Oil paper sleeve defect combined diagnosis method and system based on oil pressure and UHF (Ultra High Frequency), medium and processor

Abstract

The invention provides an oil-paper sleeve defect combined diagnosis method based on oil pressure and UHF, and the method comprises the following steps: arranging a pressure sensor at an oil taking port of a flange of an oil-paper sleeve, and arranging a UHF sensor at a position, close to the flange, outside the oil-paper sleeve; acquiring pressure data in the casing pipe in real time through a pressure sensor, and acquiring an ultrahigh frequency signal generated by partial discharge in real time through a UHF (Ultra High Frequency) sensor; s3, performing combined fault type diagnosis according to the pressure data and the ultrahigh frequency signal data; performing discharge state evaluation on the bushing in the partial discharge state based on a fuzzy hierarchy method; and based on Euclidean distance division, state evaluation is carried out on the casing pipe with overheating or missing oil leakage. Through cooperative work of the pressure sensor and the UHF sensor, various fault types possibly occurring in the sleeve can be comprehensively covered, and compared with a traditional single detection method, the accuracy and comprehensiveness of fault diagnosis are greatly improved, and the situations of misjudgment and missed judgment are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of oil-paper bushing defect diagnosis, and particularly to a combined diagnosis method, system, medium and processor for oil-paper bushing defects based on oil pressure and UHF. Background Art

[0002] In the power system, the oil-paper insulated capacitive bushing is a key device, and the reliability of its operating state is crucial for the safe and stable operation of the entire system. However, during long-term operation, the bushing is easily affected by various factors such as electricity, heat, and mechanical stress, resulting in various defects, such as partial discharge, overheating, oil leakage, etc. Once a bushing fails, it may cause a power outage accident, resulting in huge economic losses and social impacts.

[0003] Traditional bushing defect detection methods have many limitations. For example, a single detection method is difficult to comprehensively and accurately judge the type and severity of the fault. Relying solely on pressure detection, it is impossible to effectively distinguish whether the fault is caused by partial discharge or other reasons; simply relying on UHF detection, the detection effect of some early weak discharge signals or signals in an interfered environment is not good. Moreover, the existing detection technologies are not fine enough in fault state assessment, cannot timely and accurately evaluate the fault development trend, and are difficult to meet the high requirements of the power system for equipment condition monitoring. Therefore, developing an efficient and accurate oil-paper bushing defect diagnosis method has become an urgent problem to be solved in the power industry.

[0004] In view of this, a combined diagnosis method, system, medium and processor for oil-paper bushing defects based on oil pressure and UHF are needed. Summary of the Invention

[0005] Aiming at the problems in the prior art that the assessment is not fine enough and a single detection method is difficult to comprehensively and accurately judge the type and severity of the fault, the present invention provides a combined diagnosis method, system, medium and processor for oil-paper bushing defects based on oil pressure and UHF, which can comprehensively and accurately judge the type and severity of the fault. The specific technical solutions are as follows:

[0006] A combined diagnosis method for oil-paper bushing defects based on oil pressure and UHF includes the following steps:

[0007] S1: Set a pressure sensor at the oil sampling port of the oil-paper bushing flange, and set a UHF sensor outside the oil-paper bushing near the flange;

[0008] S2: Real-time collect the internal pressure data of the bushing through the pressure sensor, and real-time collect the ultra-high frequency signals generated by partial discharge through the UHF sensor;

[0009] S3: Perform combined fault type diagnosis according to the pressure data and the ultra-high frequency signal data;

[0010] S4: Evaluate the discharge state of the bushing with partial discharge based on the fuzzy analytic hierarchy process;

[0011] S5: Evaluate the state of the overheated or oil-leaking bushing based on the Euclidean distance partition.

[0012] Further, in step S4, the evaluation of the discharge state of the bushing with partial discharge based on the fuzzy analytic hierarchy process includes the following steps:

[0013] S41: Select the current change rate of UHF, the sudden increase rate of UHF, the maximum value of ΔP, and the sudden increase rate of ΔP as evaluation parameters, and construct an evaluation parameter set U;

[0014] S42: Use the analytic hierarchy process to determine the weight coefficients of each parameter in the evaluation parameter set U;

[0015] S43: Determine the membership degrees of each parameter in the evaluation parameter set U to different evaluation levels through the membership function to form an evaluation matrix;

[0016] S44: Calculate the final evaluation result using the parameter weights and the evaluation matrix, and evaluate the discharge state of the bushing according to the evaluation result.

[0017] Further, in step S42, the use of the analytic hierarchy process to determine the weight coefficients of each parameter in the evaluation parameter set U includes the following steps:

[0018] S421: Use the scaling method to compare the relative importance of the parameters in the evaluation parameter set U pairwise and form a judgment matrix;

[0019] S422: Calculate the eigenvector of the judgment matrix by the eigenvalue method and perform normalization processing on the eigenvector to obtain the weight vector of the evaluation parameter set. The weight vector contains the weight coefficients of each parameter in the evaluation parameter set;

[0020] S423: Conduct a consistency test to determine whether the judgment matrix is acceptable. If not, return to step S421 to readjust the judgment matrix.

[0021] Further, in step S422, the calculation of the eigenvector of the judgment matrix by the eigenvalue method and the normalization processing of the eigenvector to obtain the weight vector of the evaluation parameter set, where the weight vector contains the weight coefficients of each parameter in the evaluation parameter set, includes the following steps:

[0022] Calculate the maximum eigenvalue λ of the judgment matrix A max ;

[0023] Substitute λ max into the following formula to find the non-zero solution of this homogeneous linear equation system, and this non-zero solution is the eigenvector W0 :

[0024] (A - λ max )W 0 = 0;

[0025] Divide each component of the eigenvector W 0 by the norm ∥W 0 ∥ of the vector W, and the formula is as follows: 0 ∥, and the formula is as follows:

[0026]

[0027] Obtain the normalized weight vector W, and the weight vector W contains the weight coefficients of each parameter in the evaluation parameter set. The formula is as follows:

[0028] W = (w 1 , w 2 , w 3 , w 4 );

[0029] In the above formula, w 1 , w 2 , w 3 , w 4 correspond to the weight coefficients of the current change rate of UHF, the sudden increase rate of UHF, the maximum value of ΔP, and the sudden increase rate of ΔP, respectively.

[0030] Furthermore, in step S43, the method for determining the membership degrees of each parameter in the evaluation parameter set to different evaluation levels through the membership function to form an evaluation matrix includes the following steps:

[0031] S431: Set different thresholds for different evaluation parameters in the evaluation parameter set;

[0032] S432: Construct a triangular membership distribution function calculation table for different evaluation parameters in the evaluation parameter set according to the thresholds;

[0033] S433: Form an evaluation matrix according to the triangular membership distribution function calculation table. The formula is as follows:

[0034]

[0035] In the above formula, R is the evaluation matrix; r ij represents the membership degree of the i-th parameter to the j-th evaluation state.

