Multi-dimensional feature quantitative inversion method for cable defects based on local capacitance distortion mapping

Abstract

This invention provides a quantitative inversion method for multidimensional features of cable defects based on local capacitance distortion mapping. By acquiring capacitance data under normal and fault conditions, and utilizing the asymmetry and ratio relationship of local capacitance changes, it achieves decoupling and independent inversion of multidimensional defect features, enabling quantitative assessment of cable defects. The method includes: Step 1, defect angle position inversion; inputting the measured mutual capacitance change of the defect, qualitatively judging the angle position interval, constructing defect angle position feature parameters to obtain the relative angle of the defect, and inverting the defect angle position; Step 2, defect radial depth inversion; inputting the measured self-capacitance change and mutual capacitance change of the defect, and the defect angle position, and inverting the defect radial depth; Step 3, defect size and dielectric constant inversion; inputting the measured self-capacitance change and mutual capacitance change of the defect, the defect angle position, and the defect radial depth, and obtaining the defect size and dielectric constant based on size optimization and inversion.

Description

Technical Field

[0001] This invention relates to the field of cable defect detection technology, and specifically to a method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping. Background Technology

[0002] High-voltage cables are the core hub of power system transmission. During manufacturing, installation, and long-term operation, latent defects such as air gaps, impurities, and water trees can easily develop inside the cable insulation layer. These defects are difficult to detect in the early stages of operation, but under long-term electric field stress, they may cause local electric field distortion, accelerate insulation aging, and ultimately lead to cable breakdown failure.

[0003] Existing technologies for detecting defects in cable joints mainly include:

[0004] 1. Partial discharge detection: This method identifies insulation defects by monitoring discharge signals, but it is not sensitive enough to early latent air gaps or microcracks.

[0005] 2. Ultrasonic and X-ray inspection: These methods can detect some defects, but require contact or invasive operations and have limited effectiveness in detecting high-resistance insulating materials such as cross-linked polyethylene (XLPE) and ethylene propylene diene monomer (EPDM).

[0006] 3. Infrared and dielectric damage detection: suitable for overall operational status assessment, but cannot perform visual imaging and quantitative analysis of local defects.

[0007] Although existing cable insulation testing methods can detect defects to some extent, they still have the following objective drawbacks:

[0008] 1. Difficulty in detecting early latent defects

[0009] Partial discharge detection relies on obvious discharge signals and is only easily detected when defects have developed to a certain extent. It struggles to detect minute air gaps, cracks, and impurities that initially form within cable insulation. This invention uses an electrode array to acquire capacitance signals and perform defect detection and feature inversion, enabling sensitive detection of minute changes caused by defects, thus revealing latent defects.

[0010] 2. Not sensitive enough to highly insulating materials

[0011] High resistivity materials such as XLPE and EPDM are commonly used in cable joints, but traditional methods such as ultrasonic, X-ray, and dielectric loss detection are not sensitive to dielectric changes in these materials, resulting in limited detection effectiveness. This invention utilizes the sensitivity of ECT (Electro-Conductivity Electron) to capacitance changes in high resistivity materials, enabling more accurate defect identification.

[0012] 3. The detection method has limitations.

[0013] Ultrasonic and X-ray inspections often require contact with or coupling to a medium or radiation source, which is not only complex to operate but also poses safety concerns and limitations in application scenarios. This invention employs a non-contact, non-invasive inspection method, enabling offline inspection without damaging the cable structure, thus offering greater applicability.

