An electrical impedance imaging method and system for structural damage monitoring

Abstract

The present invention provides an electrical impedance imaging method and system for structural damage monitoring, wherein the method includes dividing a geometric model of a measured target into a plurality of grid cells; obtaining boundary voltage data of the resistivity distribution of the measured target; calculating a first resistivity of each grid cell based on the boundary voltage data of the resistivity distribution of the measured target; dividing the geometric model of the measured target into a potential damage region and a non-damage region based on the first resistivity of each grid cell; calculating the iteration step size of the grid cells in the potential damage region and the iteration step size of the grid cells in the non-damage region; calculating a second resistivity of each grid cell based on the iteration step size of each grid cell in the potential damage region and the non-damage region; calculating a grayscale value of each grid cell based on the second resistivity of each grid cell; and performing electrical impedance imaging of the measured target based on the grayscale value of each grid cell. The present invention improves the accuracy of the algorithm solution and the resolution of the imaging.

Description

Technical Field

[0001] The present invention belongs to the technical field of structural damage monitoring, and in particular relates to an electrical impedance imaging method and system for structural damage monitoring. Background Art

[0002] The core components of aircraft engines and gas turbines operate in extreme environments of high temperature and high pressure, and their parts are prone to damage such as mechanical fatigue and thermal ablation. Due to the complex structure and difficulty in disassembling aircraft engines and gas turbines, it is difficult to conduct rapid, real-time in-situ damage monitoring of their internal structures using ultrasonic, infrared thermal imaging, and XCT flaw detection technologies. Electrical Impedance Tomography (EIT), with its advantages of being non-invasive, simple in structure, and fast in response, is expected to make up for the shortcomings of traditional industrial flaw detection technologies and achieve real-time and accurate damage monitoring in high temperature and high pressure environments. This technology uses the voltage inversion of the structure's external boundary to invert the resistivity distribution inside the structure. Monitors can determine whether there is damage inside the structure by analyzing the resistivity distribution.

[0003] However, in the actual application of electrical impedance imaging technology, due to the limited excitation voltage data obtained, which is usually far less than the number of unknown quantities that need to be inverted, most EIT algorithms need to go through multiple iterative calculations to obtain a more accurate resistance value distribution result. Multiple iterations will result in a decrease in computational efficiency. In order to improve the speed of EIT imaging, patent CN102008303B discloses a frequency difference FNOSER imaging method for electrical impedance imaging system. The characteristic of this algorithm is that it only needs to iterate once based on the set initial resistivity to obtain a set of solutions that are approximate to the actual resistivity distribution. Therefore, the frequency difference FNOSER algorithm has a significant speed advantage compared to other algorithms and can meet the engineering needs for monitoring speed. However, its core idea is essentially the earlier EIT reconstruction algorithm NOSER algorithm, also known as the one-step Newton method. The only difference between the FNOSER algorithm and the NOSER algorithm is that the FNOSER algorithm pre-stores the Jacobi matrix and its inverse matrix to be used in the NOSER algorithm, and directly calls them when image reconstruction is required, thereby reducing the algorithm calculation amount and improving the algorithm operation efficiency. However, the FNOSER algorithm suffers from insufficient solution accuracy. Resistance images obtained using the one-step Newton method suffer from low resolution and blurred contours. Therefore, it is necessary to provide an efficient and high-precision electrical impedance imaging method that achieves high computational speed while maintaining high image reconstruction accuracy. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides an electrical impedance imaging method and system for structural damage monitoring.

[0005] In a first aspect, the present invention provides an electrical impedance tomography method for structural damage monitoring, comprising:

[0006] Divide the geometric model of the target to be measured into multiple grid units;

[0007] Obtain boundary voltage data of the resistivity distribution of the measured target;

[0008] Calculating the first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target;

[0009] Dividing the geometric model of the target under test into a potential damage area and a non-damage area according to the first resistivity of each grid cell;

[0010] Calculate the iteration step size of the mesh cells in the potential damage area and the iteration step size of the mesh cells in the damage-free area;

[0011] calculating a second resistivity of each grid cell according to an iteration step size of each grid cell in the potential damage area and the non-damage area;

[0012] Calculating a grayscale value of each grid cell according to the second resistivity of each grid cell;

[0013] The electrical impedance imaging of the target is performed according to the gray value of each grid cell.

