Sound wave diagnosis system and method for power grid equipment fault location

Abstract

The invention belongs to the technical field of fault positioning, and discloses a sound wave diagnosis system and method for power grid equipment fault positioning. The method comprises: acquiring a sound wave signal of a cable; carrying out de-noising processing on the collected sound wave signal; analyzing the de-noised sound wave signal, and judging whether a fault instruction is generated or not; if the fault instruction is generated, analyzing a corresponding fault type; according to the fault type, the propagation speed of the sound wave signal in the cable is evaluated; performing fault positioning on the cable according to the propagation speed of the sound wave signal; the method fully considers the propagation speed difference of the sound wave signal under the cable fault condition in the power distribution equipment, rapidly and accurately judges the cable fault type and positions the fault position, improves the cable fault positioning precision, improves the operation and maintenance level of a smart power grid, and guarantees the safety and reliability of power supply.

Description

Technical Field

[0001] The present invention relates to the technical field of fault location, and more specifically, to an acoustic wave diagnosis system and method for fault location of power grid equipment. Background Art

[0002] In recent years, with the development of the power system towards intelligence and informatization, the complexity of the power grid and the requirements for power supply reliability have been continuously increasing; the prevention and rapid location of power equipment failures have become the key to ensuring the safe and stable operation of the power grid; as an important power transmission and distribution equipment in the power system, cables include the transmission for connecting distribution equipment and power supply lines, and their fault diagnosis and location have always been a difficult point in power grid operation and maintenance; especially when connecting various distribution equipment, the connection of the lines is complex, and often some cable lines are set in blind areas, resulting in the situation that when a fault occurs and needs to be investigated, although the hardware structure of the distribution equipment has been investigated and no abnormalities are found, the fault still occurs, and only then is the cable fault considered. Due to the existence of blind areas in the cable lines, it is not easy to investigate faults. Common cable fault types include local faults such as short circuits, open circuits, insulation breakdowns, and partial discharges. These faults often occur in weak parts such as cable joints and terminals and have strong concealment;

[0003] Traditional cable fault location methods, such as time domain reflectometry and frequency domain analysis, are greatly affected by environmental interference and limited detection accuracy, and it is difficult to quickly and accurately locate the fault location; especially in a large power grid system, there is an associated relationship between high-voltage switchgear and metering cabinets. The high-voltage switchgear will be set up in different groups according to the power distribution requirements, and they are connected in series and parallel with each other and connected to the metering cabinets, thus forming the structure of the power distribution equipment. The location accuracy and efficiency of these methods are difficult to meet the requirements of intelligent operation and maintenance of the power grid; as an emerging cable fault location means, acoustic wave detection technology has advantages such as remote monitoring and non-destructive detection, providing effective support for cable fault diagnosis and rapid location.

[0004] Chinese Patent with Publication No. CN110161377A discloses a cable fault point location method and device; including: using a traveling wave monitoring terminal and an acoustic wave monitoring terminal to collect the time difference between the arrival time of the initial traveling wave wavefront and the arrival time of the initial acoustic wave wavefront generated after a fault occurs in the cable to be measured, wherein the traveling wave monitoring terminal and the acoustic wave monitoring terminal are installed at the same position of the cable to be measured, and the clocks of the traveling wave monitoring terminal and the acoustic wave monitoring terminal are synchronized; according to the time difference between the arrival time of the initial traveling wave wavefront and the arrival time of the initial acoustic wave wavefront, locate the fault point of the cable to be measured; this invention solves the problem that the double-ended traveling wave location method depends on the precise timing of GPS.

[0005] Although the above-mentioned technology can locate cable faults and solve the problem that the double-ended traveling wave location method depends on the precise time synchronization of GPS, in actual situations, different types of faults have different effects on the dielectric constant and mechanical properties of the cable, and thus different effects on the sound wave propagation speed. The above-mentioned technology does not consider the influence of different cable fault types on the propagation characteristics of sound waves in the cable, which will lead to large deviations in calculating the time difference, that is, there are large deviations in the positioning results, thereby reducing the positioning accuracy of faults and further affecting the safety and reliability of the smart grid.

[0006] In view of this, the present invention proposes an acoustic wave diagnosis system and method for fault location of power grid equipment to solve the above problems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art and achieve the above object, the present invention provides the following technical solutions: An acoustic wave diagnosis method for fault location of power grid equipment, including:

[0008] Collect the acoustic wave signals of the cable;

[0009] Denoise the collected acoustic wave signals;

[0010] Analyze the denoised acoustic wave signals to determine whether a fault instruction is generated;

[0011] If a fault instruction is generated, analyze the corresponding fault type;

[0012] Evaluate the propagation speed of the acoustic wave signals in the cable according to the fault type;

[0013] Locate the fault of the cable according to the propagation speed of the acoustic wave signals.

[0014] Further, the step of denoising the collected acoustic wave signals includes:

[0015] Step 1: Set the wavelet basis function and the decomposition level n;

[0016] Step 2: Perform i-layer wavelet decomposition on the acoustic wave signals using the wavelet basis function, decomposing them into a group of approximation terms and i groups of detail terms, where i ∈ [1, n];

[0017] Step 3: Calculate the signal-to-noise ratio of each group of detail terms;

[0018] Step 4: Preset the signal-to-noise threshold, compare the signal-to-noise threshold with the signal-to-noise ratios of each group of approximation terms and detail terms, remove the detail terms with signal-to-noise ratios less than the signal-to-noise threshold, and retain the detail terms with signal-to-noise ratios greater than or equal to the signal-to-noise threshold;

[0019] Step 5: Reconstruct the signal using the same wavelet basis function as in Step 2 according to the approximation terms and the i groups of detail terms after threshold processing;

[0020] Step 6: Calculate the signal-to-noise ratio of the reconstructed signal and compare it with a preset denoising threshold;

[0021] Step 7: If the signal ratio of the reconstructed signal is less than the denoising threshold, let i = i + 1 and return to Step 2, replacing the acoustic wave signal with the reconstructed signal; if the signal ratio of the reconstructed signal is greater than or equal to the denoising threshold, or i = n, then proceed to Step 8;

[0022] Step 8: Output the reconstructed signal, which is the acoustic wave signal after the denoising process is completed.

[0023] Further, the acoustic wave signal is a mechanical wave caused by minute vibrations of the substances inside the cable; a preset coupling interval is set, and m acoustic wave sensors are coupled to the cable surface according to the coupling interval;

[0024] The method for determining whether to generate a fault instruction includes:

[0025] Extract the corresponding characteristic parameters from each acoustic wave signal, where the characteristic parameters include amplitude, frequency, and duration; according to the current flow direction in the cable, different digital tags are sequentially set for the m acoustic wave sensors and marked as sensor tags, and the range of the sensor tags is [1, m]; mark the acoustic wave sensor that collects the acoustic wave signal as the acquisition sensor; sort each amplitude from largest to smallest, and mark the two acquisition sensors corresponding to the two largest amplitudes as the first acquisition sensors;

[0026] Determine whether the sensor tags of the two first acquisition sensors are adjacent; if not adjacent, no fault sensor is generated; if adjacent, mark the first acquisition sensor with the larger sensor tag as the first sensor, and mark the first acquisition sensor with the smaller sensor tag as the second sensor; mark the acquisition sensors with sensor tags larger than the first sensor as the third sensors, and mark the acquisition sensors with sensor tags smaller than the second sensor as the fourth sensors; sort the amplitudes corresponding to each third sensor from smallest to largest and generate a first sorting table; sort the amplitudes corresponding to each fourth sensor from largest to smallest and generate a second sorting table;

[0027] If the order of the first sorting table is consistent with the order of the sensor tags corresponding to the third sensors, and the order of the second sorting table is consistent with the order of the sensor tags corresponding to the fourth sensors, then generate a fault instruction.