[0014] While existing ECT (Electro-Conducting Electrode) technology overcomes the limitations of traditional methods, it lacks quantitative feature inversion results. Furthermore, current mainstream ECT data processing relies heavily on image reconstruction algorithms. These algorithms suffer from image artifacts caused by the "soft field effect," and the physical features of defects, such as angle, depth, and size, are coupled within the image, providing only a qualitative, fuzzy distribution rather than a direct representation of the defect's specific characteristics. Therefore, this paper aims to develop a method that abandons traditional image reconstruction and directly analyzes local capacitance changes to achieve cable defect identification and quantitative feature inversion, effectively overcoming the problems of existing technologies. Summary of the Invention

[0015] This invention aims to address the technical deficiencies of existing technologies by providing a quantitative inversion method for multidimensional features of cable defects based on local capacitance distortion mapping. By acquiring capacitance data under normal and fault conditions, and directly utilizing the asymmetry and ratio relationship of local capacitance changes, it achieves decoupling and independent inversion of multidimensional defect features, thereby enabling quantitative assessment of cable defect features.

[0016] This invention provides the following technical solution: a quantitative inversion method for multidimensional features of cable defects based on local capacitance distortion mapping, the method comprising the following steps:

[0017] Step 1, Defect Angle Position Inversion; Input the measured change in mutual capacitance of the defect, qualitatively determine the angle position interval, and construct the defect angle position feature parameters. Substitute the angle and position feature parameters Determine the relative angle of the defect Inversion defect angle position ;

[0018] Step 2, Defect radial depth inversion; input the measured changes in defect self-capacitance and mutual capacitance, and the defect angular position. Construct initial radial characteristic parameters Substitute the initial radial characteristic parameters Determine the angle correction factor The corrected radial depth parameters were then obtained. Inversion of defect radial depth ;

[0019] Step 3: Defect size and dielectric constant inversion; input the measured changes in defect self-capacitance and mutual capacitance, and the defect angular position. radial depth of defects A baseline feature library with different defect sizes is constructed through simulation. Based on its input, the closest radial depth data set is matched. Within the closest radial depth data set, the capacitance value change is equivalently converted to the measured self-capacitance change and measured mutual capacitance change at a standard angle, and a capacitance scaling factor is constructed. Based on capacitance scaling factor Defect size is obtained by performing size optimization and inversion. According to the capacitance scaling factor Determine the dielectric constant .

[0020] Furthermore, the measured change in mutual capacitance of defects in step 1 is acquired by constructing a capacitance matrix of defect-free and defective cables. The acquisition steps include: the detection device acquires the cable capacitance signal through an array of electrodes. Under normal operating conditions without defects, the system sequentially excites each electrode according to a preset amplitude as the excitation voltage and receives the response signal of each electrode. The response of all electrodes under different excitation conditions is recorded, and a full capacitance response matrix can be obtained. This matrix data is used to construct a symmetrical capacitance matrix under normal operating conditions. :

[0021] ,

[0022] When a defect appears inside the cable insulation layer, the distribution of the insulation medium changes. The system sequentially excites the electrodes according to a preset amplitude as the excitation voltage and receives the response signals of each electrode. It records the response of all electrodes under different excitation conditions and collects the capacitance matrix under defect conditions. Define the measured change in mutual capacitance of defects. :

[0023] ,in, , The excitation and measurement electrodes are numbered respectively. When a defect appears inside the cable insulation layer, the response of the electrode plate corresponding to the defect will be measured in terms of the change in mutual capacitance of the defect. Within the feature matrix.

[0024] Furthermore, in step 1, the qualitative judgment of the angular position range is based on the geometric symmetry of the electrode circumference distribution. Every two adjacent electrodes form a 45-degree detection range. When a defect occurs, the two adjacent electrodes closest to the defect have the largest changes in self-capacitance and mutual capacitance. The set of adjacent electrodes that have been positioned is then designated as the current electrode. The starting angle of the corresponding interval With the next electrode The corresponding interval termination angle This allows for a qualitative assessment that the defect is located within the 45-degree range.

[0025] Furthermore, in step 1, the defect angle position feature parameters are constructed. Its mathematical expression is:

[0026] ,

[0027] Among them, The previous electrode of the current electrode, The next electrode after the next electrode. This represents the change in mutual capacitance between the current electrode and the next electrode. This represents the change in mutual capacitance between the previous electrode and the current electrode. This represents the change in mutual capacitance between the next electrode and the electrode after that.