[0014] Furthermore, the calculating of the first resistivity of each grid unit according to the boundary voltage data of the resistivity distribution of the measured target includes:

[0015] The first resistivity of each grid cell is calculated according to the following formula:

[0016] σ=σ0-[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U];

[0017] Where σ is the first resistivity of each grid cell; J is the Jacobian matrix; γ is the regularization coefficient; T represents the transpose of the matrix; V(·) is the forward operator for solving the forward problem; U is the boundary voltage data of the measured target resistivity distribution; σ0 is the initial resistivity estimate;

[0018] Calculate an initial resistivity estimate using the following formula:

[0019]

[0020] Among them, U ijis the measured voltage value on the jth pair of electrodes when the ith pair of electrodes is excited by current; V ij is the theoretical value of the voltage on the jth pair of electrodes when the i-th pair of electrodes is excited by current when the resistivity distribution is unit resistivity distribution; the total number of N electrode pairs.

[0021] Furthermore, dividing the geometric model of the target under test into a potential damage area and a non-damage area according to the first resistivity of each grid unit includes:

[0022] Set the grid cell damage threshold, neighborhood radius, and minimum number of neighborhood points;

[0023] Obtain the maximum value of the first resistivity of all grid cells;

[0024] calculating a ratio of the first resistivity of each grid cell to a maximum value of the first resistivities of all grid cells;

[0025] Mark the grid cells whose ratio is greater than the grid cell damage threshold as potential damaged grid cells;

[0026] Obtain the center coordinates of each potential damage grid cell;

[0027] Arbitrarily mark a potential damage grid cell with the target label and use the center of the marked potential damage grid cell as the center of the circle;

[0028] If the number of potentially damaged grid cells within a circle with the center of the marked potentially damaged grid cell as the center and the radius of the neighborhood radius is greater than the minimum number of neighborhood points, the potentially damaged grid cells within the circle with the center of the marked potentially damaged grid cell as the center and the radius of the neighborhood radius will be marked with target labels;

[0029] Traverse all potential damage grid cells to obtain the potential damage area.

[0030] Furthermore, the step of calculating the iteration step of the grid cells in the potential damage area and the iteration step of the grid cells in the damage-free area includes:

[0031] The Goldstein step size is selected according to the following conditions:

[0032] f(σ k +αd k )∈[f(σ k )+(1-ε)αg k T d k ,f(σ k )+εαg k T d k ];

[0033] Among them, d k is the Newton step length, d k =f″(σ k ) -1 f′(σ k );g k =▽f(σ k ), ▽ is the gradient operator, σ k is the resistivity distribution result obtained under the kth iterative calculation; α is the Goldstein step size, which is a one-dimensional vector with the same dimension as the total number of grid cells, α∈[0,1]; ε is the empirical parameter k is the kth iteration; T represents the transpose of the matrix; when the grid cell is located in the potential damage area, the value of α is close to 1; when the grid cell is located in the damage-free area, the value of α is close to 0;

[0034] The product of the Newton step size and the Goldstein step size is used as the iteration step size of the corresponding grid cell.

[0035] Furthermore, the step of calculating the second resistivity of each grid cell according to the iterative step size of each grid cell in the potential damage area and the non-damage area includes:

[0036] The second resistivity of each grid cell is calculated according to the following formula:

[0037] σ′=σ0-α[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U];

[0038] Where σ′ is the second resistivity of each grid cell.

[0039] Furthermore, calculating the grayscale value of each grid cell according to the second resistivity of each grid cell includes:

[0040] The grayscale value of each grid cell is calculated according to the following formula:

[0041]

[0042] Among them, σ i is the second resistivity of the i-th grid cell; σ max is the maximum value of the second resistivity of all grid cells; σ min is the minimum value of the second resistivity of all grid cells.

[0043] In a second aspect, the present invention provides an electrical impedance imaging system for structural damage monitoring, comprising:

[0044] A grid unit division module is used to divide the geometric model of the target to be measured into multiple grid units;

[0045] A boundary voltage data acquisition module is used to obtain boundary voltage data of the resistivity distribution of the measured target;

[0046] A first calculation module is used to calculate the first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target;

[0047] A region division module is used to divide the geometric model of the target under test into a potential damage region and a non-damage region according to the first resistivity of each grid cell;

[0048] The second calculation module is used to calculate the iteration step size of the grid cells in the potential damage area and the iteration step size of the grid cells in the non-damage area;

[0049] a third calculation module, configured to calculate a second resistivity of each grid cell according to an iteration step size of each grid cell in the potential damage area and the non-damage area;

[0050] a fourth calculation module, configured to calculate a grayscale value of each grid cell according to the second resistivity of each grid cell;

[0051] The electrical impedance imaging module is used to perform electrical impedance imaging of the target under test according to the grayscale value of each grid cell.

