A method and system for predicting partial discharge faults in distribution networks

Abstract

Multiple embodiments of this specification relate to the field of electrical detection technology, and specifically to a method and system for predicting partial discharge faults in distribution networks. The method comprises the following steps: installing radio frequency current transformers on the neutral line and the main and branch lines in the distribution network to collect mutual inductance current data; installing a bracket on an inspection drone, and installing an ultrasonic acquisition device on the bracket; controlling the inspection drone to collect sound data from the marked locations on the transformer housings in the distribution network one by one along a preset first inspection path; after the inspection drone returns, installing an electromagnetic wave acquisition device on the bracket of the inspection drone; controlling the inspection drone to fly over electrical cabinets in the distribution network along a preset second inspection path to collect electromagnetic radiation data; and receiving and obtaining a prediction result for partial discharge faults based on the mutual inductance current data, sound data, and electromagnetic radiation data from a server.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the field of electrical detection technology, and specifically to a method and system for predicting partial discharge faults in a distribution network. Background Art

[0002] Partial discharge (PD) is a localized electrical discharge phenomenon that occurs within insulating materials. It can occur in solid, liquid, or gaseous dielectrics. In distribution networks, insulation degradation due to equipment aging, manufacturing defects, improper installation, or environmental factors can lead to PD. PD gradually damages the insulating material, causing degradation and ultimately leading to insulation breakdown and short circuit failures. Sustained PD shortens the service life of power equipment, increases the risk of sudden failures, and impacts power supply reliability. Equipment failures caused by PD can lead to power outages and increased repair costs. Due to the widespread distribution of distribution networks and the large number and variety of electrical equipment within them, PD detection is both costly and time-consuming, making it difficult to meet the safety requirements of distribution networks. Therefore, improved PD detection technology for distribution network PD is needed. Summary of the Invention

[0003] Multiple embodiments of this specification describe a method and system for predicting partial discharge faults in a distribution network.

[0004] In a first aspect, an embodiment of this specification provides a method for predicting partial discharge faults in a distribution network, comprising the steps of:

[0005] A radio frequency current transformer is set at the neutral line of the transformer in the distribution network and the trunk line and branch line in the distribution network to collect mutual induction current data. The radio frequency current transformer has a wireless communication module, and the wireless communication module establishes a communication connection with a preset server;

[0006] A bracket is installed on the inspection drone, and an ultrasonic acquisition device is installed on the bracket, wherein the ultrasonic acquisition device establishes a communication connection with the inspection drone, and the inspection drone establishes a communication connection with the server;

[0007] Controlling the inspection drone to collect sound data from marked locations on transformer housings in the distribution network along a preset first inspection route;

[0008] After the inspection drone returns, an electromagnetic wave collection device is installed on the bracket of the inspection drone, and the electromagnetic wave collection device establishes a communication connection with the inspection drone;

[0009] Controlling the inspection drone to fly over electrical cabinets in the distribution network along a preset second inspection path to collect electromagnetic radiation data;

[0010] The server receives and obtains a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data.

[0011] In a second aspect, an embodiment of this specification provides a system for predicting partial discharge faults in a distribution network, including:

[0012] A mutual inductance prediction module is configured to collect mutual inductance current data by installing a radio frequency current transformer at the neutral line of the transformer in the distribution network and at the trunk and branch lines in the distribution network. The radio frequency current transformer has a wireless communication module that establishes a communication connection with a preset server.

[0013] Ultrasonic prediction module, installing a bracket on the inspection drone, installing an ultrasonic acquisition device on the bracket, the ultrasonic acquisition device establishes a communication connection with the inspection drone, and the inspection drone establishes a communication connection with the server;

[0014] A first acquisition module controls the inspection drone to collect sound data from marked locations on transformer housings in the distribution network along a preset first inspection route;

[0015] An electromagnetic wave prediction module, after the inspection drone returns, an electromagnetic wave collection device is installed on the bracket of the inspection drone, and the electromagnetic wave collection device establishes a communication connection with the inspection drone;

[0016] A second collection module controls the inspection drone to fly over electrical cabinets in the distribution network along a preset second inspection path to collect electromagnetic radiation data;

[0017] The fault prediction module is configured so that the server receives and obtains a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data.

[0018] In a third aspect, embodiments of this specification provide an electronic device, including a processor and a memory;

[0019] The processor is connected to the memory;

[0020] The memory is used to store executable program code;

[0021] The processor reads the executable program code stored in the memory to run a program corresponding to the executable program code, so as to execute the method described in any one of the above aspects.

[0022] In a fourth aspect, an embodiment of this specification provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method described in any one of the above aspects is implemented.

[0023] In a fifth aspect, embodiments of this specification provide a computer program product, including a computer program, which implements the method described in any of the above aspects when executed by a processor.