[0028] Furthermore, the relative angle of the defect in step 1 Based on the inversion of the cubic polynomial mapping relation, the mapping relation satisfies:

[0029] Substitute the angle and position feature parameters Determine the relative angle of the defect ;

[0030] .

[0031] Furthermore, in step 2, the initial radial characteristic parameters are constructed. Its mathematical expression is:

[0032] ,

[0033] in, This represents the change in self-capacitance of the current electrode.

[0034] Furthermore, the angle correction factor Its mathematical expression is:

[0035] ,

[0036] in, , and These represent the fault at the actual angle, directly opposite the current electrode, and at a relative angle. The corresponding radial characteristic parameters;

[0037] Angle correction factor Based on the inversion of the cubic polynomial mapping relation, the mapping relation satisfies:

[0038] ,

[0039] .

[0040] Furthermore, the radial depth of the inversion defect in step 2 The corrected radial depth parameters were obtained through simulation fitting. With the radial depth of the defect The second-order characteristic equation:

[0041] .

[0042] Furthermore, in step 3, a baseline feature library of different defect sizes is constructed through simulation. A defect capacitance feature database under standard conditions is pre-established, the dielectric constant of the baseline defect is set, a standard angular position is selected, and multiple sets of radial depth nodes are used. At each depth node, a series of defects with continuously increasing sizes are simulated until they reach the insulation boundary. The change in baseline self-capacitance under each "depth-size" combination is recorded. and reference mutual capacitance change .

[0043] Furthermore, the size optimization and inversion steps in step 3 include:

[0044] Within the nearest radial depth data set, the dimensions are set. =Start comparing items one by one starting from 1mm;

[0045] Retrieve the reference self-capacitance and mutual capacitance changes at the current dimensions;

[0046] Calculate the ratio of the change in the self-capacitance to that of the reference self-capacitance, and the self-capacitance scaling factor. Calculate the ratio of the change in the mutual capacitance to be verified to the reference mutual capacitance, and the mutual capacitance scaling factor. ;

[0047] judge and Are they equal? ​​If yes, then the current size... If the defect size is specified, otherwise increment the size data and repeat the above calculation until a match is found.

[0048] Based on the scaling factor of the successfully matched capacitors Calculate the dielectric constant The calculation formula is:

[0049] ,

[0050] in, The dielectric constant of the insulating layer under normal conditions. The reference defect dielectric constant is denoted as .

[0051] This invention discloses a quantitative inversion method for multidimensional features of cable defects based on local capacitance distortion mapping. By acquiring capacitance data under normal and fault conditions, it directly utilizes the asymmetry and ratio relationship of local capacitance changes to decouple and independently quantify the core multidimensional features of the defect, including angular location, radial depth, size, and dielectric constant. Compared with existing cable insulation defect detection methods and traditional ECT image reconstruction technology, this method has the following technical advantages and beneficial effects:

[0052] 1. Decoupling and independent inversion of multidimensional defect features were achieved.

[0053] Existing image inversion methods couple the features of defects, such as angle, depth, size, and material, to the pixel distribution. This invention innovatively proposes a step-by-step decoupling algorithm for multi-dimensional features. By constructing a normalized parameter for the local capacitance change, it effectively cancels the interference between variables and achieves the decoupling of each physical parameter.

[0054] 2. It broke through the bottleneck of qualitative observation and achieved quantitative assessment.

[0055] Existing technologies often only provide the approximate location of defects (blurred image) or alarm signals (waveform), lacking quantitative assessment of the specific attributes of the defects themselves. This invention establishes a nonlinear polynomial mapping model and a benchmark feature database, which can directly and independently calculate and output the specific angular position, radial depth, geometric size, and dielectric constant of defects, thus achieving quantitative assessment of defect characteristics.

[0056] 3. It avoids pathological inversion problems and prevents image artifact interference.