[0052] Furthermore, the first calculation module includes:

[0053] The first calculation unit is used to calculate the first resistivity of each grid cell according to the following formula:

[0054] σ=σ0-[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U];

[0055] Where σ is the first resistivity of each grid cell; J is the Jacobian matrix; γ is the regularization coefficient; T represents the transpose of the matrix; V(·) is the forward operator for solving the forward problem; U is the boundary voltage data of the measured target resistivity distribution; σ0 is the initial resistivity estimate;

[0056] The second calculation unit is used to calculate the initial resistivity estimate according to the following formula:

[0057]

[0058] Among them, U ijis the measured voltage value on the jth pair of electrodes when the ith pair of electrodes is excited by current; V ij is the theoretical value of the voltage on the jth pair of electrodes when the i-th pair of electrodes is excited by current when the resistivity distribution is unit resistivity distribution; the total number of N electrode pairs.

[0059] Furthermore, the area division module includes:

[0060] Numerical setting unit, used to set the grid unit damage threshold, neighborhood radius and minimum number of neighborhood points;

[0061] A first obtaining unit is used to obtain a maximum value of the first resistivities of all grid cells;

[0062] a third calculating unit, configured to calculate a ratio of the first resistivity of each grid cell to a maximum value of the first resistivities of all grid cells;

[0063] A first marking unit is used to mark a grid cell corresponding to a ratio greater than a grid cell damage threshold as a potential damaged grid cell;

[0064] A second acquisition unit is used to obtain the center coordinates of each potential damage grid unit;

[0065] a second marking unit, configured to mark any potential damage grid unit with a target label and use the center of the marked potential damage grid unit as the center of a circle;

[0066] a third marking unit, configured to mark the potential damaged grid cells within the circle with the marked potential damaged grid cell center as the center and the radius as the neighborhood radius with the target label when the number of potential damaged grid cells within the circle with the marked potential damaged grid cell center as the center and the radius as the neighborhood radius is greater than the minimum number of neighborhood points;

[0067] The traversal unit is used to traverse all potential damage grid cells to obtain the potential damage area.

[0068] Furthermore, the second calculation module includes:

[0069] The step size selection unit is used to select the Goldstein step size according to the following conditions:

[0070]

[0071] Among them, d k is the Newton step length, d k =f″(σ k ) -1 f′(σ k ); is the gradient operator, σk is the resistivity distribution result obtained under the kth iterative calculation; α is the Goldstein step size, which is a one-dimensional vector with the same dimension as the total number of grid cells, α∈[0,1]; ε is the empirical parameter k is the kth iteration; T represents the transpose of the matrix; when the grid cell is located in the potential damage area, the value of α is close to 1; when the grid cell is located in the damage-free area, the value of α is close to 0;

[0072] The fourth calculation unit is configured to use the product of the Newton step size and the Goldstein step size as the iteration step size of the corresponding grid unit.

[0073] The present invention provides an electrical impedance imaging method and system for structural damage monitoring, wherein the method comprises dividing a geometric model of a measured target into a plurality of grid cells; acquiring boundary voltage data of the resistivity distribution of the measured target; calculating a first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target; dividing the geometric model of the measured target into a potential damage area and a non-damage area according to the first resistivity of each grid cell; calculating the iteration step length of the grid cells in the potential damage area and the iteration step length of the grid cells in the non-damage area; calculating the second resistivity of each grid cell according to the iteration step length of each grid cell in the potential damage area and the non-damage area; calculating the grayscale value of each grid cell according to the second resistivity of each grid cell; and performing electrical impedance imaging of the measured target according to the grayscale value of each grid cell.