[0024] The beneficial effects of the technical solutions provided by some embodiments of this specification include at least:

[0025] In multiple embodiments of this specification, a method and system for predicting partial discharge faults in a distribution network are provided. By setting up a radio frequency current transformer to collect mutual inductance current data and installing an ultrasonic acquisition device and an electromagnetic wave acquisition device on a patrol drone, it is possible to comprehensively monitor partial discharge conditions from multiple dimensions, thereby improving the accuracy and reliability of detection. Automatic inspections using patrol drones can quickly cover large areas, and are particularly suitable for widely distributed distribution network environments, significantly improving inspection efficiency. The use of automated inspection methods instead of traditional manual inspection methods not only reduces labor costs, but also reduces the safety risks brought about by staff contact with high-voltage equipment. Timely detection and handling of partial discharge problems helps prevent short-circuit faults caused by insulation breakdown, ensures the stable operation of the power system, and improves power supply reliability.

[0026] Other features and advantages of the various embodiments of this specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 This is a flow chart of a method for predicting partial discharge faults provided in an embodiment of this specification.

[0029] Figure 2 This is a schematic diagram of the inspection drone structure provided in the embodiments of this specification.

[0030] Figure 3 This is a flowchart of a first inspection path generation method provided in an embodiment of this specification.

[0031] Figure 4 This is a schematic diagram of the geometric marking points provided in the embodiments of this specification.

[0032] Figure 5 This is a schematic diagram of an electrical cabinet provided in an embodiment of this specification.

[0033] Figure 6This is a flow chart of a partial discharge fault diagnosis method provided in an embodiment of this specification.

[0034] Figure 7 This is a flow chart of a method for obtaining a delayed positioning reference table provided in an embodiment of this specification.

[0035] Figure 8 Schematic diagram of a partial discharge fault prediction system provided in an embodiment of this specification.

[0036] Figure 9 This is a schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0037] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification. However, the following embodiments are only preferred embodiments of this specification and are not exhaustive. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without making any creative work are all within the scope of protection of this specification.

[0038] Throughout this specification, the claims, and the accompanying drawings, the terms "first," "second," "third," and the like are used to distinguish between different items, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may include other steps or elements inherent to the process, method, product, or apparatus.

[0039] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, they should not be understood as limitations on this specification.

[0040] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.

[0041] Before introducing the technical solution in this specification, the application scenarios and related technologies of the technical solution are introduced.

[0042] The distribution network is a crucial component of the power system, responsible for distributing medium-voltage electricity from substations to individual users. It encompasses the entire power transmission path from the low-voltage side of the substation to the end user. The structure of a distribution network is typically planned based on the characteristics of the service area (e.g., urban, rural, or industrial) and load demand. It primarily consists of high-voltage / medium-voltage substations, medium-voltage distribution networks, distribution transformers, low-voltage distribution networks, and end users. The high-voltage / medium-voltage substation is the starting point of the distribution network, responsible for stepping down high-voltage electricity from the transmission grid to a medium-voltage level suitable for distribution. Overhead lines, using conductors supported by poles, are suitable for sparsely populated areas. Underground cables are often used in urban areas to avoid visual pollution and mitigate weather impacts. Switchyards / ring main units allow operators to control power flow for maintenance and fault isolation. Distribution transformers convert medium-voltage electricity to low-voltage electricity for use in homes and small businesses. Distribution transformers range from pole-mounted transformers mounted on utility poles to box-type transformers located above ground or in basements. The output of distribution transformers is directly supplied to consumers. The low voltage lines can also be connected to the internal wiring of users, including residential, commercial and industrial users, in the form of overhead lines or underground cables.

[0043] Partial discharge (PD) in transformers refers to a localized electrical breakdown phenomenon within a transformer's insulation system caused by the electric field strength exceeding the withstand voltage of the insulation material in a certain area. It typically occurs within or on the surface of the insulation material, as well as in air gaps between conductors. Although the energy released by a single PD is relatively small, frequent occurrences can cause gradual deterioration and damage to the insulation material, ultimately leading to failure of the entire insulation system and compromising the safe operation of the transformer. To effectively monitor and prevent PD-induced failures, modern power systems employ a variety of monitoring technologies to detect PD within transformers in real time. These technologies include, but are not limited to, Rogowski coils for acquiring electrical pulse signals generated by PD and ultrasonic sensors for capturing ultrasonic signals generated by the discharge.

[0044] Partial discharge in the electrical cabinet 30 refers to the electrical discharge phenomenon that occurs in localized areas within or on the surface of the insulating material. It typically occurs in high-voltage electrical equipment, such as switchgear and cable connectors. When the insulating material is damaged or aged, its insulation performance degrades, which may lead to the occurrence of partial discharge. Partial discharge in distribution cables is one of the important factors affecting the insulation performance of cables and their accessories, and may cause power system failures. In distribution cables, partial discharge typically occurs in areas where defects exist within or on the surface of the insulating material. These defects may be caused by flaws in the manufacturing process, improper installation, aging caused by long-term operation, or the influence of the external environment (such as moisture, pollution, etc.).