[0057] Traditional ECT detection is limited by the soft field effect of electric field. When solving the image inverse problem, it is ill-conditioned and prone to artifacts, which can lead to the masking or misjudgment of tiny defects. This invention avoids the image inversion approach and directly establishes a mapping relationship from the original capacitance change. It realizes the inversion of fault features from the algorithm and improves the positioning accuracy and sensitivity of tiny latent defects inside high insulation materials.

[0058] 4. Low algorithm complexity

[0059] Traditional image inversion algorithms (such as Landweber iteration) require high-dimensional sensitivity matrix inversion and multiple iterative operations, which are demanding on computing hardware and time-consuming. This invention adopts a method of solving polynomial equations and directly comparing with the database. The calculation process only involves basic algebraic operations, which reduces the data processing time of a single detection. Attached Figure Description

[0060] Figure 1 Flowchart of the entire process for quantitative inversion of cable defect characteristics;

[0061] Figure 2 Schematic diagram of parameters for cable, array electrodes, and defects;

[0062] Figure 3 (a) Relationship between mutual capacitance and angle change (b) Characteristic parameters Relationship with angle change;

[0063] Figure 4 (a) Relationship between correction factor R1 and relative angle α (b) Radial characteristic parameters With radial depth relation;

[0064] Figure 5 (a) Angle calculation results under radial depth variation (b) Angle calculation results under dielectric constant variation (c) Angle calculation results under defect size variation;

[0065] Figure 6 (a) Calculation results of radial depth under varying dielectric constant; (b) Calculation results of radial depth under varying angular position; (c) Calculation results of radial depth under varying defect size.

[0066] Figure 7 (a) Size output results and (b) Dielectric constant output results for defects at different radial depths;

[0067] Figure 8 Output results of (a) size and (b) dielectric constant of defects at different angles and positions. Detailed Implementation

[0068] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0069] The purpose of this invention is to provide a method for quantitative inversion and mapping of multidimensional features of cable defects based on local capacitance distortion, specifically including the following steps:

[0070] Step 1, Defect angle and location inversion;

[0071] (1) Constructing a capacitance matrix of defect-free cables and defective cables

[0072] The detection device acquires capacitance signals through an array of eight electrodes. Under normal, defect-free operation, the system sequentially excites each electrode with a preset excitation voltage of 1V and receives the response signals. By recording the responses of all electrodes under different excitation conditions, an 8x8 full capacitance response matrix (64 data points) can be obtained. These 64 sets of data are then used to construct an 8x8 symmetrical capacitance matrix under normal operating conditions. :

[0073] .

[0074] When defects appear inside the cable insulation layer, the distribution of the insulation medium changes. The capacitance matrix under the defect state is collected using the method described above. To make the defect characteristics more obvious, a characteristic matrix of capacitance change is defined. Its elements are obtained by subtracting the absolute value of the defect matrix from the absolute value of the normal operation matrix:

[0075] ,

[0076] in, , These are the excitation electrode and the measurement electrode numbers, respectively. When a defect occurs inside the cable insulation layer, the response of the electrode plate within the defect's corresponding range will be reflected in the capacitance change characteristic matrix. The characteristic matrix of capacitance change can be observed within the diagram. Changes in specific values ​​can initially pinpoint the approximate location of the fault.

[0077] (2) Quantitative inversion of cable defect angle location

[0078] This step involves analyzing the distribution pattern of local capacitance changes to construct characteristic parameters, thereby locating the angular position of the defect on the cable insulation cross-section. The specific process is as follows:

[0079] 2.1 Determine the range of defect angle location to

[0080] Based on the geometric symmetry of the electrode circumferential distribution, every two adjacent electrodes form a 45-degree detection interval. When a defect occurs, the two adjacent electrodes closest to the defect exhibit the largest changes in self-capacitance and associated mutual capacitance. The set of adjacent electrodes is then used as the current electrode. (Corresponding to the starting angle of the interval) ) and the next electrode (Corresponding interval termination angle) This allows for a qualitative assessment that the defect is located within the 45-degree range.