[0074] Compared with the traditional NOSER algorithm, this method divides the target into potentially damaged and undamaged areas and applies different iteration steps to different areas, improving the algorithm's solution accuracy and imaging resolution. Traditional EIT algorithms, such as the Newton-Raphson algorithm, the MNR algorithm, and the Landweber algorithm, require repeated calculations of the Jacobian matrix. However, the proposed algorithm only calculates the Jacobian matrix once throughout the entire process, resulting in higher computational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0076] Figure 1 A flowchart of an electrical impedance imaging method for structural damage monitoring provided by an embodiment of the present invention;

[0077] Figure 2 A schematic diagram of the structure of a damage identification algorithm provided by an embodiment of the present invention;

[0078] Figure 3 A comparison chart of imaging effects provided by an embodiment of the present invention;

[0079] Figure 4 A flowchart of a damage identification algorithm provided by an embodiment of the present invention;

[0080] Figure 5 A schematic structural diagram of an electrical impedance imaging system for structural damage monitoring provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0082] In one embodiment, if Figure 1 As shown, the present invention provides an electrical impedance imaging method for structural damage monitoring, comprising:

[0083] Step 101: Divide the geometric model of the target into a plurality of grid units.

[0084] The finite element method is used to divide the geometric model of the target to be measured into multiple grid units.

[0085] Step 102: Acquire boundary voltage data of the resistivity distribution of the measured target.

[0086] Multiple electrodes are arranged around the target to be measured. During the measurement process, a DC constant current source is used to alternately excite two adjacent electrodes. The electrode with current excitation becomes the excitation electrode. During the excitation process, the potential difference between adjacent electrodes other than the excitation electrode is collected to obtain the boundary voltage data required to generate a frame of image. An example is shown below:

[0087] Assume that multiple electrodes are arranged around a rectangular plate, and a DC constant current source is used to alternately excite two adjacent electrodes. At the same time, the potential difference between adjacent electrodes other than the excitation electrode is collected. After the measurement is completed, the electrode to which the excitation current is applied is changed, and the corresponding potential difference measurement is performed. When all the arranged electrodes become over-excitation electrodes, all the voltage data is recorded, which can be recorded as vector U.

[0088] The number of electrodes in this embodiment may be 16, but is not limited thereto. The present invention does not impose any specific limitation on the number of electrodes.

[0089] It should be noted that in this embodiment, in addition to being able to arrange electrodes around rectangular plates, electrodes can also be arranged around solid structural members of circular or other shapes. The specific arrangement targets depend on the structure to be monitored for damage in actual projects and are not specifically limited in this invention.

[0090] Step 103 : Calculate the first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target.

[0091] The first resistivity of each grid cell is calculated according to the following formula:

[0092] σ=σ0-[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U].

[0093] where σ is the first resistivity of each grid cell; J is the Jacobian matrix; γ is the regularization coefficient; T represents the transpose of the matrix; V(·) is the forward operator for solving the forward problem; U is the boundary voltage data of the resistivity distribution of the measured target; and σ0 is the initial resistivity estimate.

[0094] Calculate an initial resistivity estimate using the following formula:

[0095]

[0096] Among them, U ij is the measured voltage value on the jth pair of electrodes when the ith pair of electrodes is excited by current; V ij is the theoretical value of the voltage on the jth pair of electrodes when the i-th pair of electrodes is excited by current when the resistivity distribution is unit resistivity distribution; the total number of N electrode pairs.

[0097] Step 104 : Divide the geometric model of the target to be measured into a potential damage area and a non-damage area according to the first resistivity of each grid cell.

[0098] First, the grid unit damage threshold β, the neighborhood radius r, and the minimum number of neighborhood points MinPts are set; wherein the grid unit damage threshold needs to be selected according to the actual material properties, and in this embodiment it is 0.7.

[0099] Get the maximum value among the first resistivities of all grid cells.

[0100] The ratio of the first resistivity of each grid cell to the maximum value of the first resistivities of all grid cells is calculated.

[0101] The grid cells corresponding to the ratios greater than the grid cell damage threshold are marked as potential damaged grid cells.

[0102] Get the center coordinates of each potential damage grid cell.

[0103] Arbitrarily mark a potential damage grid cell with the target label and take the center of the marked potential damage grid cell as the center of the circle.

[0104] When the number of potential damaged grid cells within a circle with the center of the marked potential damaged grid cell as the center and a radius of the neighborhood radius is greater than the minimum number of neighborhood points, the potential damaged grid cells within a circle with the center of the marked potential damaged grid cell as the center and a radius of the neighborhood radius will be marked with target labels.

[0105] Traverse all potential damage grid cells to obtain the potential damage area.

