Liquid identification method based on electrical capacitance tomography technology

Abstract

The invention relates to the technical field of flow detection, in particular to a liquid identification method based on an electrical capacitance tomography technology. A DN mapping matrix is artificially constructed by using capacitance data measured in an ECT system, the size of the DN mapping matrix is controlled by adjusting parameters: calibrating a dielectric constant, the change of an empty field dielectric constant is simulated, and the empty field dielectric constant is adjusted to be close to a measured dielectric constant. And constructing a container radius-truncation radius comparison table, and quickly selecting an optimal truncation radius value according to the radius of the measured container. And designing a segmented interval correction method, and searching the calibration dielectric constant in the technical scheme in an iteration mode. According to the technical scheme, the ECT liquid identification method based on the Calderon method is designed. According to the method, the calculation speed is high, the single dielectric constant calculation time of the algorithm is about 0.1 s, and the single liquid identification time is 4 s. The identification precision is high, and the error between the calculated value of the dielectric constant and the measured dielectric constant value does not exceed 0.01 F / m.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flow detection, and relates to a liquid identification method based on electrical capacitance tomography technology. BACKGROUND

[0002] Safety is the eternal theme of civil aviation, and ensuring aviation safety is the basis for the survival and development of civil aviation. Aviation safety inspection is one of the important means to ensure aviation safety, and searching for aviation contraband can effectively prevent hijacking (bombing) of aircraft and other aviation safety incidents, and can greatly improve the safety of civil aviation flights and protect the lives and property safety of passengers and crew. How to effectively identify liquid contraband is a major research problem in the current aviation security system. At present, the identification method for liquid at domestic airport security is still a gap, therefore, the present application proposes to use liquid detection technology based on ECT technology, which has the characteristics of fast calculation speed and relatively high identification accuracy, in order to further improve the liquid identification accuracy of aviation security and improve the efficiency of aviation security and reduce the pressure of aviation security.

[0003] ECT technology is a multiphase flow detection technology, which has the characteristics of low cost, non-invasive and high safety. Currently commonly used ECT algorithms, such as linear back projection (LBP) algorithm and Tikhonov regularization algorithm, are mostly relative algorithms based on sensitivity theory. Direct algorithm is different from ECT relative algorithm, and can directly calculate the dielectric constant distribution based on the boundary information of the sensitive field. The Calderon method has become a popular direction in the field of ECT direct algorithm research because of its fast calculation speed and ability to calculate the real value of the dielectric constant. In 1980, Calderon analyzed the model of electrical tomography technology from a mathematical point of view, and creatively proposed a new linearization criterion for solving two-dimensional conductivity problems with low conductivity contrast [1] . This method is called Calderon method, which theoretically demonstrates the possibility of inverting the distribution in the sensitive field by collecting boundary information. In 2004, Kundesen et al. gave an algorithm for solving the nonlinear Fourier transform (D-bar) equation in the plane based on the Calderon theory, and used this method to collect useful images in the EIT system [2] . In 2009, Cao Zhang of Tianjin University applied the Calderon algorithm to ECT inversion calculation, and realized the ECT image reconstruction based on the Calderon method [3] . At present, direct imaging algorithms based on Dbar algorithm and Calderon method have been applied in practice, such as human thoracic imaging [4] and flame imaging [5] .

[0004] The computational accuracy of the Calderon method is mainly affected by two factors: Factor 1, the magnitude of the change in dielectric constant when the sensitive field changes from an empty field to a full field; the smaller the change, the higher the computational accuracy. Factor 2, the value of the parameter cutoff radius; the closer the value is to the optimal value, the higher the computational accuracy. This invention provides solutions to these two influencing factors: 1. an improved method based on the DN mapping relationship; 2. an improved method based on a container radius-cutoff radius comparison table. Furthermore, a method for estimating and calibrating the dielectric constant is proposed by combining these improved methods, realizing an ECT liquid identification method based on the Calderon method. Finally, simulation experiments verify that this method can quickly and accurately identify the measured liquid. Summary of the Invention

[0005] Purpose of the invention: To solve the problem of poor calculation accuracy of the Calderon method and realize ECT liquid identification based on the method.

[0006] The technical solution of the present invention is as follows: This invention designs a non-invasive, fast-response liquid identification method based on ECT technology. The technical solution is as follows: A liquid identification method based on capacitance tomography technology. Step (1). Use the capacitance data obtained by measuring in the ECT system to artificially construct the DN mapping matrix. Control the size of the DN mapping matrix by adjusting the parameter: calibrating the dielectric constant, and use this to simulate the change of the empty field dielectric constant, and adjust it to make the empty field dielectric constant close to the measured dielectric constant.

[0007] Step (2). Construct a container radius-cutoff radius comparison table and quickly select the optimal cutoff radius value based on the radius of the container being tested.

[0008] Step (3). Design a segmented interval correction method and search for the calibration dielectric constant in technical solution (1) through iteration.

[0009] Step (4). Combining steps (1), (2), and (3) of the integrated technical solution, design an ECT liquid identification method based on the Calderon method.