[0028] Further, the method for analyzing the fault type includes:

[0029] Use the characteristic parameters corresponding to each acoustic wave signal as analysis data, input the analysis data into a trained type analysis model, output the corresponding type tag, where the type tag is the digital tag corresponding to the fault type, and obtain the fault type of the cable according to the type tag;

[0030] The training process of the type analysis model includes:

[0031] Set different digital tags for different fault types and mark them as type tags; pre-collect g groups of analysis data, and set corresponding type tags for all g groups of analysis data, where g is an integer greater than 1;

[0032] Convert the analysis data and the corresponding type tags into a corresponding set of feature vectors; each set of feature vectors is used as the input of the type analysis model, and the type analysis model outputs a set of predicted type tags corresponding to each set of analysis data, with the actual type tag corresponding to each set of analysis data as the prediction target, and the actual type tag is the type tag pre-collected corresponding to the analysis data; use minimizing the sum of prediction errors of all analysis data as the training target; train the type analysis model until the sum of prediction errors reaches convergence and then stop training; the type analysis model is a deep neural network model.

[0033] Further, the method for evaluating the propagation speed of the acoustic wave signal in the cable includes:

[0034] Preset a fault range and a normal propagation speed, where the fault range includes a fault amplitude range and a fault frequency range; subtract the minimum value corresponding to the maximum value in the fault amplitude range to obtain an amplitude difference; subtract the minimum value corresponding to the maximum value in the fault frequency range to obtain a frequency difference; compare the amplitude corresponding to the first sensor with the amplitude corresponding to the second sensor, mark the acoustic wave sensor with the larger amplitude as the proximal sensor, and mark the acoustic wave sensor with the smaller amplitude as the distal sensor; obtain the amplitude corresponding to the proximal sensor and mark it as the proximal amplitude; obtain the frequency corresponding to the proximal sensor and mark it as the proximal frequency; subtract the maximum value in the fault amplitude range from the proximal amplitude to obtain a changed amplitude; subtract the maximum value in the fault frequency range from the proximal frequency to obtain a changed frequency;

[0035] Divide the changed amplitude by the amplitude difference to obtain an amplitude reduction coefficient; divide the changed frequency by the frequency difference to obtain a frequency reduction coefficient; use the amplitude reduction coefficient, the frequency reduction coefficient, and the predicted type tag as test data, input the test data into the trained coefficient prediction model to predict the corresponding speed reduction coefficient, and mark it as a single-fault reduction coefficient; the training process of the coefficient prediction model is the same as that of the type analysis model, and both are deep neural network models;

[0036] Multiply the single-fault reduction coefficient by the normal propagation speed to obtain the single-fault propagation speed.

[0037] Further, obtain the times when the proximal sensor and the distal sensor collect the acoustic wave signals respectively; mark the time when the proximal sensor collects the acoustic wave signal as the first time, and mark the time when the distal sensor collects the acoustic wave signal as the second time; subtract the first time from the second time to obtain the time difference; multiply the time difference by the single fault propagation speed to obtain the distance difference; subtract the distance difference from the coupling interval and then divide the result by 2 to obtain the first distance; the first distance is the distance between the proximal sensor and the cable fault location; add the distance difference to the first distance to obtain the second distance, and the second distance is the distance between the distal sensor and the cable fault location; locate the cable fault location according to the first distance and the second distance.

[0038] Further, the method further includes: if multiple acoustic wave signals are collected by the same acoustic wave sensor, classify all the collected acoustic wave signals to obtain P signal sets, and each signal set corresponds to a cable fault location respectively.

[0039] The method for obtaining the P signal sets includes:

[0040] Obtain the collection time of each acoustic wave signal; sort each collection time from earliest to latest; obtain the sensor label corresponding to the acoustic wave sensor corresponding to each acoustic wave signal; replace each collection time with the corresponding sensor label to generate a third sorted list; analyze the third sorted list in ascending order, add the first two sensor labels in the third sorted list to a signal set, and starting from the third sensor label, classify each sensor label in turn according to the classification rule to obtain P signal sets. There are no identical sensor labels in each signal set, and P is an integer greater than 1.

[0041] Further, the classification rule includes:

[0042] Mark the currently classified sensor label as the current label.

[0043] If the current label is not adjacent to each sensor label in all the signal sets, add the current label to a new signal set.

[0044] If the current label is adjacent to the sensor label in a signal set with only one sensor label, add the current label to the corresponding signal set.

[0045] If the current label is adjacent to the sensor labels in multiple signal sets respectively, obtain the collection times corresponding to all the sensor labels adjacent to the current label, sort them, and add the current label to the signal set corresponding to the sensor label with the earliest collection time.

[0046] If the current label is adjacent to the sensor label in a signal set, add the current label to the corresponding signal set.

[0047] Further, according to the fault type of each cable fault, analyze the propagation speed of the acoustic wave signal in the cable;

[0048] According to the third sorting table, obtain the acoustic wave signals corresponding to the acquisition time of each sensor tag in each signal set, extract the corresponding characteristic parameters, and then obtain the corresponding type tags for each signal set respectively; according to the characteristic parameters corresponding to the first sensor tag and the second sensor tag in each signal set, calculate the amplitude reduction coefficient and frequency reduction coefficient corresponding to each signal set; use the amplitude reduction coefficient, frequency reduction coefficient and type tag corresponding to each signal set as research data, input the research data into the trained coefficient prediction model, predict the corresponding speed reduction coefficient, and mark it as the multi-fault reduction coefficient; multiply the multi-fault reduction coefficient by the normal propagation speed to obtain the multi-fault propagation speed; the single-fault propagation speed is the propagation speed of the acoustic wave signal in the cable under the condition of multiple cable faults.

[0049] An acoustic wave diagnosis system for fault location of power grid equipment, implementing the acoustic wave diagnosis method for fault location of power grid equipment, includes:

[0050] An acoustic wave acquisition module, used to acquire the acoustic wave signals of the cable;

[0051] An acoustic wave processing module, used to perform denoising processing on the acquired acoustic wave signals;

[0052] A fault judgment module, used to analyze the denoised acoustic wave signals and judge whether to generate a fault instruction;

[0053] A type analysis module, used to analyze the corresponding fault type if a fault instruction is generated;

[0054] A speed evaluation module, used to evaluate the propagation speed of the acoustic wave signal in the cable according to the fault type;

[0055] A fault location module, used to perform fault location on the cable according to the propagation speed of the acoustic wave signal.