[0106] Alternatively, as Figure 4 As shown, all grid cells marked as potential damage are simplified into data points, each of which contains the cell number and the coordinates of the center point of the grid cell.

[0107] Among all the data points, first randomly mark one data point with the target label, such as Figure 2 As shown in a, in this embodiment, it is represented by black dots. If the number of data points in the circle with a radius of r and a randomly marked data point as the center is greater than MinPts, the mark will propagate to the surroundings, and the propagated data points will obtain the same target label as the randomly marked data point, as shown in Figure 2 As shown in b; if the number of data points in the circle is less than MinPts, the mark propagation is terminated.

[0108] For a data point that has newly obtained a target label, the ability to continue to spread the target label is determined by checking whether the number of other data points within the circular domain with a radius of r is greater than MinPts. At the same time, it is stipulated that the data points that have spread the target label to the surrounding area will also lose the ability to spread, such as Figure 2 As shown in c and d.

[0109] When all data points with the same target label as the randomly labeled data points no longer have the ability to propagate, the algorithm will generate a new label to propagate among the data points that have not been propagated, such as Figure 2 As shown in Figure e. After all data points have been traversed, data points with the same label are considered to belong to the same cluster. Each cluster of data points is represented as a damage area in the macro sense, as shown in Figure e. Figure 2 As shown in Figure 5, scattered data points far away from the cluster do not meet the label propagation conditions (the number is less than MinPts) and cannot obtain labels. Therefore, they will be judged as noise points or error data points, and the areas corresponding to these points will be judged as damage-free areas.

[0110] After processing the data points using the damage identification algorithm described above, the potential damage area can be delineated.

[0111] Step 105 : Calculate the iteration step size of the grid cells in the potential damage area and the iteration step size of the grid cells in the non-damage area.

[0112] If the data point corresponding to a grid cell has the same label as the data point in the potential damage area, the grid cell is considered to be in the potential damage area.

[0113] For the grid cells within the potential damage area, the Newton method is used to iterate them again.

[0114] The Goldstein step size is selected according to the following conditions:

[0115]

[0116] Among them, d k is the Newton step length, d k =f″(σ k ) -1 f′(σ k ); is the gradient operator, σ k is the resistivity distribution result obtained under the kth iterative calculation; α is the Goldstein step size, which is a one-dimensional vector with the same dimension as the total number of grid cells, α∈[0,1]; ε is the empirical parameter k is the kth iteration; T represents the transpose of the matrix; when the grid cell is located in the potential damage area, the value of α is close to 1; when the grid cell is located in the damage-free area, the value of α is close to 0;

[0117] The product of the Newton step size and the Goldstein step size is used as the iteration step size of the corresponding grid cell.

[0118] Step 106 : Calculate the second resistivity of each grid cell according to the iteration step size of each grid cell in the potential damage area and the non-damage area.

[0119] The second resistivity of each grid cell is calculated according to the following formula:

[0120] σ′=σ0-α[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U].

[0121] Where σ′ is the second resistivity of each grid cell.

[0122] Step 107 : Calculate the grayscale value of each grid cell according to the second resistivity of each grid cell.

[0123] The grayscale value of each grid cell is calculated according to the following formula:

[0124]

[0125] Among them, σ i is the second resistivity of the i-th grid cell; σ max is the maximum value of the second resistivity of all grid cells; σ min is the minimum value of the second resistivity of all grid cells.

[0126] Step 108 : Perform electrical impedance imaging of the target according to the grayscale value of each grid cell.

[0127] For this embodiment, the imaging speed of the conventional NOSER algorithm is 15-16 ms / frame, while the imaging speed of the electrical impedance imaging method for structural damage monitoring provided by the present invention is 16-17 ms / frame, which does not incur significant time cost. Figure 3 It can be seen from the comparison in that the electrical impedance imaging method for structural damage monitoring provided by the present invention has obvious advantages in image contrast and boundary clarity, and effectively eliminates artifacts in the edge area of ​​the image.

[0128] Based on the same inventive concept, an embodiment of the present invention also provides an electrical impedance imaging system for structural damage monitoring. Since the principle of solving the problem by this system is similar to that of the aforementioned electrical impedance imaging method for structural damage monitoring, the implementation of this system can refer to the implementation of the electrical impedance imaging method for structural damage monitoring, and the repeated parts will not be repeated.