[0010] Furthermore, step (1) includes: First, setting up the ECT system. An ECT (Electronic Tolerance) device should consist of at least three parts: a capacitance measurement module, a data acquisition and transmission module, and a computer for data processing. The capacitance measurement module should contain no fewer than eight independent electrodes, and its structure includes electrode separators, a metal shielding shell, individual electrodes, an internal shielding cover, and a sensitive field. In the data acquisition and transmission module, the sensor cable is an RG174 coaxial cable, with one end connected to the sensor and the other end connected to the acquisition system via an SMB RF connector. The data acquisition system can be connected to the computer using either a USB data cable or a network cable. When using a network cable, the Ethernet IPv4 parameters should be configured before connecting to the data acquisition system. The computer for data processing uses the Calderon method to calculate the dielectric constant, employing Matlab as the calculation software.

[0011] The next step, the second step, is ECT system calibration.

[0012] After completing the ECT system setup, the first step is to calibrate the ECT system. First, set the N independent electrodes in the capacitance measurement module to be marked counter-clockwise. , ,..., You can choose from them. The corresponding electrodes. Measured using the ECT capacitance measurement method. Each independent capacitance value is represented by a symbol. Indicates the first The first independent electrode and the first The capacitance value between individual electrodes.

[0013] After obtaining all After determining the individual capacitance values, a DN mapping can be constructed. The DN mapping uses the symbol... The calculation formula is as follows: (1) in Indicates the first The self-capacitance of each independent electrode is calculated using the following formula: (2) Symbols for DN mapping in empty field It indicates that the superscript The subscript represents the radius of the sensitive field. DN when empty field Mapping, The specific formula is as follows: (3) The N(N-1) / 2 independent capacitance values ​​in the formula are the capacitance values ​​measured when the ECT sensitive field is an empty field.

[0014] After obtaining the empty field DN mapping Then, the container filled with purified water was placed in the ECT sensitive field, and the entire ECT measurement was performed using the ECT measurement method described above. Individual capacitance values, and construct the DN mapping when the object field is used. .

[0015] Introduce the calibration dielectric constant as a new parameter, and use the symbol Indicate. Order and make use of the open space DN mapping and object field DN mapping Construction deviation DN mapping Matrix, denoted as , The calculation formula is: (4) The next step is to construct the charge change matrix. Constructing a voltage matrix in Matlab The matrix structure is as follows: (5) in ,in Indicates the first s The angle of each electrode, where i is a complex number. j Indicates the first j One voltage excitation mode, Indicates the first s The electrode at the first j The excitation voltage under the voltage excitation mode, the first s The excitation voltage of each electrode in the first voltage excitation mode is , No. j Each voltage excitation mode corresponds to a voltage matrix. V The j List.

[0016] From voltage matrix And deviation DN mapping matrix Calculate the charge change matrix Due to voltage V ,capacitance C With charge q Existence Relationship Therefore, the first j The voltage excitation mode is the first s The change in charge of each electrode is A vector consisting of the changes in charge of N electrodes. The calculation formula is: (6) Using the charge change matrix Calculate scattering transformation Complex variables To be evenly distributed in The point is the center of the circle, and the radius is... within a two-dimensional circular region One point, Not less than 1000 and Both are real numbers, representing complex variables respectively. The real and imaginary coordinates of the points within the circular region, and the radius This is called the cutoff radius. The value ranges from 4 to 6. Scattering transformation of formula (12) Discretization is performed, and the calculation formula is as follows: (7) in The area of ​​the individual electrodes is expressed in units of... ; and They are respectively and Taylor expansion, parameters The next step involves changes in scattering. Calculate the dielectric constant distribution in the sensitive field The calculation formula is: (8) in Represents the coordinates of the sensitive field, where the coordinates of the center of the sensitive field are... , Represents the dielectric constant in the sensitive field coordinates The value of is extracted. The maximum dielectric constant value calculated within the sensitive field is denoted as . dielectric constant The calculation results and the cutoff radius of the scattering transform Related to adjusting the cutoff radius Make Let the cutoff radius be... for , indicating that the container radius is The corresponding calibrated cutoff radius.

[0017] Furthermore, by calibrating containers with different radii, the container radius can be obtained. With cutoff radius The reference table is used and applied to the actual liquid identification process, recording the sensitive field radius as equal to... r Mapping the empty field DN at that time, and setting the radius r Corresponding empty field DN mapping With cutoff radius valueR r The data is stored together in the hardware device for retrieval in subsequent measurements.

[0018] Furthermore, the measurement steps of the ECT capacitance measurement method are as follows: 4) For the N independent electrodes, the first one... Electrode Applying voltage is 1 The DC excitation is applied, and all other electrodes are grounded.

[0019] 5) Measure the capacitance between the excitation electrode and the remaining N-1 electrodes.

[0020] 6) After completing the measurement in step 2), measure the first... Electrode Applying voltage is 1 The DC excitation was applied, with all other electrodes grounded. The capacitance between the excitation electrode and the remaining N-1 electrodes was measured. and The capacitance values ​​between and The capacitance values ​​between them are equal, so only N-2 capacitance values ​​need to be measured.

[0021] 4) Repeat the above steps until all are obtained. Each independent capacitance value.