[0056] The technical effects and advantages of the acoustic wave diagnosis system and method for fault location of power grid equipment of the present invention:

[0057] 1. By collecting cable acoustic signals, denoising the acoustic signals using wavelet decomposition technology, and extracting the characteristic parameters of the acoustic signals to determine whether there is a fault in the cable; using deep learning technology to analyze and identify the characteristic parameters of the acoustic signals in the faulty cable, identifying the cable fault type, and accurately calculating the propagation speed of the acoustic signals based on the influence of different cable fault types on the propagation speed of the acoustic signals in the cable, so as to achieve rapid fault location; fully considering the differences in the propagation speed of acoustic signals under different cable fault conditions, quickly and accurately judging the cable fault type and locating the fault position, improving the cable fault location accuracy, thereby enhancing the operation and maintenance level of the smart grid and ensuring the safety and reliability of power supply.

[0058] 2. According to the propagation characteristics of the acoustic signals, combined with the acquisition time of the acoustic signals and the positional relationship of the acoustic sensors, each acoustic signal is reasonably classified to accurately obtain the acoustic signals caused by each cable fault; and through deep learning technology, analyze the propagation speed of the acoustic signals under the combined influence of multiple cable faults, providing an accurate basis for subsequent fault location; to improve the applicability of the location method in complex fault scenarios, achieve precise fault location in the case of multiple cable faults, effectively solve the location difficulties in the case of multiple faults, further improve the fault location accuracy, and thus improve the safety and reliability of the smart grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Schematic diagram of the installation of acoustic sensors in the capacitor cabinet of the distribution equipment;

[0060] Figure 2 Schematic diagram of the installation of acoustic sensors in the isolating switch cabinet of the distribution equipment;

[0061] Figure 3 Schematic diagram of the installation of acoustic sensors in the distribution cabinet of the distribution equipment;

[0062] Figure 4 Schematic diagram of an acoustic diagnosis system for fault location of power grid equipment in Embodiment 1 of the present invention;

[0063] Figure 5 Schematic diagram of the marking of acoustic sensors in Embodiment 1 of the present invention

[0064] Figure 6 Schematic diagram of an acoustic diagnosis system for fault location of power grid equipment in Embodiment 2 of the present invention;

[0065] Figure 7 Schematic diagram of multiple cable faults in Embodiment 2 of the present invention;

[0066] Figure 8 Flowchart of an acoustic diagnosis method for fault location of power grid equipment in Embodiment 3 of the present invention.

[0067] Reference numerals: 1. Acoustic wave sensor. Specific embodiments

[0068] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. 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.

[0069] Embodiment 1

[0070] Please refer to Figures 1 to 4 As shown, a sonic diagnostic system for fault location of power grid equipment in this embodiment is used to locate and diagnose cable line faults between power distribution equipment, and includes an acoustic wave acquisition module, an acoustic wave processing module, a fault judgment module, a type analysis module, a speed evaluation module, and a fault location module; each module is connected by wired and / or wireless means to realize data transmission between modules;

[0071] The acoustic wave acquisition module is used to acquire the acoustic wave signal of the cable. The acoustic wave acquisition module is arranged in the power distribution equipment. As shown in Figures 1 to 3 As shown, for different power distribution equipment, the positions of the acoustic wave sensors 1 are different. At least 2 acoustic wave sensors that can cover the corresponding positions can be set at each corresponding wiring interface.

[0072] The acoustic wave signal is a mechanical wave caused by the tiny vibration of the substances inside the cable; a preset coupling interval is set, and m acoustic wave sensors are coupled to the cable surface according to the coupling interval, and the acoustic wave signal of the cable is acquired through the acoustic wave sensors; the acoustic wave sensors are, for example, piezoelectric acoustic wave sensors, fiber optic acoustic wave sensors, etc.; the coupling interval is preset by those skilled in the art according to the actual length and structure of the cable.

[0073] It should be noted that when the cable is in normal operation, the generated acoustic wave signal has very small energy, usually between 20HZ and 500HZ, and the amplitude is relatively low. The structure and material of the cable itself have a certain acoustic wave absorption and blocking effect, so the generated acoustic wave signal is difficult to be acquired by the acoustic wave sensor; while when the cable fails, obvious high-frequency acoustic wave signals will be generated to be acquired by the acoustic wave sensor.

[0074] The acoustic wave processing module is used to perform denoising processing on the acquired acoustic wave signal.

[0075] The steps of performing denoising processing on the acquired acoustic wave signal include:

[0076] Step 1: Set the wavelet basis function and the decomposition level n;

[0077] Wavelet basis functions such as Daubechies wavelets, Symlet wavelets, Coiflet wavelets, etc.; the decomposition level n is set by those skilled in the art by collecting the corresponding acoustic signals multiple times during historical cable faults and setting the decomposition level n according to the mean value of the frequencies of the multiple acoustic signals.

[0078] Step 2: Perform i-level wavelet decomposition on the acoustic signal using a wavelet basis function, decomposing it into a set of approximation terms and i sets of detail terms, where i ∈ [1, n]; when performing wavelet decomposition on the acoustic signal for the first time, i is 1.

[0079] Step 3: Calculate the signal-to-noise ratio of each set of detail terms;

[0080] The calculation method of the signal-to-noise ratio is prior art and will not be elaborated here; the approximation terms mainly reflect the low-frequency trend components of the signal and basically do not contain noise. Therefore, only the signal-to-noise ratio of the detail terms needs to be calculated for subsequent denoising processing.

[0081] Step 4: Preset a signal-to-noise threshold, compare the signal-to-noise threshold with the signal-to-noise ratio of each set of approximation terms and detail terms, remove the detail terms with a signal-to-noise ratio less than the signal-to-noise threshold, and retain the detail terms with a signal-to-noise ratio greater than or equal to the signal-to-noise threshold;

[0082] The signal-to-noise threshold is set by those skilled in the art by collecting environmental noise signals multiple times during the normal operation of historical cables and calculating the corresponding signal-to-noise ratios, and taking the mean value of the multiple signal-to-noise ratios as the signal-to-noise threshold.

[0083] Step 5: According to the approximation terms and the i sets of detail terms after threshold processing, perform signal reconstruction using the same wavelet basis function as in Step 2.

[0084] Step 6: Calculate the signal-to-noise ratio of the reconstructed signal and compare it with the preset denoising threshold;

[0085] The denoising threshold is selected by those skilled in the art in a test site with low environmental noise, collecting acoustic signals during cable faults multiple times, and calculating the corresponding signal-to-noise ratios, and taking the mean value of the multiple signal-to-noise ratios as the denoising threshold.

[0086] Step 7: If the signal ratio of the reconstructed signal is less than the denoising threshold, let i = i + 1 and return to Step 2, replacing the acoustic signal with the reconstructed signal; if the signal ratio of the reconstructed signal is greater than or equal to the denoising threshold, or i = n, then go to Step 8.

[0087] Step 8: Output the reconstructed signal, which is the acoustic signal after the denoising process is completed.

[0088] It should be noted that the purpose of denoising the collected acoustic wave signals is to remove the environmental noise in the acoustic wave signals, improve the signal-to-noise ratio of the acoustic wave signals, highlight the essential characteristics of the acoustic wave signals, thereby improving the quality of the acoustic wave signals and enhancing the accuracy and reliability of subsequent cable fault diagnosis.

[0089] The fault judgment module is used to analyze the denoised acoustic wave signals and judge whether a fault instruction is generated.