[0036] Furthermore, in step S44, the method for calculating the final evaluation result by using the parameter weights and the evaluation matrix and evaluating the discharge state of the casing includes the following steps:

[0037] The calculation formula of the evaluation result is as follows:

[0038] B 11 = w 1 × r 11 + w 2 × r 21 + w 3 × r 31 + w 4 × r 41 ;

[0039] B 12 = w 1 × r 12 + w 2 × r 22 + w 3 × r 32 + w 4 × r 42 ;

[0040] B = (B 11 B 12 );

[0041] If B 11 > B 12 , then it is judged that the partial discharge state is in the "attention" state; if B 11 < B 12 , then it is judged that the partial discharge state is in the "warning" state;

[0042] In the above formula, B 11 represents the membership degree of the equipment in the "attention" state after comprehensively considering various parameters; B 12 represents the membership degree of the equipment in the "warning" state.

[0043] Furthermore, in step S5, the state assessment of the overheated or oil-leaking sleeve based on the Euclidean distance division includes the following steps:

[0044] S51: Determine the thresholds of various influencing factors in the state assessment model of sleeve overheating and oil leakage;

[0045] S52: Extract the characteristic parameters of the pressure curve to construct a feature vector, calculate its Euclidean distance from the state vectors corresponding to different defects, and the defect state corresponding to the state vector with a closer distance is the defect state of the curve.

[0046] A combined diagnosis system for oil-paper sleeve defects based on oil pressure and UHF, applied to the combined diagnosis method for oil-paper sleeve defects based on oil pressure and UHF as described above, includes:

[0047] Setting module, which is used to set the pressure sensor at the oil extraction port of the oil-paper sleeve flange and set the UHF sensor outside the oil-paper sleeve near the flange;

[0048] An acquisition module is used to acquire the internal pressure data of the casing in real time through a pressure sensor, and to acquire the ultra-high frequency signal generated by partial discharge in real time through a UHF sensor;

[0049] A diagnosis module, which is used to perform joint fault type diagnosis based on pressure data and UHF signal data;

[0050] A first evaluation module, which is used to evaluate the discharge state of the bushing in the partial discharge state based on the fuzzy hierarchy process;

[0051] The second evaluation module is used to evaluate the state of the overheated or leaking oil casing based on Euclidean distance division.

[0052] A computer-readable storage medium includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned oil-paper casing defect joint diagnosis method based on oil pressure and UHF.

[0053] A processor is used to run a program, wherein the program executes the above-mentioned oil-paper casing defect joint diagnosis method based on oil pressure and UHF when running.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. Comprehensive and accurate diagnosis of fault types: By making the pressure sensor and UHF sensor work together, real-time pressure data and UHF signal data are collected, and joint fault type diagnosis is performed based on different value combinations of the two. This method can comprehensively cover various types of faults that may occur in the casing, such as internal discharge of the casing, oil leakage, lead wire discharge, local overheating, and normal operation. Compared with the traditional single detection method, it greatly improves the accuracy and comprehensiveness of fault diagnosis and reduces the cases of misjudgment and missed judgment.

[0056] 2. Fine evaluation of fault status: Based on the fuzzy hierarchy method, the discharge status of the bushing with partial discharge is evaluated, and multiple parameters such as UHF current change rate, UHF sudden increase rate, ΔP maximum value and ΔP sudden increase rate are selected to construct an evaluation parameter set. The weight coefficient of each parameter is determined by the hierarchical analysis method, and then the evaluation matrix is ​​formed through the membership function, and finally the evaluation level of the partial discharge status is calculated. This process can finely evaluate the severity and development trend of partial discharge, divide the discharge status into two levels of "attention" and "warning", and provide more targeted decision-making basis for operation and maintenance personnel.

[0057] 3. Quantitatively evaluate specific faults: For the bushing with overheating or oil leakage, using the method based on Euclidean distance division, by determining relevant thresholds, extracting the characteristic parameters of the pressure curve to construct a feature vector, and calculating the Euclidean distance between it and the different state vectors of the corresponding defects, so as to judge whether the fault is in the "attention" or "warning" state. This quantitative evaluation method helps to conduct a more accurate state assessment of overheating and oil leakage faults, timely discover potential serious problems, take corresponding measures in advance, and ensure the safe and stable operation of the equipment.

[0058] 4. Improve the detection efficiency and reliability: The entire technical solution integrates a variety of detection means and evaluation methods, realizing multi-dimensional detection and analysis of the defects of oil-paper bushings. From fault type diagnosis to fault state evaluation, a complete detection system is formed, which can obtain comprehensive and accurate equipment state information in a short time, improving the detection efficiency. At the same time, data sharing and mutual verification among various parts further enhance the reliability of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally marked with similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to actual scale.

[0060] Figure 1 It is a schematic flow chart of a joint diagnosis method for oil-paper bushing defects based on oil pressure and UHF;

[0061] Figure 2 It is a schematic diagram of the experimental circuit structure;

[0062] Figure 3 It is a schematic diagram of the system structure of a joint diagnosis method for oil-paper bushing defects based on oil pressure and UHF. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0064] It should be understood that when used in this specification and the appended claims, the terms "comprises" and "comprising" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0065] It should also be understood that the terms used in the specification of the present invention are merely for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0066] It should be further understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0067] Embodiment 1

[0068] The solution of this application is verified through an experimental platform. The experimental circuit is as Figure 2 shown. Among them, the test power supply consists of a voltage regulator T1 and a power frequency non - partial discharge test transformer T2, with a rated capacity of 55 kVA and a turns ratio of 100 kV / 400 kV; the protection resistor R is 10 kΩ and the rated current is 100 A, which plays a current - limiting role when the test sample breaks down; the capacitance value of the coupling capacitor C is 500 pF; Z is the detection impedance. The test bushing is fixed on the riser tank through a flange to simulate the actual operating environment of the bushing. The monitoring system mainly includes a pulse current detection, UHF detection, and pressure detection system. Among them, the pulse current detection is mainly used to provide assistance for the data analysis of UHF and pressure monitoring. The UHF (abbreviation for Ultra High Frequency, meaning ultra - high frequency) antenna (hereinafter referred to as: UHF sensor) is placed near the flange, and the pressure sensor is installed at the oil sampling port of the bushing flange.

[0069] As Figure 1 shown, a combined diagnosis method for oil - paper bushing defects based on oil pressure and UHF, which is applied to the defect diagnosis of oil - paper insulated capacitive bushings, includes the following steps:

[0070] S1: Set the pressure sensor at the oil sampling port of the oil - paper bushing flange, and set the UHF sensor outside the oil - paper bushing near the flange.

[0071] The pressure sensor is set at the oil sampling port of the bottom flange of the oil - paper bushing body because when a fault occurs inside the bushing, such as partial discharge or overheating resulting in gas generation, the bottom can feel the pressure change first and most directly, ensuring that the obtained pressure data accurately reflects the actual internal situation. At the same time, installing at the bottom can also avoid interference from other components or uneven gas distribution due to too high installation position, which affects the measurement accuracy.

[0072] The UHF sensor is arranged outside the oil-paper bushing near the flange. The flange is an area where the electric field is relatively concentrated. When partial discharge occurs inside the bushing, the generated UHF signal can be more effectively transmitted to the outside through the area near the flange. Installing the UHF sensor here can receive the UHF signal generated by partial discharge to the greatest extent, reduce signal transmission loss and external interference, and improve the monitoring sensitivity and accuracy.

[0073] S2: Real-time collect the internal pressure data of the bushing through the pressure sensor, and real-time collect the UHF signal generated by partial discharge through the UHF sensor.