[0022] Furthermore, step (2) includes liquid detection: The container holding the liquid to be tested has a radius of... The container is placed in the ECT sensitive field, at which point the capacitance measurement module is in the object field state; Measurement Each independent capacitance value is used to set the calibration dielectric constant. Construct the object-field DN mapping matrix ; Constructing Deviation DN Mapping , denoted as: (9) Calculate the charge change matrix under the object field state ; Constructing complex variables The distribution of makes Variables Evenly distributed in a circle centered at (0,0) with a radius of... A two-dimensional circular region, with a cutoff radius Container radius obtained from ECT system calibration procedure With cutoff radius Obtained from the reference table. The cutoff radius... Substituting the values ​​into the formula, we can calculate the dielectric constant distribution within the sensitive field and obtain the initial dielectric constant distribution under the object-field state. DN mapping ( x , y ), ( x , y ) represents the coordinate value of the sensitive field. Substituting the coordinate value into the coordinate value will give the dielectric constant value of that coordinate point.

[0023] Furthermore, step (3) includes dielectric constant calibration: In the context of aviation security checks, the true dielectric constant is distributed within the interval [1, 120]. ε Divide the interval into steps based on the step size. n Sub-intervals λ n ,Right now: (10) in β 1=1, β n+1 = β n Inter , + λ 120 For each sub-interval, the lower limit of that interval is taken as the calibration dielectric constant value corresponding to that interval. The calibration dielectric constant value is then substituted into equation (9) to obtain... n The dielectric constant distribution function is denoted as . n λ = ε n ( z ), where the maximum value of the dielectric constant is , denoted as max( ε n ).

[0024] The closer the calibrated dielectric constant value is to the true dielectric constant, the higher the calculation accuracy of the improved Calderon method. Therefore, calibrating the dielectric constant is crucial. β n With the true dielectric constant ε The closer they are, the higher the maximum value. ε n )and ε absolute deviation The smaller, the definition For the first n The error coefficient of each sub-interval is calculated using the following formula: (11) Then when the true dielectric constant is β i and β i+1 Between, that is βi < ε < β i+1 At this point, we have: .make ,but When step length λ Enough hours ,but The minimum value is: (12) Among them, the minimum value The corresponding calibrated dielectric constant β i and its adjacent rated dielectric constant β i+1 The calculated value used to estimate and correct the dielectric constant is given by the following formula: (13) ε corr The value is used to distinguish the type of substance being measured. e corr For calculated values ε corr Compared with the true value ε The more corrections are performed, the greater the error between them. e corr The smaller, ε corr The closer the result is to the true value, the more accurate the recognition result. In equation (13), ε corr This indicates the dielectric constant correction value. ε est This represents the estimated dielectric constant. e corr Indicates the correction error, where, e corr ≤ λ / λ / Inter 2. This completes one correction process of the segmented interval correction method.

[0025] Furthermore, the absolute deviation between the corrected dielectric constant value and the true dielectric constant does not exceed [a certain value]. Figure 1 2.

[0026] Furthermore, if it is necessary to further improve the correction accuracy, the sub-intervals can be re-evaluated. Figure 2 i = [ β i , β i+1 The intervals are divided into multiple sub-intervals, and a new round of correction is performed.

[0027] Technical effects: This method has the following advantages: (1) Fast calculation speed: the algorithm takes about 0.1s to calculate the dielectric constant and 4s to identify liquids.

[0028] (2) High recognition accuracy, the error between the calculated dielectric constant and the measured dielectric constant value does not exceed 0.01F / m.

[0029] (3) Wide recognition range: It can identify liquids with dielectric constant in the range of [1, 120] F / m, and the range basically covers the prohibited liquids commonly found in aviation security checks. Attached Figure Description

[0030] Figure 3 This is a schematic diagram of the ECT system structure of this application; Figure 4 This is a schematic diagram of the capacitor module structure of this application; Figure 5 This is a schematic diagram illustrating the process of constructing the container radius-truncation radius comparison table in this application; Figure 6 This is a schematic diagram of the segmented interval correction method of this application; Figure 7 This is a schematic diagram of the ECT liquid identification method based on the Calderon method of this application; Figure 8 A schematic diagram of the capacitive sensor modeling structure of this application is shown, where (a) is a schematic diagram of the empty field and (b) is a schematic diagram of the object field; Figure 9 The graph shows the maximum calculated value of the dielectric constant and the error coefficient after the first calibration as a function of the calibrated dielectric constant. Figure 1 The graph shows the maximum calculated value of the dielectric constant and the error coefficient after the second correction as a function of the calibrated dielectric constant. Figure 2 The graph shows the maximum calculated value of the dielectric constant and the error coefficient after the third calibration as a function of the calibrated dielectric constant. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] The features and illustrative embodiments of various aspects of the present invention will now be described in detail. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention may be practiced without requiring some of these specific details. The following description of embodiments is merely intended to provide a better understanding of the invention by illustrating examples of the invention. The invention is by no means limited to any specific setup and method set forth below, but covers any improvements, substitutions, and modifications to the structures, methods, and devices without departing from the spirit of the invention. In the following description, well-known structures and techniques are not shown to avoid unnecessarily obscuring the invention.