[0090] The method for judging whether a fault instruction is generated includes:

[0091] Extract the corresponding characteristic parameters from each acoustic wave signal. The characteristic parameters include amplitude, frequency, and duration. The amplitude represents the strength of the acoustic wave signal, that is, the sound pressure intensity, which is used to measure the power of the sound source generating the acoustic wave. The frequency represents the spectral components of the acoustic wave signal. The duration is the time length from the start to the end of the acoustic wave signal, reflecting the time-domain characteristics of the acoustic wave. Extracting the corresponding characteristic parameters from the acoustic wave signal is a prior art and will not be elaborated here.

[0092] According to the current flow direction in the cable, different digital tags are sequentially set for m acoustic wave sensors and marked as sensor tags, and the range of the sensor tags is [1, m]. The acoustic wave sensor that collects the acoustic wave signal is marked as the acquisition sensor. Sort each amplitude from large to small, and mark the two acquisition sensors corresponding to the two largest amplitudes as the first acquisition sensors.

[0093] Judge whether the sensor tags of the two first acquisition sensors are adjacent. If they are not adjacent, no fault sensor is generated. If they are adjacent, as Figure 5 shown, mark the first acquisition sensor with the larger sensor tag as the first sensor, and mark the first acquisition sensor with the smaller sensor tag as the second sensor. Mark the acquisition sensors with sensor tags larger than the first sensor as the third sensors, and mark the acquisition sensors with sensor tags smaller than the second sensor as the fourth sensors. Sort the amplitudes corresponding to each third sensor from small to large and generate a first sorting table. Sort the amplitudes corresponding to each fourth sensor from large to small and generate a second sorting table.

[0094] If the order of the first sorting table is consistent with the order of the sensor tags corresponding to the third sensors, and the order of the second sorting table is consistent with the order of the sensor tags corresponding to the fourth sensors, a fault instruction is generated.

[0095] Exemplarily, acoustic wave sensors with sensor tags 4, 5, 6, 7, 8, and 9 have collected acoustic wave signals. Among them, the acoustic wave sensors with sensor tags 6 and 7 have the largest corresponding amplitudes. Therefore, the acoustic wave sensor with sensor tag 7 is the first sensor, and the acoustic wave sensor with sensor tag 6 is the second sensor; the acoustic wave sensors with sensor tags 8 and 9 are the third sensors, and the acoustic wave sensors with sensor tags 4 and 5 are the fourth sensors; at this time, the amplitude corresponding to the acoustic wave sensor with sensor tag 4 is less than the amplitude corresponding to the acoustic wave sensor with sensor tag 5, and the amplitude corresponding to the acoustic wave sensor with sensor tag 9 is greater than the amplitude corresponding to the acoustic wave sensor with sensor tag 8. Therefore, the order in the first sorting table is 8, 9, and the order in the second sorting table is 4, 5; since the order of the sensor tags corresponding to the third sensors is also 8, 9, and the order of the sensor tags corresponding to the fourth sensors is also 4, 5, a fault instruction is generated.

[0096] It should be noted that the reason for judging whether there is a fault instruction based on the sensor tag is that when the cable fails, obvious high-frequency acoustic wave signals will be generated and collected by the acoustic wave sensors. In addition, due to complex geological conditions, the cable will also generate acoustic wave signals during terrain changes, such as earthquakes, road construction, landslides, etc.; the difference between the two is that when the acoustic wave signal is generated by a cable fault, since the cable fault only occurs at one point or multiple points at different times, the acoustic wave signal propagates from the fault point to both sides, and the acoustic wave signal gradually weakens as the propagation distance increases. The two sensors with the largest corresponding amplitudes are the two sensors closest to both sides of the fault point, and the remaining amplitudes should decrease as the distance between the corresponding sensor and the fault point increases, and the time when each sensor receives the acoustic wave signal is different; when the acoustic wave signal is generated by terrain changes, the amplitude corresponding to each sensor is irregular, and the time when each sensor receives the acoustic wave signal is basically the same; therefore, by analyzing the amplitude change law of the acoustic wave signal through the sensor tag, it can be judged whether the acoustic wave signal is caused by a cable fault.

[0097] A type analysis module is used to analyze the corresponding fault type if a fault instruction is generated.

[0098] The methods for analyzing the fault type include:

[0099] Taking the characteristic parameters corresponding to each acoustic wave signal as analysis data, inputting the analysis data into a trained type analysis model, outputting the corresponding type label, where the type label is the digital label corresponding to the fault type, and obtaining the fault type of the cable according to the type label; the fault types are, for example, short circuit, open circuit, insulation breakdown, partial discharge, etc.

[0100] The specific training process of the type analysis model includes:

[0101] Set different digital tags for different fault types and mark them as type tags; collect g groups of analysis data in advance, and set corresponding type tags for all g groups of analysis data, where g is an integer greater than 1; the type tags corresponding to the analysis data are determined by those skilled in the art during the historical cable fault diagnosis process. Collect g groups of analysis data, and under the conditions of each group of analysis data, diagnose the corresponding fault type in sequence, and set the type tags of the corresponding diagnosed fault type for each group of analysis data; set the corresponding type tags for the g groups of analysis data in sequence;

[0102] Convert the analysis data and the corresponding type tags into a corresponding set of feature vectors; each set of feature vectors is used as the input of the type analysis model. The type analysis model takes a set of predicted type tags corresponding to each group of analysis data as the output, and takes the actual type tag corresponding to each group of analysis data as the prediction target. The actual type tag is the type tag corresponding to the analysis data collected in advance; take minimizing the sum of the prediction errors of all analysis data as the training target; among them, the calculation formula of the prediction error is η K =(β K -ε K ) 2 , where η K is the prediction error, K is the group number of the feature vectors corresponding to the analysis data, β K is the predicted type tag corresponding to the Kth group of analysis data, and ε K is the actual type tag corresponding to the Kth group of analysis data; train the type analysis model until the sum of the prediction errors reaches convergence and then stop training.

[0103] The above type analysis model is specifically a deep neural network model; it includes an input layer, a hidden layer, and an output layer; each hidden layer includes multiple neurons, and there are connections between each neuron and the neurons in the next layer. The connections contain weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer. The activation function introduces non-linearity, allowing the network to learn more complex patterns and features.

[0104] It should be noted that the reason for judging the cable fault type according to the characteristic parameters is that different cable fault types correspond to different characteristic parameters; when the cable fault is a short - circuit fault, a large current will be generated in the cable, triggering a strong arc discharge. At this time, the amplitude of the acoustic wave signal generated is very high, the duration is short, and there are many high - frequency components in the spectrum, showing a broadband characteristic; when the cable fault is an open - circuit fault, the conductor inside the cable is disconnected, and a short - term arc discharge will occur. At this time, the amplitude of the acoustic wave signal generated is smaller than that of the short - circuit fault, the duration is also short, and the spectrum of the acoustic wave signal is concentrated in the low - frequency band, showing a narrow - band characteristic; when the cable fault is an insulation breakdown fault, partial discharge will occur inside the cable, the duration of the acoustic wave signal is long, the spectrum distribution is concentrated in the middle - low - frequency band, and the amplitude is low; when the cable fault is a partial discharge fault, a weak local discharge acoustic wave signal will be generated, the duration is short, the amplitude is low, and the spectrum distribution is concentrated in the middle - high - frequency band.

[0105] A speed evaluation module, which is used to evaluate the propagation speed of the acoustic wave signal in the cable according to the fault type.