[0074] S21: Before data collection, the pressure sensor and the UHF sensor need to be calibrated. For the pressure sensor, a high-precision pressure calibration device can be used to measure at different pressure values, compare the measurement results with the standard pressure values, record the deviations and make corrections. For the UHF sensor, a dedicated signal generator can be used to emit UHF signals with known intensity and frequency, test the accuracy of the received signals within the receiving range of the sensor, and adjust the sensor parameters to ensure the measurement accuracy. The calibration period should also be specified, such as calibrating regularly before the start of the experiment or at regular intervals, to ensure the reliability of the sensor throughout the monitoring process.

[0075] S22: Perform filtering and denoising preprocessing on the collected data. The collected pressure data and UHF signal data may contain noise and interference signals. For the pressure data, due to factors such as mechanical vibration and electromagnetic interference in the on-site environment, data fluctuations may occur. Methods such as moving average filtering and Kalman filtering can be used to remove the noise, making the pressure change curve smoother for subsequent analysis. For the UHF signal, due to the complex electromagnetic environment on site, there may be a large number of interference signals. Signal processing technologies such as wavelet transform can be used to filter according to the characteristic frequency band of the UHF signal, extract the effective signal, improve the signal-to-noise ratio of the signal, and avoid misjudgment caused by noise.

[0076] S23: Store and back up the preprocessed data. The real-time collected data needs to be stored properly for subsequent analysis. The data storage format should be specified, such as using formats like CSV, HDF5, etc., which are convenient for data reading and processing. The choice of storage device is also crucial. A large-capacity hard disk array or cloud storage service can be used to ensure the security and scalability of the data. At the same time, a data backup strategy should be formulated to regularly back up the collected data to prevent data loss caused by storage device failures or other unexpected situations. The backed-up data should be stored in different geographical locations to improve the data disaster recovery ability.

[0077] Furthermore, the pressure sensor continuously collects the internal pressure data of the casing to obtain the pressure change value ΔP. Record the data at a set time interval (such as every second or every few minutes) for subsequent analysis of the pressure change trend. ΔP is the pressure change value inside the casing. When ΔP > 0.1 kPa, it indicates that there is a fault inside the equipment, and for the convenience of application, it is recorded as 1; when -0.1 kPa < ΔP < 0.1 kPa, it indicates that there is no fault inside the equipment, and at this time it is recorded as 0; when ΔP < -0.1 kPa, it indicates that there is a problem with the equipment's sealing, such as oil leakage, and it is recorded as -1.

[0078] Furthermore, the UHF sensor continuously monitors whether there is a UHF signal generated by partial discharge. Record the signal state as 1 when there is a signal and as 0 when there is no signal. Also record the signal state information at a certain time interval. There are two states in UHF monitoring: having a signal and having no signal, corresponding to the existence and non-existence of partial discharge inside the oil-minus equipment respectively, and these two states are recorded as 1 and 0 respectively.

[0079] S3: Conduct a combined fault type diagnosis based on the pressure data and the UHF signal data.

[0080] S31: According to the different value combinations of the collected ΔP and UHF, determine the fault type by referring to the fault type identification coding table.

[0081] Furthermore, when UHF = 1 and ΔP = 1, it is diagnosed as internal discharge of the casing; when UHF = 0 and ΔP = -1, it is diagnosed as problems such as oil leakage.

[0082] Furthermore, the specific judgment logic of the fault type identification coding table is as follows:

[0083]

[0084] When UHF = 1 and ΔP = 1, the UHF signal indicates that there is a discharge fault inside the equipment, and the increase in pressure further verifies that the discharge causes the pressure change, most likely local discharge inside the casing.

[0085] If UHF = 1 and ΔP = -1, the UHF signal indicates the existence of partial discharge, but the pressure decreases, which may be a sealing fault of the equipment, such as air leakage or oil leakage.

[0086] When UHF = 1 and ΔP = 0, although theoretically the discharge should cause the pressure to rise, in the early stage of discharge, with small discharge amount or low frequency, the pressure change may not be obvious, and it can be diagnosed as a minor discharge fault. Specifically, in terms of the fault type, it is the lead wire discharge.

[0087] If UHF = 0 and ΔP = 1, the UHF signal rules out the discharge fault, and the increase in pressure indicates the existence of other faults, which can be diagnosed as an overheating fault.

[0088] When UHF = 0 and ΔP = -1, after excluding temperature interference, if the pressure drops, it can be diagnosed that there is a fault in the equipment's sealing performance and there is an oil leakage situation.

[0089] If UHF = 0 and ΔP = 0, there is no UHF signal and the pressure remains unchanged, indicating that there are no discharge, overheating, and sealing problems in the equipment, and it is in a normal operating state.

[0090] S4: Based on the fuzzy hierarchy method, evaluate the discharge state of the bushing in the partial discharge state. Specifically, it includes the following steps:

[0091] S41: Select the current change rate of UHF, the sudden increase rate of UHF, the maximum value of ΔP, and the sudden increase rate of ΔP as evaluation parameters, and construct an evaluation parameter set U. These parameters reflect the characteristics of partial discharge and the pressure change situation from different angles, and can more comprehensively evaluate the partial discharge state.

[0092] Furthermore, the current change rate of UHF reflects the change of the UHF signal at the current moment relative to the previous moment. The determination steps of the current change rate of UHF are as follows:

[0093]

[0094] Assume that at two consecutive sampling times t n and t n-1 , the signal intensities collected by the UHF sensor are S n and S n-1 , and r UHF-current is the current change rate of UHF. The current change rate of UHF can intuitively show the real-time change trend of the UHF signal. When the current change rate of UHF is large, it means that the partial discharge activity may have a relatively obvious increase in the short term, which may indicate that the discharge fault is developing rapidly and requires key attention. For example, during the operation of the equipment, if it is found that the current change rate of UHF continues to increase, it may indicate that the discharge area inside the bushing is expanding or the discharge intensity is increasing continuously.

[0095] Furthermore, the sudden increase rate of UHF is used to measure the degree of rapid change of the UHF signal in a short time. The determination steps of the sudden increase rate of UHF are as follows:

[0096]

[0097] In the above formula, set a relatively short time window Δt; within this time window, S 1 is the UHF signal intensity at the starting time t 1 ; S 2 is the UHF signal intensity at the ending time t 2Signal strength. The UHF sudden increase rate can effectively capture sudden occurrences of partial discharge. When this parameter shows a large value, it indicates that the partial discharge signal has suddenly increased within a short period, which is often related to sudden faults inside the bushing, such as the instantaneous breakdown of the insulating material leading to a sharp rise in the discharge intensity. It is an important warning signal for possible serious faults in the equipment.

[0098] Furthermore, the maximum value of ΔP, ΔP max The acquisition method is to record all data points of the pressure change value ΔP during a period of time (such as one cycle of equipment operation or a specific monitoring time period) while the pressure sensor continuously collects the internal pressure data of the bushing, and then find the maximum value from them. This maximum value is the maximum value of ΔP. For example, during a one-day monitoring period, the pressure change value is recorded every 1 minute to obtain a series of ΔP data, and the maximum value is determined after comparison. The maximum value of ΔP reflects the maximum amplitude of the internal pressure change of the bushing caused by partial discharge and other reasons during the monitoring period. A larger maximum value of ΔP indicates that the equipment internal may have experienced a relatively intense physical process, such as severe partial discharge generating a large amount of gas, which in turn causes a significant increase in pressure. It is one of the important indicators for evaluating the severity of partial discharge.