[0033] In the description of this invention, it should be noted that the directions or positional relationships indicated by terms such as "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" are based on the stated directions or positional relationships and are only for the convenience of describing and simplifying the invention, and should not be construed as limiting the invention. Furthermore, the use of ordinal numbers (e.g., "first and second," etc.) is for distinguishing objects and is not limited to this order, and should not be construed as indicating or implying relative importance.

[0034] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly, encompassing both direct connection and indirect connection via an intermediate medium. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances.

[0035] It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other, and the various embodiments can be referenced and cited in each other. The present invention will now be described in detail with reference to and in conjunction with the embodiments.

[0036] Example 1: Step 1: Setting up the ECT system The structural diagram of the ECT device is attached. Mapping, As shown, it should include at least three parts: a capacitance measurement module A, a data acquisition and transmission module B, and a computer C for data processing. The capacitance measurement module 1 is attached... MappingAs shown, it should contain no fewer than 8 independent electrodes, and the structure includes an electrode partition 1, a metal shielding shell 2, independent electrodes 3, an internal shielding cover 4, and a sensitive field 5. In the data acquisition and transmission module 2, the sensor cable is an RG174 coaxial cable, with one end connected to the sensor and the other end connected to the acquisition system via an SMB RF connector. The data acquisition system can be connected to the computer using either a USB data cable or a network cable. When using a network cable, the Ethernet IPv4 parameters are set, and the data acquisition system is connected. The computer C used for data processing calculates the dielectric constant using the Calderon method, employing Matlab as the calculation software.

[0037] Step 2: ECT system calibration.

[0038] After completing the ECT system setup, the first step is to calibrate the ECT system. First, set the N independent electrodes in the capacitance measurement module to be marked counter-clockwise. , ,..., You can choose from them. The corresponding electrodes. Measured using traditional ECT capacitance measurement methods. Each independent capacitance value is represented by a symbol. Indicates the first The first independent electrode and the first The capacitance value between N independent electrodes. The measurement steps of the conventional ECT capacitance measurement method are as follows: 1) For the N independent electrodes... Electrode Applying voltage is 1 1) Apply DC excitation, and ground all other electrodes. 2) Measure the capacitance between the excitation electrode and the remaining N-1 electrodes. 3) After completing step 2), measure the capacitance between the excitation electrode and the remaining N-1 electrodes. Electrode Applying voltage is 1 The DC excitation was applied, with all other electrodes grounded. The capacitance between the excitation electrode and the remaining N-1 electrodes was measured. and The capacitance values ​​between and The capacitance values ​​between them are equal, so only N-2 capacitance values ​​need to be measured. 4) Repeat the above steps until all are obtained. Each independent capacitance value.

[0039] After obtaining all After obtaining individual capacitance values, a DN can be constructed. Mapping DN Mapping, Use symbols The calculation formula is as follows: (1) in Indicates the first The self-capacitance of each independent electrode is calculated using the following formula: (2) DN in an empty field DN mapping Use symbols It indicates that the superscript The subscript represents the radius of the sensitive field. DN when empty field DN mapping The specific formula is as follows: (3) The N(N-1) / 2 independent capacitance values ​​in the formula are the capacitance values ​​measured when the ECT sensitive field is an empty field.

[0040] After obtaining the empty field DN mapping Then, the container filled with purified water was placed in the ECT sensitive field, and the entire ECT measurement was performed using the traditional ECT measurement method described above. Individual capacitance values, and construct the DN mapping when the object field is used. .

[0041] Introduce the calibration dielectric constant as a new parameter, and use the symbol This indicates that, since the relative permittivity of pure water at 25°C is close to 78.3 (F / m), in order to improve the calculation accuracy of the algorithm in the permittivity range of 78.3 and its neighborhood, let... and make use of the open space DN mapping and object field Figure 3 Construction deviation ε Matrix, denoted as , The calculation formula is: (4) The next step is to construct the charge change matrix. Constructing a voltage matrix in Matlab The matrix structure is as follows: (5) in ,in Indicates the first s The angle of each electrode, where i is a complex number. j Indicates the first j One voltage excitation mode, Indicates the first s The electrode at the first j The excitation voltage in the voltage excitation mode, for example: the first voltage excitation mode. s The excitation voltage of each electrode in the first voltage excitation mode is , No. j Each voltage excitation mode corresponds to a voltage matrix. V The j List.