[0106] The methods for evaluating the propagation speed of the acoustic wave signal in the cable include:

[0107] Preset a fault range and a normal propagation speed. The fault range includes a fault amplitude range and a fault frequency range; the fault range is obtained by those skilled in the art by collecting the acoustic wave signals emitted by the cable multiple times during the historical cable fault process and extracting the corresponding characteristic parameters. The maximum and minimum values among multiple amplitudes are used as the fault amplitude range, and the maximum and minimum values among multiple frequencies are used as the fault frequency range; the normal propagation speed is obtained by those skilled in the art using the pulse reflection method during the historical normal operation of the cable.

[0108] Subtract the minimum value from the maximum value in the fault amplitude range to obtain the amplitude difference; subtract the minimum value from the maximum value in the fault frequency range to obtain the frequency difference; compare the amplitude corresponding to the first sensor with the amplitude corresponding to the second sensor, mark the acoustic wave sensor with the larger amplitude as the proximal sensor, and mark the acoustic wave sensor with the smaller amplitude as the distal sensor; obtain the amplitude corresponding to the proximal sensor and mark it as the proximal amplitude; obtain the frequency corresponding to the proximal sensor and mark it as the proximal frequency; subtract the maximum value in the fault amplitude range from the proximal amplitude to obtain the changed amplitude; subtract the maximum value in the fault frequency range from the proximal frequency to obtain the changed frequency.

[0109] Divide the change amplitude by the amplitude difference to obtain the amplitude reduction coefficient; divide the change frequency by the frequency difference to obtain the frequency reduction coefficient; use the amplitude reduction coefficient, the frequency reduction coefficient, and the predicted type label as test data, and input the test data into the trained coefficient prediction model to predict the corresponding speed reduction coefficient, which is marked as the single-fault reduction coefficient; the specific training process of the coefficient prediction model is the same as that of the type analysis model, and both are deep neural network models;

[0110] Multiply the single-fault reduction coefficient by the normal propagation speed to obtain the single-fault propagation speed; the single-fault propagation speed is the propagation speed of the acoustic wave signal in the cable under the condition of a single cable fault.

[0111] It should be noted that different types of faults will affect the dielectric constant and mechanical properties of the cable, thereby affecting the propagation speed of the acoustic wave signal in the cable; for example, a short-circuit fault will cause the insulator in the cable to break down, resulting in a decrease in the dielectric constant, and at the same time, the conductor in the cable will melt and deform; an open-circuit fault will cause the insulator in the cable to lose charge, resulting in an increase in the dielectric constant, and at the same time, the insulation layer in the cable will crack or detach; the change in the dielectric constant will affect the electric field distribution inside the cable, thereby changing the propagation speed of the acoustic wave signal inside the cable; the change in the mechanical properties will change the stiffness and density of the cable, thereby affecting the propagation speed of the acoustic wave signal inside the cable.

[0112] A fault location module is used to locate the cable fault according to the propagation speed of the acoustic wave signal.

[0113] Obtain the time when the acoustic wave signal is collected by the proximal sensor and the distal sensor respectively; mark the time when the acoustic wave signal is collected by the proximal sensor as the first time, and mark the time when the acoustic wave signal is collected by the distal sensor as the second time; the first time and the second time are obtained from the timestamps recorded when the corresponding acoustic wave sensors collect the acoustic wave signals; subtract the first time from the second time to obtain the time difference; multiply the time difference by the single-fault propagation speed to obtain the distance difference; subtract the distance difference from the coupling interval and then divide by 2 to obtain the first distance; the first distance is the distance between the proximal sensor and the cable fault; add the distance difference to the first distance to obtain the second distance, and the second distance is the distance between the distal sensor and the cable fault; locate the cable fault according to the first distance and the second distance.

[0114] In this embodiment, by collecting cable acoustic signals, denoising the acoustic signals using wavelet decomposition technology, and extracting the characteristic parameters of the acoustic signals to determine whether a fault occurs in the cable; using deep learning technology to analyze and identify the characteristic parameters of the acoustic signals in the faulty cable, identifying the cable fault type, and accurately calculating the propagation speed of the acoustic signals according to the influence of different cable fault types on the propagation speed of the acoustic signals in the cable, so as to achieve rapid fault location; fully considering the propagation speed difference of the acoustic signals under different cable fault conditions, quickly and accurately judging the cable fault type and locating the fault position, improving the cable fault location accuracy, thereby enhancing the intelligent power grid operation and maintenance level and ensuring the safety and reliability of power supply.

[0115] Embodiment 2

[0116] Please refer to Figure 6 As shown, this embodiment further improves the design on the basis of Embodiment 1. The method in Embodiment 1 is for fault location when there is one fault in the cable. When multiple faults occur in the cable in a short time, since the acoustic signals will propagate in the cable, there will be a situation where a single acoustic wave sensor collects multiple acoustic signals in a short time; at this time, using the method in Embodiment 1 cannot effectively distinguish which fault the acoustic signals collected by the acoustic wave sensor are specifically caused by, resulting in the inability to accurately locate the fault; therefore, this embodiment provides an acoustic diagnosis system for fault location of power grid equipment, which also includes a signal classification module and a speed analysis module;

[0117] The signal classification module is used to classify all the collected acoustic signals if a single acoustic wave sensor collects multiple acoustic signals, and obtain P signal sets, and each signal set corresponds to a cable fault respectively.

[0118] The method for obtaining P signal sets includes:

[0119] Obtain the acquisition time of each acoustic signal; sort each acquisition time from earliest to latest; obtain the sensor label corresponding to the acoustic wave sensor corresponding to each acoustic signal; replace each acquisition time with the corresponding sensor label to generate a third sorting table; analyze according to the forward order of the third sorting table, add the first two sensor labels in the third sorting table to a signal set, and starting from the third sensor label, classify each sensor label in turn according to the classification rule, and obtain P signal sets in total. There are no identical sensor labels in each signal set, and P is an integer greater than 1.

[0120] The classification rule includes:

[0121] Mark the currently classified sensor label as the current label;

[0122] If the current tag is not adjacent to each sensor tag in all signal sets, add the current tag to a new signal set;

[0123] If the current tag is adjacent to the sensor tag in a signal set with only one sensor tag, add the current tag to the corresponding signal set;

[0124] If the current tag is adjacent to the sensor tags in multiple signal sets respectively, obtain the acquisition times corresponding to all the sensor tags adjacent to the current tag, sort them, and add the current tag to the signal set corresponding to the sensor tag with the earliest acquisition time;

[0125] If the current tag is adjacent to the sensor tag in a signal set, add the current tag to the corresponding signal set.