[0129] In another embodiment, an electrical impedance imaging system for structural damage monitoring provided by an embodiment of the present invention is as follows: Figure 5 As shown, including:

[0130] The grid unit division module 10 is used to divide the geometric model of the target to be measured into multiple grid units.

[0131] The boundary voltage data acquisition module 20 is used to acquire boundary voltage data of the resistivity distribution of the measured target.

[0132] The first calculation module 30 is configured to calculate the first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target.

[0133] The region division module 40 is configured to divide the geometric model of the target to be measured into a potential damage region and a non-damage region according to the first resistivity of each grid cell.

[0134] The second calculation module 50 is used to calculate the iteration step size of the grid cells in the potential damage area and the iteration step size of the grid cells in the non-damage area.

[0135] The third calculation module 60 is configured to calculate the second resistivity of each grid cell according to the iteration step size of each grid cell in the potential damage area and the non-damage area.

[0136] The fourth calculation module 70 is configured to calculate the grayscale value of each grid cell according to the second resistivity of each grid cell.

[0137] The electrical impedance imaging module 80 is used to perform electrical impedance imaging of the target according to the grayscale value of each grid cell.

[0138] Optionally, the first calculation module includes:

[0139] The first calculation unit is used to calculate the first resistivity of each grid cell according to the following formula:

[0140] σ=σ0-[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U].

[0141] where σ is the first resistivity of each grid cell; J is the Jacobian matrix; γ is the regularization coefficient; T represents the transpose of the matrix; V(·) is the forward operator for solving the forward problem; U is the boundary voltage data of the resistivity distribution of the measured target; and σ0 is the initial resistivity estimate.

[0142] The second calculation unit is used to calculate the initial resistivity estimate according to the following formula:

[0143]

[0144] Among them, U ij is the measured voltage value on the jth pair of electrodes when the ith pair of electrodes is excited by current; V ij is the theoretical value of the voltage on the jth pair of electrodes when the i-th pair of electrodes is excited by current when the resistivity distribution is unit resistivity distribution; the total number of N electrode pairs.

[0145] Optionally, the area division module includes:

[0146] The numerical setting unit is used to set the grid cell damage threshold, neighborhood radius and minimum number of neighborhood points.

[0147] The first acquisition unit is configured to acquire a maximum value among the first resistivities of all grid cells.

[0148] The third calculation unit is configured to calculate a ratio of the first resistivity of each grid cell to a maximum value among the first resistivities of all grid cells.

[0149] The first marking unit is used to mark the grid unit corresponding to the ratio being greater than the grid unit damage threshold as a potential damaged grid unit.

[0150] The second acquisition unit is used to obtain the center coordinates of each potential damage grid unit.

[0151] The second marking unit is used to arbitrarily mark a potential damage grid unit with a target label and take the center of the marked potential damage grid unit as the center of a circle.

[0152] The third marking unit is used to mark the potential damaged grid cells within the circle with the center of the marked potential damaged grid cell as the center and the radius as the neighborhood radius with the target label when the number of potential damaged grid cells within the circle with the center of the marked potential damaged grid cell as the center and the radius as the neighborhood radius is greater than the minimum number of neighborhood points.

[0153] The traversal unit is used to traverse all potential damage grid cells to obtain the potential damage area.

[0154] Optionally, the second calculation module includes:

[0155] The step size selection unit is used to select the Goldstein step size according to the following conditions:

[0156]

[0157] Among them, d k is the Newton step length, d k =f″(σ k ) -1 f′(σ k ); is the gradient operator, σ k is the resistivity distribution result obtained under the kth iterative calculation; α is the Goldstein step size, which is a one-dimensional vector with the same dimension as the total number of grid cells, α∈[0,1]; ε is the empirical parameter k is the kth iteration; T represents the transpose of the matrix; when the grid cell is located in the potential damage area, the value of α is close to 1; when the grid cell is located in the damage-free area, the value of α is close to 0;

[0158] The fourth calculation unit is configured to use the product of the Newton step size and the Goldstein step size as the iteration step size of the corresponding grid unit.

[0159] Optionally, the third calculation module includes:

[0160] The fifth calculation unit is configured to calculate the second resistivity of each grid cell according to the following formula:

[0161] σ′=σ0-α[[J(σ0)] T J(σ0)+γdiag(J T J)] -1 [J(σ0)][V(σ0)-U].