[0042] From voltage matrix And deviation DN mapping matrix Calculate the charge change matrix Due to voltage V ,capacitance C With charge q Existence Relationship Therefore, the first j The voltage excitation mode is the first s The change in charge of each electrode is A vector consisting of the changes in charge of N electrodes. The calculation formula is: (6) Using the charge change matrix Calculate scattering transformation Complex variables To be evenly distributed in The point is the center of the circle, and the radius is... within a two-dimensional circular region One point, Generally not less than 1000 and Both are real numbers, representing complex variables respectively. The real and imaginary coordinates of the points within the circular region, and the radius In this application, it is referred to as the cutoff radius, the cutoff radius of The value range is generally between 4 and 6. Scattering transformation of formula (12) Discretization is performed, and the calculation formula is as follows: (7) in The area of ​​the individual electrodes is expressed in units of... ; and They are respectively and Taylor expansion, parameters The next step involves changes in scattering. Calculate the dielectric constant distribution in the sensitive field The calculation formula is: (8) in Represents the coordinates of the sensitive field, where the coordinates of the center of the sensitive field are... , Represents the dielectric constant in the sensitive field coordinates The value of is extracted. The maximum dielectric constant value calculated within the sensitive field is denoted as . dielectric constant The calculation results and the cutoff radius of the scattering transform Related to adjusting the cutoff radius Make Let the cutoff radius be... for , indicating that the container radius is The corresponding calibration cutoff radius is then determined. The above process constitutes the complete calibration process for the ECT system. Further calibration of containers with different radii will yield the container radius. With cutoff radius A reference table was used, and applied to the actual liquid identification process. To facilitate subsequent measurements, the radius of the sensitive field was recorded as equal to... r Mapping the empty field DN at that time, and setting the radius r Corresponding empty field DN mapping With cutoff radius value R r The data is stored together in the hardware device and can be retrieved in subsequent measurements. The method for constructing the container radius-cutoff radius comparison table is attached. λ As shown. The third step is liquid detection. The container holding the liquid to be tested has a radius of... The container is placed in the ECT sensitive field, at which point the capacitance measurement module is in the object field state; Measurement Each independent capacitance value is used to set the calibration dielectric constant. Construct the object-field DN mapping matrix ; Constructing Deviation DN Mapping , denoted as: (9) Calculate the charge change matrix under the object field state ; Constructing complex variables The distribution of makes Variables Evenly distributed in a circle centered at (0,0) with a radius of... A two-dimensional circular region, where the cutoff radius is... Container radius obtained from ECT system calibration procedure With cutoff radius Obtained from the reference table. The cutoff radius... Substituting the values ​​into the formula, we can calculate the dielectric constant distribution within the sensitive field and obtain the initial dielectric constant distribution under the object-field state. Inter ( x, y ), ( x , y ) represents the coordinate value of the sensitive field. Substituting the coordinate value into the coordinate value will give the dielectric constant value of that coordinate point.

[0043] Step 4: Dielectric constant correction Because the dielectric constant of the liquid being measured during ECT measurement is uncertain and variable, manually setting a suitable calibrated dielectric constant is quite difficult. To solve this problem, this invention proposes a dielectric constant estimation and correction method based on an iterative method, which is called the piecewise interval correction method. This method allows the calculated dielectric constant value to converge quickly to the true dielectric constant value of the liquid being measured.

[0044] In the context of aviation security checks, the true dielectric constant is typically distributed within the range [1, 120]. + λ Divide the interval into steps based on the step size. n Sub-intervals n λ = ε n ,Right now: (10) in β 1=1, β n+1 = β n ε , ε 120 For each sub-interval, the lower limit of that interval is taken as the calibration dielectric constant value corresponding to that interval. The calibration dielectric constant value is then substituted into equation (9) to obtain... n The dielectric constant distribution function is denoted as . ε n ( z ), where the maximum value of the dielectric constant is , denoted as max( ε n ).

[0045] The closer the calibrated dielectric constant value is to the true dielectric constant, the higher the calculation accuracy of the improved Calderon method. Therefore, calibrating the dielectric constant is crucial. β n With the true dielectric constant < ε < β The closer they are, the higher the maximum value. λ n )and ε absolute deviation The smaller, the definition For the first n The error coefficient of each sub-interval is calculated using the following formula: (11) Then when the true dielectric constant is β i and β i+1 Between, that is β i ε i+1 At this point, we have: .make ,but When step length ε Enough hours ,but The minimum value is: (12) Among them, the minimum value The corresponding calibrated dielectric constant β i and its adjacent rated dielectric constant β i+1 It can be used to estimate and correct the calculated value of the dielectric constant. The calculation formula is as follows: (13) ε corr The value is used to distinguish the type of substance being measured. e corr For calculated values ε corr Compared with the true value ε The more corrections are performed, the greater the error between them. e corr The smaller, ≤ λ / λ / Inter corr The closer the result is to the true value, the more accurate the recognition result. In equation (13), Figure 4 corr This indicates the dielectric constant correction value. Figure 5 est This represents the estimated dielectric constant. e corr Indicates the correction error, where, e corr Figure 6 2. This completes one correction process using the segmented interval correction method. The absolute deviation between the corrected dielectric constant value and the true dielectric constant does not exceed [a certain value]. Figure 6 2. If further improvement in correction accuracy is required, the sub-intervals can be re-calibrated. ε = ε i = [ β i , β i+1 The interval is divided into multiple sub-intervals, and a new round of correction is performed. The correction process of the segmented interval correction method is shown in the attached figure. εThe flowchart of the ECT liquid recognition method based on the Calderon method is shown in the attached figure. ε As shown.