[0126] Exemplarily, there are acoustic wave sensors with sensor tags 1 - 8 on the cable, and cable faults occur between sensor tags 2 and 3 and between sensor tags 5 and 6; mark the cable fault between sensor tags 2 and 3 as the first fault, and mark the cable fault between sensor tags 5 and 6 as the second fault; the generation time of the first fault is earlier than that of the second fault, and the order of the third sorting table is 5, 6, 3, 2, 4, 7, 4, 1, 3, 8; add 5 and 6 to a signal set and mark it as the first set; since 3 is not adjacent to 5 and 6 in the first set, add 3 to a new signal set and mark it as the second set; since 2 is adjacent to 3 in the second set and there is only one 3 in the second set, add 2 to the second set; since 4 is adjacent to 5 in the first set and 3 in the second set, but the acquisition time of 5 is earlier than that of 3, add 4 to the first set; since 7 is adjacent to 6 in the first set, add 7 to the first set; since 4 is adjacent to 5 in the first set and 3 in the second set, but there is already a 4 in the first set, add 4 to the second set; since 1 is adjacent to 2 in the second set, add 1 to the second set; since there is already a 3 in the second set and 3 is adjacent to 4 in the first set, add 3 to the first set; since 8 is adjacent to 7 in the first set, add 8 to the first set; among them, the first set includes 5, 6, 4, 7, 3, 8, and the second set includes 3, 2, 4, 1, all of which conform to the propagation law of acoustic wave signals. The specific schematic diagram is as Figure 7 shown. The dotted line in the figure is the propagation range of the acoustic wave signal generated by the first fault, and the thick solid line is the propagation range of the acoustic wave signal generated by the second fault.

[0127] A velocity analysis module for analyzing the propagation velocity of acoustic wave signals in the cable according to the fault type of each cable fault.

[0128] According to the third sorting table, obtain the acoustic wave signals corresponding to the acquisition times of each sensor tag in each signal set, extract the corresponding characteristic parameters, and then obtain the corresponding type tags for each signal set respectively; according to the characteristic parameters corresponding to the first sensor tag and the second sensor tag in each signal set, calculate the amplitude reduction coefficient and frequency reduction coefficient corresponding to each signal set; use the amplitude reduction coefficient, frequency reduction coefficient and type tag corresponding to each signal set as research data, input the research data into the trained coefficient prediction model, predict the corresponding speed reduction coefficient, and mark it as the multi-fault reduction coefficient; multiply the multi-fault reduction coefficient by the normal propagation speed to obtain the multi-fault propagation speed; the single-fault propagation speed is the propagation speed of the acoustic wave signal in the cable under the condition of multiple cable faults.

[0129] It should be noted that when multiple identical faults occur in the cable, the subsequent fault location is the same as the fault location method when a single fault occurs in the cable, that is, the P signal sets are respectively regarded as P cables with a single fault for fault location.

[0130] In this embodiment, according to the propagation characteristics of the acoustic wave signal, combined with the acquisition time of the acoustic wave signal and the position relationship of the acoustic wave sensors, each acoustic wave signal is reasonably classified to accurately obtain the acoustic wave signal caused by each cable fault; and through deep learning technology, analyze the propagation speed of the acoustic wave signal under the combined influence of multiple cable faults, providing an accurate basis for subsequent fault location; to improve the applicability of the location method in complex fault scenarios, realize accurate fault location in the case of multiple cable faults, effectively solve the location difficulties in the case of multiple faults, further improve the fault location accuracy, and thus improve the safety and reliability of the smart grid.

[0131] Embodiment 3

[0132] Please refer to Figure 8 As shown, for the parts not described in detail in this embodiment, see the descriptions in Embodiment 1 and Embodiment 2. Provide an acoustic wave diagnosis method for fault location of power grid equipment. The method includes:

[0133] Collect the acoustic wave signals of the cable;

[0134] Perform denoising processing on the collected acoustic wave signals;

[0135] Analyze the denoised acoustic wave signals to determine whether a fault instruction is generated;

[0136] If a fault instruction is generated, analyze the corresponding fault type;

[0137] According to the fault type, evaluate the propagation speed of the acoustic wave signal in the cable;

[0138] Fault locate the cable according to the propagation speed of the acoustic wave signal.

[0139] Further, the step of denoising the collected acoustic wave signal includes:

[0140] Step 1: Set the wavelet basis function and the decomposition level n;

[0141] Step 2: Perform i-level wavelet decomposition on the acoustic wave signal using the wavelet basis function, decomposing it into a group of approximation terms and i groups of detail terms, where i ∈ [1, n];

[0142] Step 3: Calculate the signal-to-noise ratio of each group of detail terms;

[0143] Step 4: Preset a signal-to-noise threshold, compare the signal-to-noise threshold with the signal-to-noise ratios of each group of approximation terms and detail terms, remove the detail terms with a signal-to-noise ratio less than the signal-to-noise threshold, and retain the detail terms with a signal-to-noise ratio greater than or equal to the signal-to-noise threshold;

[0144] Step 5: According to the approximation terms and the i groups of detail terms after threshold processing, perform signal reconstruction using the same wavelet basis function as in Step 2;

[0145] Step 6: Calculate the signal-to-noise ratio of the reconstructed signal and compare it with the preset denoising threshold;

[0146] Step 7: If the signal ratio of the reconstructed signal is less than the denoising threshold, let i = i + 1 and return to Step 2, replacing the acoustic wave signal with the reconstructed signal; if the signal ratio of the reconstructed signal is greater than or equal to the denoising threshold, or i = n, then go to Step 8;

[0147] Step 8: Output the reconstructed signal, which is the acoustic wave signal after the denoising process is completed.

[0148] Further, the acoustic wave signal is a mechanical wave caused by the minute vibration of the internal substances of the cable; preset a coupling interval, and couple m acoustic wave sensors to the surface of the cable according to the coupling interval;

[0149] The method for determining whether to generate a fault instruction includes:

[0150] Extract the corresponding characteristic parameters from each acoustic wave signal, where the characteristic parameters include amplitude, frequency, and duration; according to the current flow direction in the cable, set different digital tags for the m acoustic wave sensors in sequence and label them as sensor tags, where the range of the sensor tags is [1, m]; label the acoustic wave sensor that collects the acoustic wave signal as the acquisition sensor; sort each amplitude from largest to smallest, and label the two acquisition sensors corresponding to the two largest amplitudes as the first acquisition sensors;

[0151] Determine whether the sensor tags of the two first-sampling sensors are adjacent; if not, no faulty sensor is generated; if adjacent, mark the first-sampling sensor with the larger sensor tag as the first sensor, and mark the first-sampling sensor with the smaller sensor tag as the second sensor; mark the acquisition sensors with sensor tags larger than the first sensor as the third sensors, and mark the acquisition sensors with sensor tags smaller than the second sensor as the fourth sensors; sort the amplitudes corresponding to each third sensor from small to large, and generate a first sorting table; sort the amplitudes corresponding to each fourth sensor from large to small, and generate a second sorting table;

[0152] If the order of the first sorting table is consistent with the order of the sensor tags corresponding to the third sensors, and the order of the second sorting table is consistent with the order of the sensor tags corresponding to the fourth sensors, then a fault instruction is generated.

[0153] Further, the method for analyzing the fault type includes:

[0154] Take the characteristic parameters corresponding to each acoustic wave signal as analysis data, input the analysis data into the trained type analysis model, and output the corresponding type label. The type label is the digital label corresponding to the fault type, and obtain the fault type of the cable according to the type label;

[0155] The training process of the type analysis model includes:

[0156] Set different digital labels for different fault types and mark them as type labels; pre-collect g sets of analysis data, and set corresponding type labels for the g sets of analysis data, where g is an integer greater than 1;

[0157] Convert the analysis data and the corresponding type labels into a corresponding set of feature vectors; each set of feature vectors is used as the input of the type analysis model. The type analysis model outputs a set of predicted type labels corresponding to each set of analysis data, and takes the actual type label corresponding to each set of analysis data as the prediction target. The actual type label is the type label pre-collected corresponding to the analysis data; take minimizing the sum of the prediction errors of all analysis data as the training target; train the type analysis model until the sum of the prediction errors reaches convergence and then stop training; the type analysis model is a deep neural network model.