[0081] 2.2 Constructing the characteristic parameters of the defect angle and position

[0082] like Figure 3 As shown in (a), within the detection period of 0-45 degrees, as the defect angle position increases, the change in mutual capacitance between the current electrode and the next electrode exhibits a nonlinear pattern of first increasing and then decreasing, while the change in mutual capacitance between adjacent intervals shows a monotonically increasing or decreasing trend. Based on this asymmetric response characteristic, spatial angle feature parameters that can effectively amplify the signal difference caused by the angle shift are constructed by utilizing the asymmetry of capacitance changes between adjacent intervals. .

[0083] definition The previous electrode of the current electrode. This is the electrode after the next electrode. Extract the following three sets of mutual capacitance changes:

[0084] Change in mutual capacitance between the current electrode and the next electrode ;

[0085] Change in mutual capacitance between the previous electrode and the current electrode ;

[0086] Change in mutual capacitance between the next electrode and the electrode after that .

[0087] Constructing normalized feature parameters Its mathematical expression is:

[0088] .

[0089] 2.3 Relative Angles Based on Cubic Polynomial Mapping Relationships Inversion

[0090] Establish the relative angular position of the defect The independent variable is the characteristic parameter obtained by calculation. For nonlinear mapping models of dependent variables, such as Figure 3 As shown in (b). The fitting analysis reveals that this mapping relationship follows a cubic multinomial distribution:

[0091] .

[0092] Calculated from the measured data at the time of the defect. Substituting the value into the above equation yields the unique valid relative angle value. .

[0093] 2.4 Inversion of absolute angle of cable defects

[0094] The final defect's angular position The relative angle is calculated by adding the reference angle of the starting electrode in the current interval. constitute:

[0095] .

[0096] The obtained relative angle Substituting into the above formula, the final angular position of the defect can be obtained. This allows for the inversion and localization of the defect's angle and position.

[0097] To verify the effectiveness of the inversion method for defect angular location features, the radial depth, size, and dielectric constant of the defect were changed, and the calculated characteristic parameters were... Values, and the angle values ​​obtained through analysis. Set the defect's angle position to a fixed 100 degrees, and then change the following three sets of feature parameters:

[0098] The radial depths are set to 34.44mm, 36.44mm, 38.44mm, 40.44mm, 42.44mm, 44.44mm, and 46.44mm.

[0099] The dielectric constants were set to 1, 1.3, 1.6, 1.9, 2.1, 2.5, 2.8, 3.1, 3.4, and 3.7, respectively.

[0100] Set the dimensions to 1mm, 2mm, 3mm, 4mm, 5mm, 6mm, 7mm, and 8mm respectively;

[0101] Substitute the measured data corresponding to the different defect conditions above into the following categories:

[0102] ,

[0103] ,

[0104] ,

[0105] The measured angle position results are as follows Figure 5 As shown in (a), (b), and (c), the calculated characteristic parameters The value remains relatively stable, and the obtained angle value Within a reasonable engineering tolerance range, this proves that the analytical method can achieve independent inversion of the defect angle location.

[0106] Step 2, Defect radial depth inversion;

[0107] This step constructs characteristic parameters using the ratio of the difference in capacitance change to the sum of the values, and introduces an angle correction factor to eliminate deviations caused by non-aligned electrode defects, thereby locating the radial depth of the defect on the cable insulation cross-section. The specific process is as follows:

[0108] (1) Constructing the initial radial characteristic parameters

[0109] When a defect occurs, the ratio of the changes in self-capacitance and mutual capacitance between the two nearest adjacent electrodes will change according to a specific pattern with varying radial depth. Based on the defect location interval obtained in the previous step, the set of adjacent electrodes located at the current electrode is defined as the current electrode. (Corresponding to the starting angle of the interval) ) and the next electrode Extract the following two sets of capacitance changes:

[0110] Change in mutual capacitance between the current electrode and the next electrode ;

[0111] Current change in self-capacitance of the electrode .

[0112] Construct normalized feature parameters that can characterize the radial depth from the fault point to the center of the cable conductor. :

[0113] .