[0162] Where σ′ is the second resistivity of each grid cell.

[0163] Optionally, the fourth calculation module includes:

[0164] The sixth calculation unit is used to calculate the grayscale value of each grid cell according to the following formula:

[0165]

[0166] Among them, σ i is the second resistivity of the i-th grid cell; σ max is the maximum value of the second resistivity of all grid cells; σ min is the minimum value of the second resistivity of all grid cells.

[0167] For more specific working processes of the above modules, please refer to the corresponding contents disclosed in the above embodiments, which will not be repeated here.

[0168] The present invention has been described in detail above with reference to specific embodiments and exemplary examples. However, these descriptions should not be construed as limiting the present invention. Those skilled in the art will appreciate that various equivalent substitutions, modifications, or improvements may be made to the technical solutions and implementations of the present invention without departing from the spirit and scope of the present invention, all of which fall within the scope of the present invention. The scope of protection of the present invention shall be determined by the appended claims.

Claims

1. An electrical impedance tomography method for structural damage monitoring, characterized in that: include: Divide the geometric model of the target to be measured into multiple grid units; Obtain boundary voltage data of the resistivity distribution of the measured target; Calculating the first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target specifically includes: The first resistivity of each grid cell is calculated according to the following formula: ； in, σ is the first resistivity of each grid cell; J is the Jacobian matrix; γ is the regularization coefficient; T Represents the transpose of a matrix; V (•) Forward operator for solving the forward problem; U Boundary voltage data of the resistivity distribution of the measured target; σ 0 is the initial resistivity estimate; Calculate an initial resistivity estimate using the following formula: ； in, U ij For the i When the electrodes are excited by current, j The measured voltage value on the electrode; V ij When the resistivity distribution is unit resistivity distribution, i When the current is excited to the electrode, j Theoretical value of voltage on the counter electrode; N The total number of electrode pairs; Dividing the geometric model of the target under test into a potential damage area and a non-damage area according to the first resistivity of each grid cell; Calculate the iteration step size of the mesh cells in the potential damage area and the iteration step size of the mesh cells in the damage-free area; calculating a second resistivity of each grid cell according to an iteration step size of each grid cell in the potential damage area and the non-damage area; Calculating a grayscale value of each grid cell according to the second resistivity of each grid cell; The electrical impedance imaging of the target is performed according to the gray value of each grid cell.

2. The electrical impedance tomography method for structural damage monitoring according to claim 1, characterized in that: The method of dividing the geometric model of the target under test into a potential damage area and a non-damage area according to the first resistivity of each grid cell includes: Set the grid cell damage threshold, neighborhood radius, and minimum number of neighborhood points; Obtain the maximum value of the first resistivity of all grid cells; calculating a ratio of the first resistivity of each grid cell to a maximum value of the first resistivities of all grid cells; The grid cells whose ratio is greater than the grid cell damage threshold are marked as potential damaged grid cells; Obtain the center coordinates of each potential damage grid cell; Arbitrarily mark a potential damage grid cell with the target label and use the center of the marked potential damage grid cell as the center of the circle; If the number of potentially damaged grid cells within a circle with the center of the marked potentially damaged grid cell as the center and the radius of the neighborhood radius is greater than the minimum number of neighborhood points, the potentially damaged grid cells within the circle with the center of the marked potentially damaged grid cell as the center and the radius of the neighborhood radius will be marked with target labels; Traverse all potential damage grid cells to obtain the potential damage area.

3. The electrical impedance tomography method for structural damage monitoring according to claim 1, characterized in that: The calculating of the iterative step length of the grid cells in the potential damage area and the iterative step length of the grid cells in the damage-free area includes: The Goldstein step size is selected according to the following conditions: ； in, d k is the Newton step length, ; , is the gradient operator, f ( σ )= ; σ k For the k The resistivity distribution results obtained by the iterative calculation; α is the Goldstein step size, a one-dimensional vector with the same dimension as the total number of grid cells, α ∈[0, 1]; ε is the empirical parameter ; k For the k iterations; T represents the transpose of the matrix; when the grid cell is located in the potential damage area, α The value of is close to 1; when the grid cell is located in the damage-free area, α The value of is close to 0; The product of the Newton step size and the Goldstein step size is used as the iteration step size of the corresponding grid cell.