[0099] Furthermore, similar to the UHF sudden increase rate, the ΔP sudden increase rate is also calculated for the sharp change of the pressure change value within a short time. Determine a short time window Δt, let the pressure change value at the start time of the window be ΔP 1 , and the end time be ΔP 2 , then the determination steps of the ΔP sudden increase rate are as follows:

[0100]

[0101] In the above formula, r ΔP-sudden is the ΔP sudden increase rate. The ΔP sudden increase rate can quickly reflect the sudden change of pressure. When this value is large, it means that the internal pressure of the bushing has fluctuated sharply within a short time, which may be related to the sudden intensification of partial discharge or other sudden faults (such as sudden damage of internal components leading to abnormal pressure changes), and it is of great significance for timely discovering potential serious problems of the equipment.

[0102] Combine the calculated current change rate of UHF, UHF sudden increase rate, ΔP max and ΔP sudden increase rate these four parameters together to form an evaluation parameter set U, that is:

[0103] U = {r UHF-current , r UHF-sudden , ΔP max , r ΔP-sudden};

[0104] This evaluation parameter set U comprehensively describes the characteristics of UHF signals and pressure changes during partial discharge from different perspectives, providing basic data for subsequent steps such as using the analytic hierarchy process to determine the weight coefficients of each parameter and constructing an evaluation matrix, thereby achieving an accurate assessment of the partial discharge state.

[0105] S42: Use the analytic hierarchy process to determine the weight coefficients of each parameter in the evaluation parameter set U, that is, construct a judgment matrix by comparing the relative importance of each parameter, calculate the weight vector, and perform a consistency test to ensure that the weight assignment is reasonable and reliable. The specific steps are as follows:

[0106] S421: Use the scaling method to compare the relative importance of the parameters in the evaluation parameter set U pairwise and form a judgment matrix.

[0107] Determine the comparison scale: Adopt the 1-9 scaling method to compare the relative importance of the four parameters of the current change rate of UHF, the sudden increase rate of UHF, ΔP max and the sudden increase rate of ΔP in the evaluation parameter set U pairwise. For example, if it is considered that the current change rate of UHF is equally important as the sudden increase rate of UHF, the scaling value is taken as 1; if the current change rate of UHF is slightly more important than the sudden increase rate of UHF, the scaling value is taken as 3; if much more important, the scaling values are taken as 5, 7, 9, etc., and the intermediate states take 2, 4, 6, 8.

[0108] Form judgment matrix A: Assume the four parameters are A, B, C, D respectively, and the form of the constructed judgment matrix A is as follows:

[0109]

[0110] In the above formula, where a ij represents the scaling value of the comparison between parameter i and parameter j, and In actual operation, the scaling values need to be accurately filled according to the judgment of the relative importance of each parameter. Parameter i or parameter j is one of the four parameters of the current change rate of UHF, the sudden increase rate of UHF, ΔP max and the sudden increase rate of ΔP above.

[0111] S422: Calculate the eigenvector of the judgment matrix A by the eigenvalue method and perform normalization processing on the eigenvector to obtain the weight vector of the evaluation parameter set U, and the weight vector contains the weight coefficients of each parameter in the evaluation parameter set.

[0112] Solving the eigenvector W is achieved by solving the homogeneous linear equation system (A - λ max I)W 0 = 0 (where I is the identity matrix). In specific calculations, first calculate the largest eigenvalue λ of the judgment matrix A max , then substitute λ max into (A - λ max )W0 In the equation = 0, find the non - zero solution of this homogeneous linear equation system, and this non - zero solution is the eigenvector W 0 . After that, in order to make the sum of the components of the eigenvector W 0 equal to 1, that is, to normalize it, divide each component of W 0 by the norm ∥W 0 ∥ of the vector W 0 to obtain the normalized weight vector W

[0113]

[0114] S4221: First, calculate the largest eigenvalue λ of the judgment matrix A max .

[0115] Calculation steps: First, select the initial vector V 0 = [1, 1, 1, 1] T . Calculate V 1 = AV 0 to obtain the vector V 1 . Then calculate (You can choose any i, such as i = 1). Then normalize V 1 to obtain where ∥V 1 ∥ is the norm of V 1 . Repeat the above steps to calculate and iterate continuously until |λ k+1 - λ k | is less than the pre - set precision (such as 10 -6 ). At this time, λ k + 1 is the largest eigenvalue λ of the judgment matrix A max .

[0116] S4222: Secondly, calculate the eigenvector W max corresponding to the largest eigenvalue λ 0 . Generally, the eigenvector W 0 can be obtained by calculating AW max = λ 0 W 0 .

[0117] (1). Calculate A - λ max I: Given the judgment matrix and the already obtained largest eigenvalue λ max , calculate A - λ max I, where I is a 4 - order identity matrix Then:

[0118]

[0119] (2). For \(A - \lambda\) max Perform elementary row operations on \(I\) to transform it into row echelon form: Using the elementary row operations of matrices, transform \(A - \lambda\) max \(I\) into a row echelon form matrix. For example, through operations such as row addition (adding a multiple of one row to another row), row swapping (swapping the positions of two rows), and row multiplication (multiplying a row by a non-zero constant). Suppose after a series of operations, the row echelon form matrix obtained is (The actual transformation result depends on the elements of matrix \(A\) and \(\lambda\) max value). Suppose the matrix \(A\) is judged as:

[0120]

[0121] The largest eigenvalue \(\lambda\) has been obtained max , suppose \(\lambda\) max = 4.01 (obtained through actual calculation, only an example here), then calculate \(A - \lambda\) max \(I\):

[0122]

[0123] First step: Multiply the first row by to make the leading term become 1.

[0124]

[0125] Second step: Subtract 0.5 times the first row from the second row, subtract times the first row from the third row, subtract

[0126]

[0127] Simplify to get:

[0128]

[0129] Third step: Multiply the second row by to make the leading term of the second row become 1;

[0130]

[0131] Fourth step: Add times the second row to the first row, subtract 0.73 times the second row from the third row, and subtract 0.39 times the second row from the fourth row;

[0132]

[0133] Fifth step: Multiply the third row by to make the leading term of the third row become 1;

[0134]

[0135] Step 6: Multiply the first row by 1.93 and add it to the third row, multiply the second row by 0.93 and add it to the third row, and subtract 1.11 times the third row from the fourth row;

[0136]

[0137] Step 7: Multiply the fourth row by to make the leading term of the fourth row equal to 1;

[0138]

[0139] Step 8: Subtract 8 times the fourth row from the first row, subtract 4.35 times the fourth row from the second row, and add 1.54 times the fourth row to the third row;

[0140]

[0141] After the above series of elementary row operations, A - λ max I is transformed into a row echelon form matrix.