[0114] (2) Constructing the angular position offset correction factor

[0115] As the defect angle shifts, the change in self-capacitance of the current electrode gradually decreases, while the change in mutual capacitance first increases and then decreases, leading to changes in characteristic parameters. Changes have occurred. Therefore, a correction factor is introduced. It is defined as the ratio of the eigenvalue at a specific angle to the eigenvalue when facing the current electrode:

[0116] ,in, , and These represent the fault at the actual angle, directly opposite the current electrode, and at a relative angle. The corresponding radial characteristic parameters;

[0117] Establish the relative angular position of the defect As independent variable, with correction factor For nonlinear mapping models of dependent variables, such as Figure 3 As shown in (a). The fitting analysis reveals that this mapping relationship follows a cubic multinomial distribution:

[0118] .

[0119] Known Substituting into the above formula, we can obtain the corresponding correction factor. At this point, the modified radial characteristic parameters are defined. :

[0120] .

[0121] (3) Radial depth based on second-order polynomial mapping relationship Inversion

[0122] After eliminating the interference caused by angular offset, the corrected characteristic parameters are established. With radial depth The baseline mapping model between them, such as Figure 3 As shown in (b). The results were obtained through simulation fitting. With radial depth The second-order characteristic equation:

[0123] .

[0124] The calculated Substituting into the equation, we obtain the unique and effective radial depth of the defect. .

[0125] To verify the effectiveness of the radial depth feature inversion method for defects, the angular position, size, and dielectric constant of the defects were changed, and the resulting correction factor was calculated. Radial characteristic parameters And the radial depth obtained from the analysis Set the radial depth of the defect to a fixed value of 40.44 mm or 52.44 mm, and then change the following three sets of characteristic parameters:

[0126] The dielectric constants were set to 1, 1.3, 1.6, 1.9, 2.1, 2.5, 2.8, 3.1, 3.4, and 3.7, respectively.

[0127] Set the angle positions to 0, 5, 10, 15, 20, 25, 30, 35, 40, and 45 respectively;

[0128] Set the dimensions to 1mm, 2mm, 3mm, 4mm, 5mm, 6mm, 7mm, and 8mm respectively;

[0129] Substitute the measured data corresponding to the different defect conditions above into the following categories:

[0130] ,

[0131] ,

[0132] ,

[0133] ,

[0134] The measured radial depth results are as follows Figure 6 As shown in (a), (b), and (c), the calculated correction factors Value, radial characteristic parameter The value remains relatively stable, and the obtained radial depth Within a reasonable engineering tolerance range, this demonstrates that the analytical method can achieve independent inversion of the radial depth of defects.

[0135] Step 3: Defect size and dielectric constant inversion;

[0136] (1) Simulate the construction of different defect sizes The benchmark feature library

[0137] A database of defect capacitance characteristics under standard conditions was pre-established. The dielectric constant of the baseline defect was set to 1, and standard angular positions of 0° and 22.5° were selected, along with five radial depth nodes of 40.44mm, 44.44mm, 46.44mm, 48.44mm, and 52.44mm. At each depth node, a series of defects with continuously increasing sizes (1mm, 2mm, 3mm... until reaching the insulation boundary) were simulated, and the change in baseline self-capacitance under each depth-size combination was recorded. and reference mutual capacitance change .

[0138] (2) Based on the relative angle and radial depth inversion feature set and the measured capacitance, match and optimize the data set and construct the capacitance to be verified. and

[0139] In the benchmark feature library, based on the radial depth obtained in step 2 Among the five sets of data corresponding to radial depth nodes 40.44mm, 44.44mm, 46.44mm, 48.44mm, and 52.44mm, the data set that is closest to the obtained radial depth of the defect is selected as the optimal data set.

[0140] In the optimization data set, based on the relationship between the self-capacitance and mutual capacitance corresponding to a relative angle of 22.5° and a standard angle of 0° within the set, the measured change in self-capacitance is... and mutual capacitance change The equivalent value converted to the standard angle is denoted as the measured self-capacitance to be verified. and measured mutual capacitance .