[0046] Simulation verification The method will be verified through simulation here: The simulation experiment uses a gas-liquid two-phase flow model as the research object. The model simulates the measurement of cylindrical containers in aviation security inspection, using a 12-electrode capacitive sensor. The sensor structure is designed as a three-layer structure, including a metal shielding layer, a sensitive field region, and the liquid to be measured. The materials used are copper, air, and the filling material to be measured, respectively. The corresponding dielectric constants are set to 2.2 F / m, 1 F / m, and ε F / m, where the ε value is set manually according to the experimental purpose. The diameter of the sensitive field is 100 mm, and the outer diameter of the shielding layer is set to 110 mm. The air field and object field models of the capacitive sensor are shown in the attached figures. ε As shown in (a) and (b). This experiment neglects the influence of the container shell on the calculation, therefore... Figure 7 (b) The measured liquid fills the entire sensitive field region within the medium field.

[0047] The experiment was conducted on an Intel(R) Core(TM) i7-12650H processor (2.30 GHz) and Matlab 2020b. A model of an ECT sensor with 12 electrodes was built using COMSOL 5.3. After the model was built, finite element analysis was performed to obtain relevant data, which was then imported into Matlab software. The dielectric constant was then calculated using the Calderon method based on the discrete integral formula.

[0048] Preset some parameters of the Calderon method, among which Z The planar subdivision has 316 pixels. K The planar partitioning has 7860 pixels, and the truncation radius is... R =2.1.

[0049] Verification of corrected dielectric constant The experimental setup sets the true dielectric constant to be [value missing]. Figure 7 At 63.33, the absolute error Δ calculated by the traditional Calderon method is... ε =18.62. The interval [1, 120] is divided into 12 sub-intervals using the segmented interval correction method. , respectively , ... , Set the lower limit of each sub-interval to the calibration dielectric constant value corresponding to that sub-interval. β i They are respectively: β 1=1、β 2=10、...、 β 11 =100、 β 12 =110, and supplement the boundary value calibration dielectric constant value. β 13 =120 is used for data processing and calculating different... β i The maximum calculated value of the dielectric constant under the following conditions is max( ε and the error coefficients of the corresponding sub-intervals. ,max( ε )and The graph showing the variation of the calibrated dielectric constant is attached. Figure 8 As shown, the specific experimental data are shown in Tables 1 and 2.

[0050] From the appendix Figure 8 As can be seen, with the calibration dielectric constant β i The error coefficient increases continuously from 1 to 120, with each interval having its own error coefficient. The trend of change is first decreasing and then increasing, and at... β The value reaches its minimum at 7 = 60, therefore it can be determined that the true dielectric constant is within the range [7, 60, ... Within this range, the estimated value of the dielectric constant is... ε est =65.00, the absolute deviation between the estimated dielectric constant and the true value. e est =1.67, the corrected value for the dielectric constant is ε corr =65.00±5.00, the maximum error of the correction value is e max =6.67.

[0051] To further improve calculation accuracy, Divided into 10 sub-intervals, namely , ... , Repeat the above steps of calibrating the dielectric constant and calculating the numerical value, and perform a second correction, max( ε )and The graph showing the variation of the calibrated dielectric constant is attached. Figure 9 As shown, the specific experimental data are shown in Tables 3 and 4.

[0052] Similarly, attached Figure 9 As the calibration dielectric constant increases, the error coefficient of each sub-interval increases. First decrease, then increase, and when βi = At 63, the error coefficient is the smallest, and it can be determined that the true dielectric constant is within the range Within this range, the estimated value of the dielectric constant is... ε est =63.50, the absolute deviation between the estimated and true dielectric constant values. e est =0.17, the correction value for the dielectric constant is ε corr =63.50±0.50, the maximum error of the correction value is e max =0.67.

[0053] Further Divide into 10 sub-intervals and perform a third correction, max( ε )and The graph showing the variation of the calibrated dielectric constant is attached. ​ As shown, the specific experimental data are shown in Tables 5 and 6.

[0054] From the appendix ​ As shown, the third correction data can determine that the true dielectric constant is within the range Within this range, the estimated value of the dielectric constant is... ​ est =63.35, the absolute deviation between the estimated and true dielectric constant values. e est =0.02, the correction value for the dielectric constant is ​ corr =63.35±0.05, the maximum error of the correction value is e max =0.07. The summary of the three correction data is shown in Table 7.

[0055] The data in Table 7 demonstrates that the piecewise interval correction method can bring the calculated dielectric constant value to the true dielectric constant value. Furthermore, both the estimation error and the correction error between the calculated and true values ​​gradually decrease with increasing correction counts. After three corrections, the calculation error is reduced from the original 18.62 to below 0.1, a reduction of approximately 99.46%. The estimated dielectric constant after three corrections is set as the calibration dielectric constant, i.e., let... β Substituting 63.35 into the algorithm, we obtain the maximum dielectric constant value at this point, max( ​ The result is 63.3729, with an error of 0.0429, which basically meets the requirements of the liquid detection environment in aviation security inspections. Experimental results show that the ECT liquid identification method proposed in this invention can accurately calculate the dielectric constant value of high-contrast dielectric constant materials.