[0142] (3). Solve the homogeneous linear equations: The homogeneous linear equations corresponding to the row echelon form matrix B are:

[0143]

[0144] Let x 4 = t (t is any non-zero real number, generally take t = 1 for convenience of calculation), and solve the equations from bottom to top. From x 3 + b 34 x 4 = 0, we can get x 3 = -b 34 t; Substitute x 3 = -b 34 t into x 2 + b 23 x 3 + b 24 x 4 = 0, and solve to get x 2 = (-b 24 + b 23 b 34 )t; Then substitute the values of x 2 and x 3 into x 1 + b 12 x 2 + b 13 x 3 + b 14 x 4 = 0, and solve to get x 1= [-b 14 + b 13 b 34 -b 12 (-b 24 + b 23 b 34 )]t. Thus, the general solution of the system of equations is obtained:

[0145]

[0146] This general solution is the eigenvector corresponding to the eigenvalue λ of the matrix A - λ max I, where t represents all possible values of the eigenvector. When t = 1, the obtained vector is a particular solution and can be used as an initial value of the eigenvector W. max

[0147] S4223: Finally, normalize the eigenvector W 0 so that the sum of its elements is 1. The obtained normalized vector is the weight vector W of each parameter.

[0148] Normalization to obtain the weight vector W: To obtain the weight vector W that meets the requirements, it is necessary to normalize the eigenvector obtained above. Let the obtained eigenvector be Calculate its modulus Then perform normalization, and the weight vector At this time, the obtained W is the normalized eigenvector corresponding to the largest eigenvalue λ of the judgment matrix A max , that is, the weight vector W formed by the weight coefficients of each parameter in the evaluation parameter set, satisfying

[0149] S423: Consistency check.

[0150] Calculate the consistency index CI: where n is the order of the judgment matrix (n = 4 in this embodiment). The closer λ max is to n, the closer the CI value is to 0, indicating that the consistency of the judgment matrix is better.

[0151] Find the random consistency index RI: According to the matrix order n, consult the standard value table of the random consistency index RI. For n = 4, RI usually takes 0.90.

[0152] Calculate the consistency ratio CR: When CR < 0.1, it is considered that the judgment matrix has acceptable consistency and the weight vector is reasonable and reliable; if CR ≥ 0.1, the judgment matrix needs to be adjusted again until the consistency requirement is met.

[0153] After the above calculations and checks, a set of reasonable weight coefficients w is finally obtained i, corresponding to the current UHF change rate, UHF sudden increase rate, maximum ΔP value, and ΔP sudden increase rate respectively. These weight coefficients indicate the relative importance of each parameter in evaluating the partial discharge state. For example, if the calculated weight of the current UHF change rate is 0.35, the weight of the UHF sudden increase rate is 0.25, the weight of the maximum ΔP value is 0.2, and the weight of the ΔP sudden increase rate is 0.2, this means that when evaluating the partial discharge state, the current UHF change rate has a relatively large impact on the evaluation result, and other parameters also play their roles according to the weight ratio. These weight coefficients will play a key role in the subsequent step S44 when calculating the final evaluation result B using the formula B = W·R, multiplying with the evaluation matrix R to determine the evaluation level of the partial discharge state.

[0154] S43: Determine the membership degrees of each parameter in the evaluation parameter set U to different evaluation levels through the membership function to form an evaluation matrix. Set the thresholds of each influencing factor based on a large amount of experimental data, construct a triangular membership distribution function calculation table, determine the membership degrees of each parameter to different evaluation levels (attention and warning), and form an evaluation matrix R.

[0155] S431: Set different thresholds for different evaluation parameters in the evaluation parameter set respectively: In the evaluation of the partial discharge state of oil-paper insulated capacitive bushings, for the four evaluation parameters of the current UHF change rate, UHF sudden increase rate, maximum ΔP value, and ΔP sudden increase rate, set different thresholds respectively. Taking the current UHF change rate as an example, according to the statistical analysis of a large amount of experimental data, when the current UHF change rate is less than 5%, it can be considered that the partial discharge activity is relatively stable; when it is between 5% - 15%, there is a certain degree of partial discharge development trend; when it is greater than 15%, the partial discharge activity may be relatively intense. Similarly, for the UHF sudden increase rate, set the thresholds to 10% and 30%; for the maximum ΔP value, set the thresholds to 0.5 kPa and 1.0 kPa according to the bushing type and operating environment; for the ΔP sudden increase rate, set the thresholds to 10% and 30%. These thresholds are not fixed and will be dynamically adjusted according to different bushing equipment characteristics and operating experience.

[0156] S432: Construct a triangular membership distribution function calculation table for different evaluation parameters in the evaluation parameter set according to the thresholds.

[0157] Calculation of the membership degree of the current UHF change rate: Construct a triangular membership distribution function. For the "Attention" state, when the current UHF change rate is less than or equal to 5%, the membership degree is 1; when it is between 5% and 10%, the membership degree linearly decreases from 1 to 0; when it is greater than or equal to 10%, the membership degree is 0. For the "Warning" state, when the current UHF change rate is less than or equal to 10%, the membership degree is 0; when it is between 10% and 15%, the membership degree linearly increases from 0 to 1; when it is greater than or equal to 15%, the membership degree is 1. For example, if the current UHF change rate is 8%, its membership degree for the "Attention" state is 0.4, and its membership degree for the "Warning" state is 0.6.

[0158] Calculation of the membership degree of the sudden increase rate of UHF: Similarly, for the "Attention" state, when the sudden increase rate of UHF is less than or equal to 10%, the membership degree is 1; between 10% and 20%, the membership degree linearly decreases from 1 to 0; when it is greater than or equal to 20%, the membership degree is 0. For the "Warning" state, when the sudden increase rate of UHF is less than or equal to 20%, the membership degree is 0; between 20% and 30%, the membership degree linearly increases from 0 to 1; when it is greater than or equal to 30%, the membership degree is 1. Suppose the sudden increase rate of UHF is 25%, its membership degree for the "Attention" state is 0.2, and its membership degree for the "Warning" state is 0.8.

[0159] Calculation of the membership degree of the maximum value of ΔP: Construct a function with 0.5 kPa and 1.0 kPa as thresholds. For the "Attention" state, when the maximum value of ΔP is less than or equal to 0.5 kPa, the membership degree is 1; between 0.5 kPa and 0.75 kPa, the membership degree linearly decreases from 1 to 0; when it is greater than or equal to 0.75 kPa, the membership degree is 0. For the "Warning" state, when the maximum value of ΔP is less than or equal to 0.75 kPa, the membership degree is 0; between 0.75 kPa and 1.0 kPa, the membership degree linearly increases from 0 to 1; when it is greater than or equal to 1.0 kPa, the membership degree is 1. If the maximum value of ΔP is 0.8 kPa, its membership degree for the "Attention" state is 0.2, and its membership degree for the "Warning" state is 0.8.

[0160] Calculation of the membership degree of the sudden increase rate of ΔP: When the sudden increase rate of ΔP is less than or equal to 10%, the membership degree of the "Attention" state is 1; between 10% and 20%, the membership degree linearly decreases from 1 to 0; when it is greater than or equal to 20%, the membership degree is 0. For the "Warning" state, when the sudden increase rate of ΔP is less than or equal to 20%, the membership degree is 0; between 20% and 30%, the membership degree linearly increases from 0 to 1; when it is greater than or equal to 30%, the membership degree is 1. If the sudden increase rate of ΔP is 22%, its membership degree for the "Attention" state is 0.4, and its membership degree for the "Warning" state is 0.6.