[0158] Further, the method for evaluating the propagation speed of the acoustic wave signal in the cable includes:

[0159] Preset a fault range and a normal propagation speed. The fault range includes a fault amplitude range and a fault frequency range; subtract the minimum value corresponding to the maximum value in the fault amplitude range to obtain an amplitude difference; subtract the minimum value corresponding to the maximum value in the fault frequency range to obtain a frequency difference; compare the amplitude corresponding to the first sensor with the amplitude corresponding to the second sensor, mark the acoustic wave sensor with the larger amplitude as the proximal sensor, and mark the acoustic wave sensor with the smaller amplitude as the distal sensor; obtain the amplitude corresponding to the proximal sensor and mark it as the proximal amplitude; obtain the frequency corresponding to the proximal sensor and mark it as the proximal frequency; subtract the maximum value in the fault amplitude range from the proximal amplitude to obtain a changed amplitude; subtract the maximum value in the fault frequency range from the proximal frequency to obtain a changed frequency;

[0160] Divide the changed amplitude by the amplitude difference to obtain an amplitude reduction coefficient; divide the changed frequency by the frequency difference to obtain a frequency reduction coefficient; use the amplitude reduction coefficient, the frequency reduction coefficient, and the predicted type label as test data, input the test data into the trained coefficient prediction model, predict the corresponding speed reduction coefficient, and mark it as the single-fault reduction coefficient; the training process of the coefficient prediction model is the same as that of the type analysis model, and both are deep neural network models;

[0161] Multiply the single-fault reduction coefficient by the normal propagation speed to obtain the single-fault propagation speed.

[0162] Further, respectively obtain the times when the proximal sensor and the distal sensor collect acoustic wave signals; mark the time when the proximal sensor collects the acoustic wave signal as the first time, and mark the time when the distal sensor collects the acoustic wave signal as the second time; subtract the first time from the second time to obtain a time difference; multiply the time difference by the single-fault propagation speed to obtain a distance difference; subtract the distance difference from the coupling interval and then divide by 2 to obtain a first distance; the first distance is the distance between the proximal sensor and the cable fault location; add the distance difference to the first distance to obtain a second distance, and the second distance is the distance between the distal sensor and the cable fault location; locate the cable fault location according to the first distance and the second distance.

[0163] Further, the method further includes: if multiple acoustic wave signals are collected by the same acoustic wave sensor, classify all the collected acoustic wave signals to obtain P signal sets, and each signal set corresponds to a cable fault location respectively;

[0164] The method for obtaining the P signal sets includes:

[0165] Obtain the acquisition time of each acoustic wave signal; sort each acquisition time from earliest to latest; obtain the sensor label corresponding to the acoustic wave sensor corresponding to each acoustic wave signal; replace each acquisition time with the corresponding sensor label to generate a third sorting table; analyze in ascending order according to the third sorting table, add the first two sensor labels in the third sorting table to a signal set, and starting from the third sensor label, classify each sensor label in turn according to the classification rules, and obtain a total of P signal sets, where there are no identical sensor labels in each signal set, and P is an integer greater than 1.

[0166] Further, the classification rules include:

[0167] Mark the currently classified sensor label as the current label;

[0168] If the current label is not adjacent to each sensor label in all signal sets, add the current label to a new signal set;

[0169] If the current label is adjacent to the sensor label in a signal set with only one sensor label, add the current label to the corresponding signal set;

[0170] If the current label is adjacent to the sensor labels in multiple signal sets respectively, obtain the acquisition times corresponding to the sensor labels adjacent to the current label and sort them, and add the current label to the signal set corresponding to the sensor label with the earliest acquisition time;

[0171] If the current label is adjacent to the sensor label in a signal set, add the current label to the corresponding signal set.

[0172] Further, according to the fault type of each cable fault, analyze the propagation speed of the acoustic wave signal in the cable;

[0173] According to the third sorting table, obtain the acoustic wave signals corresponding to the acquisition times of each sensor label in each signal set, extract the corresponding characteristic parameters, and then obtain the type label corresponding to each signal set respectively; according to the characteristic parameters corresponding to the first sensor label and the second sensor label in each signal set, calculate the amplitude reduction coefficient and frequency reduction coefficient corresponding to each signal set; use the amplitude reduction coefficient, frequency reduction coefficient and type label corresponding to each signal set as research data, input the research data into the trained coefficient prediction model, predict the corresponding speed reduction coefficient, and mark it as the multi-fault reduction coefficient; multiply the multi-fault reduction coefficient by the normal propagation speed to obtain the multi-fault propagation speed; the single-fault propagation speed is the propagation speed of the acoustic wave signal in the cable under the condition of multiple cable faults.

[0174] In addition, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be run by a processor to execute instructions corresponding to the method steps provided by the present application, such as: a method for acoustic diagnosis of power grid equipment faults. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0175] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0176] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should all be included in the protection scope of the present invention.

Claims

1. An acoustic wave diagnostic method for locating faults in power grid equipment, characterized in that: include: Collect the acoustic wave signal of the cable; De-noising the collected sound wave signal; Analyze the sound wave signal after denoising to determine whether a fault instruction is generated; If a fault instruction is generated, the corresponding fault type is analyzed; According to the fault type, evaluate the propagation speed of the acoustic signal in the cable; The cable fault is located based on the propagation speed of the acoustic wave signal.

2. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 1, characterized in that: The step of performing denoising processing on the collected sound wave signal comprises: Step 1: Set the wavelet basis function and the number of decomposition layers n; Step 2: Use wavelet basis function to perform i-layer wavelet decomposition on the acoustic signal, decomposing it into a group of approximate items and i groups of detail items, i∈[1,n]; Step 3: Calculate the signal-to-noise ratio of each group of detail items; Step 4: preset a signal-to-noise threshold, compare the signal-to-noise threshold with the signal-to-noise ratio of each group of approximate items and detail items, remove detail items with a signal-to-noise ratio less than the signal-to-noise threshold, and retain detail items with a signal-to-noise ratio greater than or equal to the signal-to-noise threshold; Step 5: Based on the i groups of detail items after the approximate items and threshold processing, the same wavelet basis function as in step 2 is used to reconstruct the signal; Step 6: Calculate the signal-to-noise ratio of the reconstructed signal and compare it with the preset denoising threshold; Step 7: If the signal ratio of the reconstructed signal is less than the denoising threshold, set i=i+1 and return to step 2 to replace the acoustic wave signal with the reconstructed signal; if the signal ratio of the reconstructed signal is greater than or equal to the denoising threshold, or i=n, ​​proceed to step 8; Step 8: Output the reconstructed signal, which is the sound wave signal after the denoising process is completed.

3. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 2, characterized in that: The acoustic wave signal is a mechanical wave caused by a tiny vibration of the material inside the cable; A coupling interval is preset, and m acoustic wave sensors are coupled to the surface of the cable according to the coupling interval; The method for determining whether a fault instruction is generated comprises: Extract the corresponding characteristic parameters from each acoustic wave signal, including amplitude, frequency and duration; set different digital labels for the m acoustic wave sensors in turn according to the current flow direction in the cable, and mark them as sensor labels, with the sensor label range being [1, m]; mark the acoustic wave sensor that collects the acoustic wave signal as the collection sensor; sort each amplitude from large to small, and mark the two collection sensors corresponding to the first two amplitudes as the first collection sensors; Determine whether the sensor tags of two first-collected sensors are adjacent; if not, do not generate a faulty sensor; if adjacent, mark the first-collected sensor with a larger sensor tag as the first sensor, and mark the first-collected sensor with a smaller sensor tag as the second sensor; mark the acquisition sensor with a larger sensor tag than the first sensor as the third sensor, and mark the acquisition sensor with a smaller sensor tag than the second sensor as the fourth sensor; sort the amplitude corresponding to each third sensor from small to large, and generate a first sorting table; sort the amplitude corresponding to each fourth sensor from large to small, and generate a second sorting table; If the order of the first sorting table is consistent with the order of the sensor tags corresponding to the third sensor, and the order of the second sorting table is consistent with the order of the sensor tags corresponding to the fourth sensor, a fault instruction is generated.

4. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 3, characterized in that: Methods for analyzing fault types include: The characteristic parameters corresponding to each acoustic wave signal are used as analysis data, and the analysis data is input into the trained type analysis model to output the corresponding type label. The type label is a digital label corresponding to the fault type, and the fault type of the cable is obtained according to the type label; The training process of the type analysis model includes: Different digital labels are set for different fault types and marked as type labels; g groups of analysis data are collected in advance, and corresponding type labels are set for the g groups of analysis data, where g is an integer greater than 1; The analysis data and the corresponding type labels are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of the type analysis model, and the type analysis model takes a set of predicted type labels corresponding to each set of analysis data as output, and takes the actual type labels corresponding to each set of analysis data as prediction targets, where the actual type labels are the pre-collected type labels corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training target; the type analysis model is trained until the sum of prediction errors converges; the type analysis model is stopped; the type analysis model is a deep neural network model.

5. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 4, characterized in that: The method for evaluating the propagation speed of an acoustic wave signal in a cable comprises: Preset the fault range and normal propagation speed, the fault range includes the fault amplitude range and the fault frequency range; subtract the corresponding minimum value from the maximum value in the fault amplitude range to obtain the amplitude difference; subtract the corresponding minimum value from the maximum value in the fault frequency range to obtain the frequency difference; compare the amplitude corresponding to the first sensor with the amplitude corresponding to the second sensor, mark the acoustic wave sensor corresponding to the large amplitude as the proximal sensor, and mark the acoustic wave sensor corresponding to the small amplitude as the distal sensor; obtain the amplitude corresponding to the proximal sensor and mark it as the proximal amplitude; obtain the frequency corresponding to the proximal sensor and mark it as the proximal frequency; subtract the maximum value in the fault amplitude range from the proximal amplitude to obtain the change amplitude; subtract the maximum value in the fault frequency range from the proximal frequency to obtain the change frequency; The amplitude reduction coefficient is obtained by dividing the change amplitude by the amplitude difference; the frequency reduction coefficient is obtained by dividing the change frequency by the frequency difference; the amplitude reduction coefficient, the frequency reduction coefficient and the predicted type label are used as test data, and the test data is input into the trained coefficient prediction model to predict the corresponding speed reduction coefficient and mark it as a single fault reduction coefficient; the training process of the coefficient prediction model is consistent with the training process of the type analysis model, and both are deep neural network models; Multiply the single fault reduction factor by the normal propagation speed to obtain the single fault propagation speed.

6. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 5, characterized in that: The time when the near-end sensor and the far-end sensor collect the acoustic signal is obtained respectively; the time when the near-end sensor collects the acoustic signal is marked as the first time, and the time when the far-end sensor collects the acoustic signal is marked as the second time; the first time is subtracted from the second time to obtain the time difference; the time difference is multiplied by the single fault propagation speed to obtain the distance difference; the distance difference is subtracted from the coupling interval, and then divided by 2 to obtain the first distance; the first distance is the distance between the near-end sensor and the cable fault; The first distance is added to the distance difference to obtain a second distance, where the second distance is the distance between the remote sensor and the cable fault location; the cable fault location is located according to the first distance and the second distance.

7. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 6, characterized in that: The method further includes: if the same acoustic wave sensor collects multiple acoustic wave signals, classifying all collected acoustic wave signals to obtain P signal sets, each signal set corresponding to a cable fault; The method of obtaining P signal sets includes: Acquire the acquisition time of each sound wave signal; sort each acquisition time from early to late; acquire the sensor tag corresponding to the sound wave sensor corresponding to each sound wave signal; replace each acquisition time with the corresponding sensor tag to generate a third sorting table; analyze according to the positive order of the third sorting table, add the first two sensor tags in the third sorting table to a signal set, and start from the third sensor tag, classify each sensor tag in turn according to the classification rule, and obtain a total of P signal sets, wherein there is no identical sensor tag in each signal set, and P is an integer greater than 1.

8. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 7, characterized in that: The classification rules include: Mark the sensor label currently being classified as the current label; If the current tag is not adjacent to any sensor tag in all signal sets, then add the current tag to a new signal set; If the current tag is adjacent to a sensor tag in a signal set where only one sensor tag exists, the current tag is added to the corresponding signal set; If the current tag is adjacent to sensor tags in multiple signal sets, the acquisition time corresponding to all sensor tags adjacent to the current tag is obtained, and they are sorted, and the current tag is added to the signal set corresponding to the sensor tag with the earliest acquisition time; If the current tag is adjacent to a sensor tag in a signal set, the current tag is added to the corresponding signal set.

9. The acoustic wave diagnosis method for locating faults in power grid equipment according to claim 8, characterized in that: According to the fault type of each cable fault, analyze the propagation speed of the acoustic wave signal in the cable; According to the third sorting table, the acoustic wave signal corresponding to the acquisition time of each sensor tag in each signal set is obtained, and the corresponding characteristic parameters are extracted, and then the type labels corresponding to each signal set are respectively obtained; according to the characteristic parameters corresponding to the first sensor tag and the second sensor tag in each signal set, the amplitude reduction coefficient and the frequency reduction coefficient corresponding to each signal set are calculated; the amplitude reduction coefficient, the frequency reduction coefficient and the type label corresponding to each signal set are used as research data, and the research data are input into the trained coefficient prediction model to predict the corresponding speed reduction coefficient, and marked as a multi-fault reduction coefficient; Multiply the multiple fault reduction factor by the normal propagation speed to obtain the multiple fault propagation speed; The single fault propagation velocity is the propagation velocity of the acoustic wave signal in the cable when there are multiple cable faults.

10. An acoustic wave diagnosis system for locating faults in power grid equipment, implementing an acoustic wave diagnosis method for locating faults in power grid equipment as claimed in any one of claims 1 to 9, characterized in that: include: An acoustic wave acquisition module, used for collecting acoustic wave signals of the cable; The sound wave processing module is used to perform denoising on the collected sound wave signals; A fault judgment module is used to analyze the sound wave signal after noise removal and determine whether to generate a fault instruction; A type analysis module, used to analyze the corresponding fault type if a fault instruction is generated; A velocity assessment module is used to assess the propagation velocity of the acoustic wave signal in the cable according to the fault type; The fault location module is used to locate the fault of the cable according to the propagation speed of the acoustic wave signal.

Images