[0141] (3) Based on the construction of capacitance scaling factor Size optimization and inversion

[0142] Within the data set at the corresponding depth, compare each item sequentially, starting from a dimension of 1 mm:

[0143] Define the self-capacitance scaling factor: ,

[0144] Define the mutual capacitance scaling factor: .

[0145] judge and Are they equal? ​​If not, try increasing the size data and repeat the above calculation; if they match, the match is successful. The corresponding size data at this point... This refers to the physical size of the defect.

[0146] (4) Quantitative inversion of the true dielectric constant

[0147] After the matching is successful in step (3) above, the proportional scaling factor is obtained. This characterizes the difference in dielectric constant between the actual defect and the reference defect. This allows for the direct calculation of the true dielectric constant of the defect, thus completing the inversion of all multidimensional characteristics. The calculation formula is as follows:

[0148] ,

[0149] in, The dielectric constant of the insulating layer under normal conditions. The reference defect dielectric constant is denoted as .

[0150] To verify the effectiveness of the method for inverting defect size and dielectric constant characteristics, the angular position and radial depth of the defect were changed. Data sets corresponding to the actual radial depth were matched, and the actual angle was equivalently mapped to the corresponding value under the standard angle based on the relationship between the actual angle and the standard angle in the database, thus obtaining the self-capacitance. and mutual capacitance Set the defect size to a fixed 5mm and the dielectric constant to 3, and then change the following two sets of characteristic parameters:

[0151] The radial depths are set to 40.44mm, 44.44mm, 46.44mm, 48.44mm, and 52.44mm.

[0152] Set the angle positions to 0, 5, 10, 15, and 20 respectively;

[0153] Starting with the corresponding 1mm dimension data from the database, substitute the data:

[0154] ,

[0155] ,

[0156] Item-by-item comparison and Are they consistent? If they are inconsistent, try increasing the size data and repeat the above calculation; if they are consistent, then the corresponding size data is the physical size of the defect. This characterizes the difference in dielectric constant between the actual defect and the reference defect. The test results are as follows: Figure 7 (a)(b) and Figure 8 As shown in (a)(b), the final output size and dielectric constant All values ​​are within a reasonable engineering tolerance range. This proves that the analysis method can achieve independent inversion of the defect angle position.

[0157] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. These changes involve related technologies well known to those skilled in the art, and all of them fall within the protection scope of the present invention.

[0158] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.

Claims

1. A quantitative inversion method for multidimensional features of cable defects based on local capacitance distortion mapping, characterized in that, The method includes the following steps: Step 1, Defect Angle Position Inversion; Input the measured change in mutual capacitance of the defect, qualitatively determine the angle position interval, and construct the defect angle position feature parameters. Substitute the angle and position feature parameters Determine the relative angle of the defect Inversion defect angle position ;in ,definition The previous electrode of the current electrode. The next electrode after the next electrode. This represents the change in mutual capacitance between the current electrode and the next electrode. This represents the change in mutual capacitance between the previous electrode and the current electrode. This represents the change in mutual capacitance between the next electrode and the electrode after that. Step 2, Defect radial depth inversion; Input the measured changes in defect self-capacitance and mutual capacitance, and the defect angle and position. Construct initial radial characteristic parameters Substitute the initial radial characteristic parameters Determine the angle correction factor The corrected radial depth parameters were then obtained. Inversion of defect radial depth ;in , This represents the change in self-capacitance of the current electrode; Step 3: Defect size and dielectric constant inversion; input the measured changes in defect self-capacitance and mutual capacitance, and the defect angular position. radial depth of defects A baseline feature library with different defect sizes is constructed through simulation. Based on its input, the closest radial depth data set is matched. Within the closest radial depth data set, the capacitance value change is equivalently converted to the measured self-capacitance change and measured mutual capacitance change at a standard angle, and a capacitance scaling factor is constructed. Based on capacitance scaling factor Defect size is obtained by performing size optimization and inversion. According to the capacitance scaling factor Determine the dielectric constant ; The size optimization and inversion steps include: Within the nearest radial depth data set, the dimensions are set. =Start comparing items one by one starting from 1mm; Retrieve the reference self-capacitance and mutual capacitance changes at the current dimensions; Calculate the ratio of the change in the self-capacitance to that of the reference self-capacitance, and the self-capacitance scaling factor. Calculate the ratio of the change in the mutual capacitance to be verified to the reference mutual capacitance, and the mutual capacitance scaling factor. ; judge and Are they equal? ​​If yes, then the current size... If the defect size is specified, otherwise increment the size data and repeat the above calculation until a match is found. Based on the scaling factor of the successfully matched capacitors Calculate the dielectric constant The calculation formula is: ， in, The dielectric constant of the insulating layer under normal conditions. The reference defect dielectric constant is denoted as .