[0161] S433: Calculate the evaluation matrix R according to the triangular membership distribution function calculation table: Arrange the membership degrees of each parameter obtained above for the "Attention" and "Warning" states to form the evaluation matrix R. Suppose the membership degrees of the currently calculated UHF change rate, UHF sudden increase rate, maximum value of ΔP, and ΔP sudden increase rate for the "Attention" state are 0.3, 0.5, 0.6, and 0.4 respectively, and the membership degrees for the "Warning" state are 0.7, 0.5, 0.4, and 0.6 respectively. Then the evaluation matrix R is:

[0162]

[0163] Through the above detailed steps, the process of determining the evaluation matrix based on the membership function is completed, providing key data support for calculating the final evaluation result B using the parameter weight W and the evaluation matrix R later, which helps to more accurately judge the partial discharge state of the oil-paper insulated capacitive bushing.

[0164] S44: Use the parameter weight W and the evaluation matrix R to calculate the final evaluation result B according to the formula B = W·R, and evaluate the discharge state of the bushing based on the evaluation result, that is, obtain the evaluation grade of the partial discharge state by referring to the evaluation set and judge whether it is in the attention state or the warning state.

[0165] Suppose the weight vector W = (w 1 、w 2 、w 3 、w 4 ), where w 1 、w 2 、w 3 、w 4 correspond to the weight coefficients of the UHF current change rate, UHF sudden increase rate, maximum value of ΔP, and ΔP sudden increase rate respectively;

[0166] Evaluation matrix where r ij represents the membership degree of the i-th parameter for the j-th evaluation state (j = 1 is the "Attention" state, j = 2 is the "Warning" state). Then the final evaluation result B is a 1×2 matrix, and the calculation process is:

[0167] B 11 = w 1 ×r 11 + w 2 ×r 21 + w 3 ×r 31 + w 4 ×r 41 ;

[0168] B 12 = w 1 ×r 12 + w2 ×r 22 +w 3 ×r 32 +w 4 ×r 42 ;

[0169] The obtained B=(B 11 B 12 ), where B 11 represents the membership degree of the equipment in the "Attention" state after comprehensively considering various parameters; B 12 represents the membership degree of the equipment in the "Early Warning" state.

[0170] Judge the evaluation level: Make a judgment by referring to the evaluation set. Compare B 11 and B 12 . If B 11 >B 12 , it is judged that the partial discharge state is in the "Attention" state, which means that although there is a partial discharge situation, it is currently in a relatively stable stage, but the development trend needs to be continuously monitored. If B 11 <B 12 , it is judged that the partial discharge state is in the "Early Warning" state, indicating that the partial discharge activity is relatively intense and there may be a serious fault risk in the equipment, and measures need to be taken in a timely manner, such as arranging maintenance, replacing components, etc. If B 11 =B 12 , this situation is relatively rare. At this time, combined with the actual situation, such as the historical data of the recent equipment operation, other monitoring indicators, etc., further comprehensively judge whether the partial discharge state is more inclined to "Attention" or "Early Warning", or set a special processing process, such as increasing the monitoring frequency, etc.

[0171] The progressive relationship between step S3 and step S4 is as follows. The above step S3 for jointly diagnosing the fault type based on the pressure data and the UHF signal data is a preliminary diagnosis. It is based on the real-time monitored UHF (Ultra High Frequency) and the internal pressure signal of the bushing, and through simple logical judgment, classifies the equipment state into several types such as internal discharge of the bushing, oil leakage problem, discharge of the outgoing line, local overheating, normal operation, etc. And the evaluation and early warning of the partial discharge state of the bushing based on the fuzzy hierarchy method in this step is to further deeply analyze the partial discharge state on the basis of the possible partial discharge problem found in the previous preliminary diagnosis (such as the case where UHF = 1 in the first diagnosis result). It conducts a more detailed evaluation of the partial discharge state through steps such as establishing a comprehensive evaluation parameter set and determining the parameter weight coefficient, divides it into two levels of attention and early warning, clarifies the severity and development trend of the partial discharge, so it is the deepening and refinement of the diagnosis result of the previous step.

[0172] Data sharing and complementary relationship between step S3 and step S4: The UHF data (with signal or without signal) and the pressure change value ΔP obtained in the combined fault type diagnosis of pressure and UHF provide some basic data for the evaluation and early warning of the partial discharge state of the bushing based on the fuzzy hierarchy method. The parameters selected for the evaluation and early warning based on the fuzzy hierarchy method, such as the current change rate of UHF, the sudden increase rate of UHF, the maximum value of ΔP, and the sudden increase rate of ΔP, are further processing and expansion of the previous monitoring data. At the same time, the evaluation results based on the fuzzy hierarchy method can be fed back into the overall fault diagnosis system, corroborating with the combined fault type diagnosis results to improve the accuracy of diagnosis. For example, if the combined diagnosis determines internal discharge of the bushing, and the fuzzy hierarchy method evaluates that the partial discharge is in a warning state, this further strengthens the judgment that there are serious problems with the equipment.

[0173] S5: Conduct state evaluation on the overheated or oil-leaking or oil-deficient bushings based on Euclidean distance partitioning.

[0174] S51: Through the analysis of a large amount of experimental data, determine the thresholds of various influencing factors in the state evaluation models for bushing overheating and oil-leaking or oil-deficient conditions, such as the maximum change amount of overheating defects, the maximum change value / rise time threshold, the maximum change amount / fall time threshold of oil-leaking or oil-deficient defects, etc. These thresholds are the key basis for judging the fault state.

[0175] 1. Determination of overheating defect threshold.

[0176] Maximum change amount threshold: By analyzing a large amount of experimental data, statistically analyze the pressure change data of different types of bushings under normal operation and overheating fault conditions. For a specific type of bushing, if it is found in multiple experiments that when the maximum change amount of pressure within a certain time interval (such as one operating cycle) exceeds ΔP 过热-变-阈值 (For example, for a common type of bushing, ΔP 过热-变-阈值 = 3 kPa), the probability of the equipment having an overheating fault increases significantly. Then, ΔP 过热-变-阈值 is determined as the maximum change amount threshold of overheating defects.

[0177] Maximum change value / rise time threshold: Also based on experimental data, calculate the ratio of the maximum change value of pressure to the rise time during the overheating process of the bushing. After analyzing multiple groups of data, it is found that when this ratio exceeds R 过热-变时-阈值 (Assume that for a certain type of bushing, R 过热-变时-阈值 = 0.5 kPa / min), it can be judged that the equipment is in an overheating fault state. This threshold reflects the rate of pressure rise, and too fast a rate often means there are overheating problems inside the equipment.

[0178] 2. Determination of oil-leaking or oil-deficient defect threshold.

[0179] Maximum change amount / drop time threshold: For oil leakage and lack of oil defects, analyze the data of the casing pressure drop in the experiment. Determine that when the ratio of the maximum change amount of the pressure to the drop time exceeds R 漏油-变时-阈值 (For example, through a large number of experiments, for a certain series of casings, R 漏油-变时-阈值 = 0.05 kPa / min), it can be considered that there is an oil leakage or lack of oil fault in the equipment. This threshold can measure the speed of the pressure drop and help judge whether there is an oil leakage in the equipment.

