for detection, wherein the triangular electrode differential capacitance sensor 1 comprises an insulating pipeline 4, two groups of upstream electrode pairs C with the same size and structure are arranged on the insulating pipeline 4 at a certain distance in an axial direction along the insulating pipeline 4 in a surrounding manner₁And a downstream electrode pair C₂The upper side of the upstream electrode pair, the space between the upstream electrode pair and the downstream electrode pair and the lower side of the downstream electrode pair are respectively provided with an annular shielding electrode 8, the outer side of the insulating pipeline 4 is sleeved with a metal shielding cover 5, the upstream electrode pair and the downstream electrode pair are respectively composed of a triangular excitation electrode 6 and a triangular detection electrode 7, the inclined sides of the excitation electrode 6 and the detection electrode 7 are oppositely arranged, the axial distance along the insulating pipeline 4 is 1-3 mm, one right-angle side is vertical to the axis of the insulating pipeline 4, the length of the right-angle side is the same as the circumference of the outer wall of the insulating pipeline 4, the other right-angle side is parallel to the axis of the insulating pipeline 4 and is 10-20 mm, two groups of right-angle sides form a rectangularThe symmetric centers form a centrosymmetric relation, the excitation electrode 6, the detection electrode 7 and the shielding electrode 8 are all made of purple copper foils and are all embedded in the insulating pipeline 4, the length of the metal shielding cover 5 is 1.5-2 times of the sum of the upstream electrode pair and the downstream electrode pair, insulating materials are filled between the insulating pipeline 4 and the metal shielding cover 5, the signal acquisition circuit 2 is a capacitance digital conversion circuit based on a PCAP01 chip and an interface circuit thereof and is connected with the upstream electrode pair and the downstream electrode pair through a single-core shielding cable and transmits acquired signals to the computer 3, the excitation electrode 6 is connected with one high-level end of the signal acquisition circuit 2, the detection electrode 7 is connected with one low-level end of the signal acquisition circuit 2, and the shielding electrode 8 and the metal shielding cover 5 are grounded after being connected in series.

The invention relates to a metal particle flow velocity measuring method based on a triangular electrode capacitance sensor, which specifically comprises the following steps: as the metal particles move within the insulated conduit 4, the particles pass through the sensitive space of the sensor 1, the sensitive space being the space between the uppermost shield electrode (8) and the lowermost shield electrode (8), producing two sets of capacitance signals Δ C containing components that are responsive to particle flow information₁(t) and Δ C₂(t), the capacitance signal is collected by the signal collecting circuit (2), the capacitance signal is transmitted to the computer (3) by the signal collecting circuit (2) through conversion, the input electric signal is subjected to correlation operation based on the cross-correlation velocity measurement principle through computer software, two paths of output signals and the frequency spectrum of a differential signal are respectively analyzed based on the spatial filtering velocity measurement principle, and the transition time tau of the two paths of signals is extracted_mAnd equivalent peak frequency f of the signals before and after the difference_m1，f_m2，f_mdAnd then preliminarily obtaining the movement velocity information of the metal particles, and using a feature layer data fusion algorithm based on a weighted average method to obtain four velocities v based on different principles_c，v_m1，v_m2，v_mdAnd performing further processing to obtain a measurement result with higher precision, wherein if the flow state of the measured particulate matter is stable and the sampling frequency of the system is higher, the flow state of the measured particulate matter is assigned to velocity data obtained by a cross-correlation algorithm to obtain a larger weight, and otherwise, the flow state of the measured particulate matter is assigned to velocity data obtained by a spatial filtering algorithm to obtain a larger weight.