[0056] Table 1 shows the calculated maximum dielectric constant and error coefficient during the first calibration. Table 2 shows the calculated maximum dielectric constant and error coefficient during the first calibration. Table 3 shows the calculated maximum dielectric constant and error coefficient during the second correction. Table 4 is a supplementary table showing the calculated maximum dielectric constant and error coefficients during the second correction. Table 5 shows the calculated maximum dielectric constant and error coefficient during the third correction. Table 6 is a supplementary table showing the calculated maximum dielectric constant and error coefficients during the third correction. Table 7 is a summary table of data from the three corrections. Table 8 shows the calculated maximum correction error values ​​for the liquid recognition experiment, in F / m. Table 9 is a statistical table of calibration time for the liquid recognition experiment, in seconds. Appendix 1

[0057] Appendix 2

[0058] Appendix 3

[0059] Appendix 4

[0060] Appendix 5

[0061] Appendix 6

[0062] Appendix 7

[0063] Appendix 8

[0064] Appendix 9

[0065] The above detailed embodiments are a description of the present invention. It should not be considered that the specific embodiments of the present invention are limited to these descriptions. For those skilled in the art, several simple deductions and substitutions can be made without departing from the concept of the present invention, and all of these should be considered to fall within the protection scope of the present invention.

Claims

1. A liquid identification method based on capacitance tomography, characterized in that, Includes the following steps: Step (1) Use the capacitance data obtained by measuring in the ECT system to artificially construct the DN mapping matrix. Control the size of the DN mapping matrix by adjusting the parameter: calibrated dielectric constant, and use this to simulate the change of the empty field dielectric constant, and adjust it to make the empty field dielectric constant close to the measured dielectric constant. Step (2) Construct a container radius-cutoff radius comparison table and quickly select the optimal cutoff radius value based on the radius of the container being tested; Step (3) Design a segmented interval correction method and search for the calibration dielectric constant in technical solution (1) through iteration; Step (4) Integrating the technical solutions of Step (1), Step (2) and Step (3), design an ECT liquid identification method based on the Calderon method.

2. The method as described in claim 1, characterized in that, Step (1) includes: First, setting up the ECT system. An ECT device should include at least three parts: a capacitance measurement module, a data acquisition and transmission module, and a computer for data processing. The capacitance measurement module should contain no fewer than eight independent electrodes, with a structure including electrode separators, a metal shielding shell, independent electrodes, an internal shielding cover, and a sensitive field. In the data acquisition and transmission module, the sensor cable is an RG174 coaxial cable, with one end connected to the sensor and the other end connected to the acquisition system via an SMB RF connector. The data acquisition system can be connected to the computer using either a USB data cable or a network cable. When using a network cable, the Ethernet IPv4 parameters should be set before connecting to the data acquisition system. The computer for data processing uses the Calderon method to calculate the dielectric constant, employing Matlab as the calculation software.

3. The method as described in claim 2, characterized in that, The second step is ECT system calibration; After completing the ECT system setup, the first step is to calibrate the ECT system. First, set the N independent electrodes in the capacitance measurement module to be marked counter-clockwise. , ,..., You can choose from them. The corresponding electrodes; measured using the ECT capacitance measurement method. Each independent capacitance value is represented by a symbol. Indicates the first The first independent electrode and the first The capacitance value between individual electrodes; After obtaining all After determining the individual capacitance values, a DN mapping can be constructed. The DN mapping uses the symbol... The calculation formula is as follows: (1) in Indicates the first The self-capacitance of each independent electrode is calculated using the following formula: (2) Symbols for DN mapping in empty field It indicates that the superscript The subscript represents the radius of the sensitive field. DN when empty field Mapping, The specific formula is as follows: (3) The N(N-1) / 2 independent capacitance values ​​in the formula are the capacitance values ​​measured when the ECT sensitive field is an empty field; After obtaining the empty field DN mapping Then, the container filled with purified water was placed in the ECT sensitive field, and the entire ECT measurement was performed using the ECT measurement method described above. Individual capacitance values, and construct the DN mapping when the object field is used. ; Introduce the calibration dielectric constant as a new parameter, and use the symbol Indicate; command and make use of the open space DN mapping and object field DN mapping Construction deviation DN mapping Matrix, denoted as , The calculation formula is: (4) The next step is to construct the charge change matrix. Constructing the voltage matrix in Matlab The matrix structure is as follows: (5) in ,in Indicates the first s The angle of each electrode, where i is a complex number. j Indicates the first j One voltage excitation mode, Indicates the first s The electrode at the first j The excitation voltage under the voltage excitation mode, the first s The excitation voltage of each electrode in the first voltage excitation mode is , No. j Each voltage excitation mode corresponds to a voltage matrix. V The j List; From voltage matrix And deviation DN mapping matrix Calculate the charge change matrix Due to voltage V ,capacitance C With charge q Existence Relationship Therefore, the first j The voltage excitation mode is the first s The change in charge of each electrode is A vector consisting of the changes in charge of N electrodes. The calculation formula is: (6) Using the charge change matrix Calculate scattering transformation Complex variables To be evenly distributed in The point is the center of the circle, and the radius is... within a two-dimensional circular region One point, Not less than 1000 and Both are real numbers, representing complex variables respectively. The real and imaginary coordinates of the points within the circular region, and the radius This is called the cutoff radius. The value ranges from 4 to 6; Scattering transformation of formula (12) Discretization is performed, and the calculation formula is as follows: (7) in The area of ​​the individual electrodes is expressed in units of... ; and They are respectively and Taylor expansion, parameters The next step involves changes in scattering. Calculate the dielectric constant distribution in the sensitive field The calculation formula is: (8) in Represents the coordinates of the sensitive field, where the coordinates of the center of the sensitive field are... , Represents the dielectric constant in the sensitive field coordinates The value of ; extract the maximum dielectric constant value calculated within the sensitive field, denoted as . dielectric constant The calculation results and the cutoff radius of the scattering transform Related to adjusting the cutoff radius Make Let the cutoff radius be... for , indicating that the container radius is The corresponding calibrated cutoff radius.