[0180] S52: Extract the characteristic parameters of the pressure curve to construct a feature vector, and calculate its Euclidean distance from the different state vectors corresponding to the defects. The defect state corresponding to the state vector with a closer distance is the defect state of the curve, and judge whether it is in the attention state or the warning state. The specific steps are as follows:

[0181] S521: Extract the characteristic parameters of the pressure curve to construct a fault feature vector.

[0182] 1. Construction of overheating fault feature vector: For overheating faults, extract two key characteristic parameters from the pressure monitoring data:

[0183] One is the maximum change amount ΔP of the pressure within the monitoring time period max-过热 , by comparing all pressure change values, find the maximum value among them;

[0184] The other is to calculate the ratio R of the maximum change value to the rise time corresponding to the maximum change value of the pressure 上升-过热 , that is, within the time period t 上升-过热 from the start of the pressure rise to the maximum change value, Combine these two parameters into an overheating fault feature vector

[0185]

[0186] 2. Construction of oil leakage and lack of oil fault feature vector: For oil leakage and lack of oil faults, mainly focus on the pressure drop process. Extract the ratio of the maximum change amount of the pressure to the drop time as the characteristic parameter, denoted as R 下降-漏油 . Assume that during the monitoring process, from the discovery of the start of the pressure drop to the time when the pressure drop reaches the maximum, the time used is t 下降一最油 , and the corresponding maximum pressure change amount is ΔP max-漏油 , then Construct an oil leakage and lack of oil fault feature vector

[0187] Since the oil leakage and lack of oil fault is mainly characterized by this one key index, the feature vector is a one-dimensional vector.

[0188] S522: Calculate the Euclidean distance based on the fault feature vector and determine the fault status.

[0189] 1. Judgment of overheat fault status: Predetermine the different state vectors of overheat faults. For example, the "attention" state vector and the "early warning" state vector:

[0190]

[0191] Calculate the feature vector and the Euclidean distance from the "attention" state vector as well as the Euclidean distance from the "early warning" state vector :

[0192]

[0193] Compare the magnitudes of d 过热-注意 and d 过热-预警 . If d 过热-注意 < d 过热-预警 , it is determined that the overheat fault is in the "attention" state; conversely, if d 过热-注意 > d 过热-预警 , it is determined that the fault is in the "early warning" state.

[0194] 2. Judgment of lack or leakage fault status: For lack or leakage faults, also set the "attention" state vector and the "early warning" state vector:

[0195] S 漏油-注意 =(R 注意-漏油-变时 );

[0196]

[0197] Using the Euclidean distance formula (for one-dimensional vectors, the Euclidean distance is the absolute value of the difference between two values), calculate the feature vector and the Euclidean distance from the "attention" state vector as well as the Euclidean distance from the "early warning" state vector :

[0198] d 漏油-注意 =|R 下降-漏油 -R 注意-最油一变时 |;

[0199] d 漏油一预警 =|R 下降-漏油 -R 预警-漏油一变时 |;

[0200] Compare d 漏油一注意 and d 漏油-预警 . If d 漏油-注意 < d 漏油-预警 , it is determined that the lack or leakage fault is in the "attention" state; if d 漏油-注意 > d漏油-预警 , it is determined that it is in the "early warning" state.

[0201] Progressive relationship between step S3 and step S5: In step S3, the combined fault type diagnosis of pressure and UHF is carried out. By real-time monitoring of UHF and internal pressure signals of the bushing, and based on different value combinations of the signals, fault types such as internal discharge of the bushing, oil leakage problem, discharge of the outgoing line, local overheating, normal operation, etc. are initially judged. Step S5, the assessment of the overheating and oil leakage states of the bushing based on the Euclidean distance, focuses on the two specific faults of overheating and oil leakage diagnosed in step S3, and further quantitatively evaluates their states to determine whether it is in the "attention" or "early warning" level. It is a progressive link in the diagnosis process from the initial judgment of the fault type to the assessment of the severity of specific faults.

[0202] Inheritance and expansion of the data basis between step S3 and step S5: The pressure monitoring data in step S3 provides the original data basis for step S5. In step S5, characteristic parameters of the pressure curve (such as the maximum change amount, maximum change value / rise time, maximum change amount / fall time, etc.) are extracted to construct a feature vector. These parameters are further analysis and refinement of the pressure data in step S3 in terms of the time dimension and change amplitude to meet the need for accurate assessment of the overheating and oil leakage states.

[0203] Example test: With the help of a simulation experiment platform, historical data of pressure and UHF at different stages of different defects (such as discharge, overheating, oil leakage) are collected, such as 20 groups of data are collected. The above diagnostic method is used to evaluate the fault type and development stage, and the evaluation results are compared with the actual situation to calculate the accuracy rate to verify the effectiveness and accuracy of the technical solution.

[0204] The evaluation results of the discharge fault are shown in Table 4-3-5:

[0205] Table 4-3-5 Evaluation results of discharge fault

[0206]

[0207]

[0208] The evaluation results of the overheating fault are shown in Table 4-3-6:

[0209] Table 4-3-6 Evaluation results of overheating fault

[0210]

[0211]

[0212] The evaluation results of the oil leakage are shown in Table 4-3-7:

[0213] Table 4-3-7 Oil leakage and lack of oil evaluation results

[0214] Sample type Maximum change amount / Drop time Evaluation result Whether the result is correct Minor oil leakage 0.034 Attention Correct Minor oil leakage 0.051 Attention Correct Minor oil leakage 0.023 Attention Correct Minor oil leakage 0.044 Attention Correct Minor oil leakage 0.067 Warning Error Serious oil leakage 0.084 Warning Correct Serious oil leakage 0.1 Warning Correct Serious oil leakage 0.13 Warning Correct Serious oil leakage 0.96 Warning Correct Serious oil leakage 0.77 Warning Correct

[0215] The accuracy rate of the test results is shown in Table 4-3-8 as follows:

[0216] Table 4-3-8 Accuracy rate of test results

[0217]

[0218] Example 2

[0219] Such as Figure 3 , a combined diagnosis system for oil-paper bushing defects based on oil pressure and UHF, which is applied to the above-mentioned combined diagnosis method for oil-paper bushing defects based on oil pressure and UHF, includes:

[0220] A setting module, which is used to set a pressure sensor at the oil sampling port of the oil-paper bushing flange and set a UHF sensor outside the oil-paper bushing near the flange;

[0221] An acquisition module, which is used to collect the internal pressure data of the bushing in real time through the pressure sensor and collect the ultra-high frequency signals generated by partial discharge in real time through the UHF sensor;

[0222] A diagnosis module, which is used to conduct a combined fault type diagnosis based on the pressure data and ultra-high frequency signal data;

[0223] A first evaluation module, which is used to evaluate the discharge state of the bushing in a partial discharge state based on the fuzzy hierarchy method;

[0224] A second evaluation module, which is used to evaluate the state of the bushing with overheating or oil leakage or lack of oil based on the Euclidean distance division.

[0225] Example 3

[0226] A computer-readable storage medium, the computer-readable storage medium includes a stored program, wherein, when the program runs, it controls the device where the computer-readable storage medium is located to execute the above-mentioned combined diagnosis method for oil-paper bushing defects based on oil pressure and UHF.