The process of processing the collected signals in the computer comprises the following steps:

Step 1, programming two groups of collected signals delta C₁(t) and Δ C₂(t) performing cross-correlation operation, and calculating correlation function R of two-way capacitance output signals of the sensor according to a correlation theorem_xy(τ), expressed as:

In the formula,. DELTA.C₁(t) and Δ C₂(t) two output signals of the sensor 1 by determining a correlation curve R_xyThe time corresponding to the peak of (τ), i.e. the transit time τ_mThe velocity v of the metal particles can be obtained_c：

v_c＝λ/τ_m＝(w₁+w₂+2d)/τ_m (14)

Wherein λ is the upstream electrode pair C₁Electrode pair C with downstream₂Axial spacing therebetween, w₁To measure the electrode width, w₂The width of a shielding electrode is set, d is the distance between an exciting electrode and a detecting electrode which form a capacitor and the shielding electrode, the speed measurement precision of the cross-correlation velocity measurement method mainly depends on the flowing state of metal particles and the sampling frequency of a system, and the more stable the flowing state of the metal particles is, the higher the sampling frequency of the system is, the higher the measured speed precision is;

Step 2, because the structural size and the geometric shape of the sensor 1 are limited, the sensor shows a certain form of spatial filtering effect on the original flow noise brought by the metal particles, and in order to better obtain a capacitance change rule caused when the metal particles pass through a sensitive space of the sensor, the axial sensitivity distribution of the triangular electrode differential capacitance sensor 1 is defined by the following equation:

In the formula: z is the total axial length of the sensitive space of the sensor, S_i(z) is a metalsensitivity of the sensor when the particles are in different axial positions z, C_i(z) are the two output signals of the sensor when the particle is at axial position z. C_εlAnd C_εhthe capacitance values of the empty tube and the full tube are corresponding to the sensor when the sensor is filled with air and particles. μ is a correction factor related to the particle size of the metal particles, defined as the ratio of the sensitive spatial volume of the sensor to the volume of the metal particles, and a, b, c are gaussian fit coefficients of the axial sensitivity profile.

When the metal particles are at different streamline positions A as shown in FIG. 3₀，A₁，A₂，A₃And (3) treating the following components: when moving along the z-axis at r-0, 1, 2, 3 mm, it can be seen from fig. 4 that the axial sensitivity profile s at a determined radial position is found_i(z) fitting can be performed using a Gaussian function, axial sensitivity s of the upstream and downstream electrode pairs₁(z) and s₂The distribution rules of (z) are completely the same, the overall double-peak rule is formed, the distance between two peaks is the axial interval lambda, and the peak value of the sensitivity curve is slightly increased at the position closer to the streamline of the pipe wall. Compared with sensors with other structures, the sensitivity change of the triangular electrode differential capacitive sensor 1 at a position far away from the pipe wall and a position close to the pipe wall is small, and the distribution of the sensitivity fields is more uniform;

Step 3, in the actual measurement process, because the motion of the conductive particles causes the steady-like field in the sensitive space of the sensor to change continuously, the output value of the sensor fluctuates continuously, in order to simplify the calculation, it is assumed that the particles at the determined positions only have a velocity component in the z-axis direction, the velocity components in other directions are zero, the spatial weight function has periodicity only in the z-axis direction and is the same in the z-axis direction, and the output capacitance signal delta C of the capacitance sensor_i(t) can be represented as

ΔC_i(t)＝k₀∫∫∫ρ(x,y,z+v_zt)·si(x,y,z)dxdydz,i＝1,2 (16)

Wherein k is₀Is a constant related to the dielectric properties and geometric characteristics of the metal particles, the output signal Δ C_i(t) is the spatial particle distribution function ρ (x, y, z, t) and the spatial sensitivity function s_iConvolution integral of (x, y, z);

Output signal Δ C by programming_i(t) Fourier transform, squaring the obtained spectral amplitude to obtain the power spectral density function P of the measured output signal_ΔCi(f) Expressed as:

in the formula, S_ΔCi(f) For the sensor output signal Δ C_i(t) Fourier transform, m is a constant related to dielectric properties, geometry, and spatial position of the particles, and time frequency f is expressed as space frequency f_zand velocity v_zThe product of (a). The method of finite element analysis is used for analyzing the spatial sensitivity distribution rule of the sensor, so that the axial distribution s of the spatial sensitivity function at the position of the determined radial position can be known_i(z) Power spectral density function P of the capacitive output signal, which can be fitted with a Gaussian function_ΔCi(f) Expressed as:

In the formula, a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve, and F (·) represents Fourier transform. According to the formula, the sensor has a low-pass filtering effect on the original flow noise of the metal particles, and the power spectral density function P of the output signal_ΔCi(f) the method comprises the steps that not only a signal component related to the fluid velocity is contained, but also a fundamental frequency direct current component with a lower frequency is contained, and the direct current component is superposed on a narrow-band frequency, so that the narrow-band frequency of a periodic signal component is reduced, the peak frequency is shifted, and the accuracy of a metal particle velocity measurement result is influenced;

step 4, aiming at the limitation of the existing frequency domain characteristic parameter extraction method, introducing equivalent peak frequency f_mConcept defined as the magnitude of the power spectral density function Pd (f)_z) With corresponding frequency f_zthe result of the weighted sum is divided by the quotient of the sum of the amplitudes, for two output signals ac which are not differentiated_i(t) equivalent peak frequency f obtained by power spectrum analysis_mi：

The corrected result can reduce the influence of frequency domain characteristic parameter extraction on the speed measurement result of the space filtering method, and in practical application, a dimensionless proportionality coefficient k and the speed v of metal particles are introduced due to the influence of particle distribution, size, speed distribution, fluid uniformity and stability_miIs defined as:

v_mi＝k·λf_mi＝k·(w₁+w₂+2d)f_mi,i＝1,2 (20)

In the formula, λ is the upstream electrode pair C of the triangular electrode differential capacitive sensor 1₁Electrode pair C with downstream₂the dimensionless scaling factor k is experimentally determined for the axial spacing therebetween. v. of_m1And v_m2Is to output signals Delta C of two paths of upstream and downstream₁(t) and Δ C₂(t) carrying out power spectrum analysis to obtain the flow velocity of the metal particles;

Step 5, two paths of output signals delta C of the triangular electrode differential capacitive sensor 1_i(t) performing a difference process to obtain a sensitivity function s of the difference output signal_d(z) is represented by:

In the formula, lambda is the axial interval between the upstream and downstream electrode pairs, and a, b and c are Gaussian fitting coefficients of an axial sensitivity distribution curve;

step 6, differential output signal Delta C is output through programming_d(t) Fourier transforming, squaring the obtained frequency spectrum amplitude to obtain power spectral density function P of the differential capacitance output signal of the measurement output signal_d(f_z) Expressed as:

As can be seen from the above equation, the difference processing can reduce the influence of the baseband frequency of the dc component in the output signal on the accuracy of extracting the frequency domain characteristic parameters;

Step 7, for the differential output signal Delta C_d(t) equivalent peak frequency f obtained by power spectrum analysis_mdi.e. by

Introducing a dimensionless scaling factor k, the velocity v of the metal particles_mdIs defined as:

v_md＝k·λf_md＝k·(w₁+w₂+2d)f_md (24)

Wherein λ is the upstream electrode pair C₁Electrode pair C with downstream₂The axial interval and the dimensionless proportionality coefficient k are determined by experiments, a phase Doppler velocimeter (PDA) is used for calibrating the coefficient k, a metal particle speed measuring device of the triangular electrode differential capacitance sensor and the PDA are synchronously measured, and the measured data of the system are recorded and stored. Taking out experimental data acquired at the same time as the PDA and measuring data of the PDA to form a data pair, selecting at least 30 pairs of data for each calibration, taking the measuring result of a metal particle speed measuring device of the angle electrode differential capacitance sensor as a horizontal coordinate, taking the measuring result of the PDA as a vertical coordinate, taking data of which the correlation coefficient is more than 0.9 as effective data, wherein the number of effective measuring points is more than half of the total data, and reaching a calibration curve through regression analysis to further obtain a calibration coefficient;

This shows that the time spectrum of the sensor completely depends on the spatial filtering characteristics, the particle velocity can be obtained by selecting a proper sensor structure and obtaining the spatial sensitivity function, and theoretical analysis and experimental verification show that the frequency spectrum characteristic of the output signal is in direct proportion to the average velocity of the particles under the condition of fixed sensor geometric characteristics.

Step 8, combining the programming with a feature layer data fusion algorithm based on a weighted average method to four speeds v_c，v_m1，v_m2，v_mdAnd performing further processing to obtain a measurement result with higher precision, wherein if the flow state of the measured particulate matter is stable and the sampling frequency of the system is higher, the velocity data obtained by the cross-correlation algorithm is assigned with higher weight, and otherwise, the velocity data obtained by the spatial filtering algorithm is assigned with higher weight.