4. The method as described in claim 3, characterized in that, By calibrating containers of different radii, the container radius can be obtained. With cutoff radius The reference table is used and applied to the actual liquid identification process, recording the sensitive field radius as equal to... r Mapping the empty field DN at that time, and setting the radius r Corresponding empty field DN mapping With cutoff radius value R r The data is stored together in the hardware device for retrieval in subsequent measurements.

5. The method as described in claim 4, characterized in that, The measurement steps of the ECT capacitance measurement method are as follows: 1) For the N independent electrodes, the first one... Electrode Applying voltage is 1 The DC excitation is applied, and all other electrodes are grounded. 2) Measure the capacitance between the excitation electrode and the remaining N-1 electrodes; 3) After completing the measurement in step 2), the first... Electrode Applying voltage is 1 The DC excitation was applied, with all other electrodes grounded. The capacitance between the excitation electrode and the remaining N-1 electrodes was measured. and The capacitance values ​​between and The capacitance values ​​between them are equal, so only N-2 capacitance values ​​need to be measured; 4) Repeat the above steps until all are obtained. Each independent capacitance value.

6. The method as described in claim 1, characterized in that, Step (2) includes liquid detection: The container holding the liquid to be tested has a radius of... The container is placed in the ECT sensitive field, at which point the capacitance measurement module is in the object field state; Measurement Each independent capacitance value is used to set the calibration dielectric constant. Construct the object-field DN mapping matrix ; Constructing Deviation DN Mapping , denoted as: (9) Calculate the charge change matrix under the object field state ; Constructing complex variables The distribution of makes Variables Evenly distributed in a circle centered at (0,0) with a radius of... A two-dimensional circular region, with a cutoff radius Container radius obtained from ECT system calibration procedure With cutoff radius Obtain from the reference table; cutoff radius Substituting the values ​​into the formula, we can calculate the dielectric constant distribution within the sensitive field and obtain the initial dielectric constant distribution under the object-field state. ε ( x , y ), ( x , y ) represents the coordinate value of the sensitive field. Substituting the coordinate value into the coordinate value will give the dielectric constant value of that coordinate point.

7. The method as described in claim 1, characterized in that, Step (3) includes dielectric constant calibration: In the context of aviation security checks, the true dielectric constant is distributed within the interval [1, 120]. λ Divide the interval into steps based on the step size. n Sub-intervals Inter n ,Right now: (10) in β 1=1, β n+1 = β n +λ , nλ= 120 For each sub-interval, the lower limit of that interval is taken as the calibration dielectric constant value corresponding to that interval. The calibration dielectric constant value is then substituted into equation (9) to obtain... n The dielectric constant distribution function is denoted as . ε n ( z ), where the maximum value of the dielectric constant is , denoted as max( ε n ); The closer the calibrated dielectric constant value is to the true dielectric constant, the higher the calculation accuracy of the improved Calderon method. Therefore, calibrating the dielectric constant is crucial. β n With the true dielectric constant ε The closer they are, the higher the maximum value. ε n )and ε absolute deviation The smaller, the definition For the first n The error coefficient of each sub-interval is calculated using the following formula: (11) Then when the true dielectric constant is β i and β i+1 Between, that is β i <ε<β i+1 At this point, we have: ;make ,but When step length λ Enough hours ,but The minimum value is: (12) Among them, the minimum value The corresponding calibrated dielectric constant β i and its adjacent rated dielectric constant β i+1 The calculated value used to estimate and correct the dielectric constant is given by the following formula: (13) ε corr The value is used to distinguish the type of substance being measured. e corr For calculated values ε corr Compared with the true value ε The more corrections are performed, the greater the error between them. e corr The smaller, ε corr The closer the result is to the true value, the more accurate the recognition result. In equation (13), ε corr This indicates the dielectric constant correction value. ε est This represents the estimated dielectric constant. e corr Indicates the correction error, where, e corr ≤ λ / 2; This completes one correction process of the segmented interval correction method.

8. The method as described in claim 7, characterized in that, The absolute deviation between the corrected dielectric constant value and the true dielectric constant shall not exceed λ / 2.

9. The method as described in claim 8, characterized in that, If further improvement in correction accuracy is required, the sub-intervals should be re-evaluated. Inter i = [ β i , β i+1 The intervals are divided into multiple sub-intervals, and a new round of correction is performed.

Images