2. The method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping according to claim 1, characterized in that, The measured change in mutual capacitance of defects in step 1 is acquired by constructing a capacitance matrix of defect-free and defective cables. The acquisition steps include: the detection device acquires the cable capacitance signal through an array of electrodes. Under normal operating conditions without defects, the system sequentially excites each electrode according to a preset amplitude as the excitation voltage and receives the response signal of each electrode. The response of all electrodes under different excitation conditions is recorded, and a full capacitance response matrix can be obtained. This matrix data is used to construct a symmetrical capacitance matrix under normal operating conditions. : ， When a defect appears inside the cable insulation layer, the distribution of the insulation medium changes. The system sequentially excites the electrodes according to a preset amplitude as the excitation voltage and receives the response signals of each electrode. It records the response of all electrodes under different excitation conditions and collects the capacitance matrix under defect conditions. Define the measured change in mutual capacitance of defects. : ,in, , The excitation and measurement electrodes are numbered respectively. When a defect appears inside the cable insulation layer, the response of the electrode plate corresponding to the defect will be measured in terms of the change in mutual capacitance of the defect. Within the feature matrix.

3. The method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping according to claim 1, characterized in that, In step 1, the qualitative judgment of the angular position range is based on the geometric symmetry of the electrode circumference distribution. Every two adjacent electrodes form a 45-degree detection range. When a defect occurs, the self-capacitance and mutual capacitance changes of the two adjacent electrodes closest to the defect are the largest. The set of adjacent electrodes that have been positioned are the current electrodes. The starting angle of the corresponding interval With the next electrode The corresponding interval termination angle This allows for a qualitative assessment that the defect is located within the 45-degree range.

4. The method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping according to claim 3, characterized in that, The relative angle of the defect in step 1 Based on the inversion of the cubic polynomial mapping relation, the mapping relation satisfies: Substitute the angle and position feature parameters Determine the relative angle of the defect ; 。 5. The method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping according to claim 1, characterized in that, The angle correction factor Its mathematical expression is: ， in, , and These represent the fault at the actual angle, directly opposite the current electrode, and at a relative angle. The corresponding radial characteristic parameters; Angle correction factor Based on the inversion of the cubic polynomial mapping relation, the mapping relation satisfies: ， 。 6. The method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping according to claim 1, characterized in that, In step 2, the radial depth of the inversion defect is determined. The corrected radial depth parameters were obtained through simulation fitting. With the radial depth of the defect The second-order characteristic equation: 。 7. The method for quantitative inversion of multidimensional features of cable defects based on local capacitance distortion mapping according to claim 1, characterized in that, In step 3, a baseline feature library with different defect sizes is constructed through simulation. A defect capacitance feature database under standard conditions is pre-established, the dielectric constant of the baseline defect is set, a standard angular position is selected, and multiple sets of radial depth nodes are used. At each depth node, a series of defects with continuously increasing sizes are simulated until they reach the insulation boundary. The change in baseline self-capacitance under each "depth-size" combination is recorded. and reference mutual capacitance change .

Images