[0227] Example 4

[0228] A processor, the processor is used to run a program, wherein, when the program runs, it executes the above-mentioned combined diagnosis method for oil-paper bushing defects based on oil pressure and UHF.

[0229] The present application provides a combined diagnosis method for oil-paper bushing defects based on oil pressure and UHF, comprising the following steps: A pressure sensor is arranged at the oil extraction port of the oil-paper bushing flange, and a UHF sensor is arranged outside the oil-paper bushing near the flange; The internal pressure data of the bushing is collected in real time through the pressure sensor, and the UHF signal generated by partial discharge is collected in real time through the UHF sensor; S3: Perform combined fault type diagnosis based on the pressure data and the UHF signal data; Evaluate the discharge state of the bushing in the partial discharge state based on the fuzzy hierarchy method; Evaluate the state of the bushing with overheating or oil leakage or oil shortage based on the Euclidean distance division. By enabling the pressure sensor and the UHF sensor to work together, various fault types that may occur in the bushing can be comprehensively covered. Compared with the traditional single detection method, the accuracy and comprehensiveness of fault diagnosis are greatly improved, and the situations of misjudgment and missed judgment are reduced.

[0230] Those of ordinary skill in the art can realize that the units of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components of each example have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0231] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0232] In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0233] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0234] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of various embodiments of the present invention, and they should all be covered within the scope of the claims and the description of the present invention.

Claims

1. A joint diagnosis method for oil-paper casing defects based on oil pressure and UHF, characterized in that: The following steps are involved: S1: Set the pressure sensor at the oil inlet of the oil-paper casing flange, and set the UHF sensor outside the oil-paper casing close to the flange; S2: The pressure sensor is used to collect the internal pressure data of the casing in real time, and the UHF sensor is used to collect the ultra-high frequency signal generated by partial discharge in real time; S3: Perform joint fault type diagnosis based on pressure data and UHF signal data; S4: Evaluate the discharge status of bushings with partial discharge based on fuzzy hierarchy process; S5: Based on the Euclidean distance division, the condition of the overheated or leaking casing is evaluated.

2. The oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to claim 1 is characterized in that: In step S4, the discharge state evaluation of the bushing with partial discharge based on the fuzzy hierarchy process includes the following steps: S41: Select UHF current change rate, UHF sudden increase rate, ΔP maximum value and ΔP sudden increase rate as evaluation parameters to construct an evaluation parameter set U; S42: Determine the weight coefficient of each parameter in the evaluation parameter set U by using the hierarchical analysis method; S43: determining the membership of each parameter in the evaluation parameter set U to different evaluation levels through a membership function to form an evaluation matrix; S44: Calculate a final evaluation result using the parameter weight and the evaluation matrix, and evaluate the discharge state of the bushing according to the evaluation result.

3. The oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to claim 2 is characterized in that: In step S42, the method of using the analytic hierarchy process to determine the weight coefficient of each parameter in the evaluation parameter set U includes the following steps: S421: using a scaling method to compare the relative importance of the parameters in the evaluation parameter set U in pairs, and forming a judgment matrix; S422: Calculate the eigenvector of the judgment matrix by the eigenroot method, and normalize the eigenvector to obtain the weight vector of the evaluation parameter set, wherein the weight vector includes the weight coefficient of each parameter in the evaluation parameter set; S423: Perform a consistency check to determine whether the judgment matrix is ​​accepted. If it is not accepted, return to step S421 to readjust the judgment matrix.

4. The oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to claim 3 is characterized in that: In step S422, the eigenvector of the judgment matrix is ​​calculated by the eigenroot method, and the eigenvector is normalized to obtain the weight vector of the evaluation parameter set, wherein the weight vector contains the weight coefficient of each parameter in the evaluation parameter set, including the following steps: Calculate the maximum eigenvalue λ of the judgment matrix A max ; Put λ max Substitute into the following formula to find the non-zero solution of the homogeneous linear equations. This non-zero solution is the eigenvector W0: (A-λ max )W0=0; Divide each component of the eigenvector W0 by the modulus of vector W0 || W0 ||, the formula is as follows: The normalized weight vector W is obtained, and the weight vector W contains the weight coefficients of each parameter in the evaluation parameter set, and the formula is as follows: W = (w1, w2, w3, w4); In the above formula, w1, w2, w3, and w4 correspond to weight coefficients of the UHF current change rate, the UHF sudden increase rate, the ΔP maximum value, and the ΔP sudden increase rate, respectively.

5. The oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to claim 4 is characterized in that: In step S43, the membership of each parameter in the evaluation parameter set to different evaluation levels is determined by the membership function to form an evaluation matrix, which includes the following steps: S431: setting different thresholds for different evaluation parameters in the evaluation parameter set; S432: constructing a triangular membership distribution function calculation table for different evaluation parameters in the evaluation parameter set according to the threshold value; S433: Form an evaluation matrix according to the triangular membership distribution function calculation table, the formula is as follows: In the above formula, R is the evaluation matrix; r ij It represents the membership of the i-th parameter to the j-th evaluation state.

6. The oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to claim 5 is characterized in that: In step S44, the final evaluation result is calculated by using the parameter weight and the evaluation matrix, and the discharge state of the bushing is evaluated according to the evaluation result, including the following steps: The calculation formula of the evaluation results is as follows: B 11 =w1×r 11 +w2×r 21 +w3×r 31 +w4×r 41 ; B 12 =w1×r 12 +w2×r 22 +w3×r 32 +w4×r 42 ; B=(B 11 B 12 ); If B 11 >B 12 , then the partial discharge state is judged to be in the "caution" state; if B 11 12 , then the partial discharge state is judged to be in the "warning" state;​ In the above formula, B 11 Indicates the degree to which the device is in the "attention" state after comprehensive consideration of various parameters; B 12 Indicates the degree to which the device is in the "warning" state.

7. The oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to claim 1 is characterized in that: In step S5, the state assessment of the overheated or leaking casing is performed based on the Euclidean distance division, including the following steps: S51: Determine the threshold of each influencing factor in the casing overheating and oil leakage status assessment model; S52: extract characteristic parameters of the pressure curve to construct a characteristic vector, calculate the Euclidean distance between the characteristic vector and different state vectors of the corresponding defect, and the defect state corresponding to the state vector with a closer distance is the defect state of the curve.

8. A joint diagnosis system for oil-paper casing defects based on oil pressure and UHF, characterized in that: The oil-paper casing defect joint diagnosis method based on oil pressure and UHF applied to any one of claims 1 to 7 comprises: A setting module is used to set the pressure sensor at the oil inlet of the oil-paper casing flange, and to set the UHF sensor outside the oil-paper casing near the flange; An acquisition module is used to acquire the internal pressure data of the casing in real time through a pressure sensor, and to acquire the ultra-high frequency signal generated by partial discharge in real time through a UHF sensor; A diagnosis module, which is used to perform joint fault type diagnosis based on pressure data and UHF signal data; A first evaluation module, which is used to evaluate the discharge state of the bushing in the partial discharge state based on the fuzzy hierarchy process; The second evaluation module is used to evaluate the state of the overheated or leaking oil casing based on Euclidean distance division.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program executes the oil-paper casing defect joint diagnosis method based on oil pressure and UHF according to any one of claims 1 to 7 when running.

Images