4. The electrical impedance tomography method for structural damage monitoring according to claim 3, characterized in that: The step of calculating the second resistivity of each grid cell according to the iterative step length of each grid cell in the potential damage area and the non-damage area includes: The second resistivity of each grid cell is calculated according to the following formula: ； in, is the second resistivity of each grid cell.

5. The electrical impedance tomography method for structural damage monitoring according to claim 4, characterized in that: Calculating the grayscale value of each grid cell according to the second resistivity of each grid cell includes: The grayscale value of each grid cell is calculated according to the following formula: ； in, σ i For the i The second resistivity of each grid cell; σ max is the maximum value of the second resistivity of all grid cells; σ min is the minimum value of the second resistivity of all grid cells.

6. An electrical impedance tomography system for structural damage monitoring, characterized in that: include: A grid unit division module is used to divide the geometric model of the target to be measured into multiple grid units; A boundary voltage data acquisition module is used to obtain boundary voltage data of the resistivity distribution of the measured target; A first calculation module is used to calculate the first resistivity of each grid cell according to the boundary voltage data of the resistivity distribution of the measured target; The first calculation module includes: The first calculation unit is used to calculate the first resistivity of each grid cell according to the following formula: ； in, σ is the first resistivity of each grid cell; J is the Jacobian matrix; γ is the regularization coefficient; T Represents the transpose of a matrix; V (•) Forward operator for solving the forward problem; U Boundary voltage data of the resistivity distribution of the measured target; σ 0 is the initial resistivity estimate; The second calculation unit is used to calculate the initial resistivity estimate according to the following formula: ； in, U ij For the i When the electrodes are excited by current, j The measured voltage value on the electrode; V ij When the resistivity distribution is unit resistivity distribution, i When the current is excited to the electrode, j Theoretical value of voltage on the counter electrode; N The total number of electrode pairs; A region division module is used to divide the geometric model of the measured target into a potential damage area and a non-damage area according to the first resistivity of each grid cell; The second calculation module is used to calculate the iteration step size of the grid cells in the potential damage area and the iteration step size of the grid cells in the non-damage area; a third calculation module, configured to calculate a second resistivity of each grid cell according to an iteration step size of each grid cell in the potential damage area and the non-damage area; a fourth calculation module, configured to calculate a grayscale value of each grid cell according to the second resistivity of each grid cell; The electrical impedance imaging module is used to perform electrical impedance imaging of the target under test according to the grayscale value of each grid cell.

7. The electrical impedance tomography system for structural damage monitoring according to claim 6, characterized in that: The area division module includes: Numerical setting unit, used to set the grid unit damage threshold, neighborhood radius and minimum number of neighborhood points; A first obtaining unit is used to obtain a maximum value of the first resistivities of all grid cells; a third calculating unit, configured to calculate a ratio of the first resistivity of each grid cell to a maximum value of the first resistivities of all grid cells; A first marking unit is used to mark a grid cell corresponding to a ratio greater than a grid cell damage threshold as a potential damaged grid cell; A second acquisition unit is used to obtain the center coordinates of each potential damage grid unit; a second marking unit, configured to mark any potential damage grid unit with a target label and use the center of the marked potential damage grid unit as the center of a circle; a third marking unit, configured to mark the potential damaged grid cells within the circle with the marked potential damaged grid cell center as the center and the radius as the neighborhood radius with the target label when the number of potential damaged grid cells within the circle with the marked potential damaged grid cell center as the center and the radius as the neighborhood radius is greater than the minimum number of neighborhood points; The traversal unit is used to traverse all potential damage grid cells to obtain the potential damage area.

8. The electrical impedance tomography system for structural damage monitoring according to claim 6, characterized in that: The second calculation module includes: The step size selection unit is used to select the Goldstein step size according to the following conditions: ； in, d k is the Newton step length, ; , is the gradient operator, f ( σ )= ; σ k For the k The resistivity distribution results obtained by the iterative calculation; α is the Goldstein step size, a one-dimensional vector with the same dimension as the total number of grid cells, α ∈[0, 1]; ε is the empirical parameter ; k For the k iterations; T represents the transpose of the matrix; when the grid cell is located in the potential damage area, α The value of is close to 1; when the grid cell is located in the damage-free area, α The value of is close to 0; The fourth calculation unit is configured to use the product of the Newton step size and the Goldstein step size as the iteration step size of the corresponding grid unit.

Images