[0045] Due to the wide distribution range of the distribution network and the large number and variety of electrical equipment in the distribution network, the partial discharge detection of the distribution network is not only costly and time-consuming, but also difficult to meet the needs of ensuring the safety of the distribution network. For this reason, it is necessary to improve the detection technology of partial discharge faults in the distribution network. The four-rotor inspection drone 11 has the characteristics of simple structure, flexible control and vertical take-off and landing, and has been widely used in many fields. In the power industry, the specially designed four-rotor inspection drone 11 can resist the electromagnetic interference generated around the power equipment. In recent years, it has been increasingly used in distribution network inspection tasks, significantly improving the efficiency of distribution network inspection.

[0046] This manual first provides a method for predicting partial discharge faults in distribution networks. Figure 1 , including the steps of:

[0047] Step S101) A radio frequency current transformer is installed at the neutral line of the transformer in the distribution network and the trunk and branch lines in the distribution network to collect mutual inductance current data. The radio frequency current transformer has a wireless communication module, and the wireless communication module establishes a communication connection with a preset server.

[0048] By detecting high-frequency current pulses generated by partial discharge activity, the insulation condition of power equipment can be monitored to prevent potential failures. Installing a radio frequency current transformer on the neutral line of the substation transformer can help monitor partial discharges occurring within low-voltage equipment (such as cables, switchgear, etc.). Because partial discharges generate high-frequency currents throughout the circuit, and these currents flow through the neutral line, monitoring here can capture partial discharge signals at multiple points within the system. Installing radio frequency current transformers on the trunk and branch lines of the distribution network can more directly monitor partial discharge conditions within specific line sections. It can help locate the line where partial discharge occurs, especially in complex network structures, and provides guidance for the detection and troubleshooting of partial discharges.

[0049] Step S102) Install a bracket 12 on the inspection drone 11, install an ultrasonic acquisition device on the bracket 12, establish a communication connection between the ultrasonic acquisition device and the inspection drone 11, and establish a communication connection between the inspection drone 11 and the server. Figure 2 As a recommended approach, bracket 12 extends beyond the front end of inspection drone 11 to a preset length and is positioned below the rotor at a preset vertical distance. The ultrasonic acquisition device can be implemented using known technologies in the art. Electromagnetic radiation protection for inspection drones can be implemented using known technologies in the art.

[0050] The inspection drone 11 is provided with a bracket 12 , and a magnetic head 13 is installed at the end of the bracket 12 . The magnetic head 13 is a permanent magnet.

[0051] This specification provides an embodiment of an ultrasonic collection device, including a shell, several electromagnets, several magnets, a sound collector, an elastic bracket, a back magnet, a communication module, a controller and a battery. The back magnet is installed on the back of the shell for cooperating with the magnetic suction head 13. Several magnets are installed in the front of the shell. The electromagnet is installed in the shell and its position matches the magnet one by one. The sound collector is installed on the shell through the elastic bracket and extends relative to the magnet. The electromagnet, communication module and sound collector are all connected to the controller, and the battery powers the remaining components.

[0052] The process of the inspection drone 11 installing the ultrasonic collection device on the transformer casing 21 is as follows: the inspection drone 11 uses the magnetic suction head 13 to attract the back magnet of an ultrasonic collection device, and through communication, the electromagnet of the corresponding ultrasonic collection device generates a magnetic field in the opposite direction of the magnet pole. The two basically cancel each other out, so that the ultrasonic collection device can be removed from the steel plate. After removal, the electromagnet is powered off to save energy.

[0053] The inspection drone 11 then flies, carrying the ultrasonic acquisition device, to the first marking point 23. Through flight control and attitude control, the magnetic head 13 on the bracket 12 drives the ultrasonic acquisition device to align with the geometric marking point 22. However, it still needs to maintain a certain distance from the housing 21. The electromagnet is controlled to generate a magnetic field in the opposite direction of the magnet's poles, so that the two magnetic fields essentially cancel each other out. The inspection drone 11 is controlled to continue approaching the housing 21, aligning the ultrasonic acquisition device with the first marking point 23 and bringing it close to the transformer housing 21. The electromagnet is then de-energized, and the ultrasonic acquisition device is now firmly attached to the transformer housing 21 via the magnet, forming a coupling with the housing 21. The inspection drone 11 then retreats. The attraction between the magnetic head 13 and the back magnet is less than the attraction between the magnet and the transformer housing 21, causing the magnetic head 13 to detach from the back magnet. The inspection drone 11 also needs to synchronize time with the ultrasonic acquisition device. This synchronization is accomplished using techniques known in the art.

[0054] According to the positions of the geometric marking points 22, the inspection drone 11 continues to sequentially place multiple ultrasonic acquisition devices at the multiple geometric marking points 22. After collecting ultrasonic data for a preset time period, the inspection drone 11 removes the ultrasonic acquisition device from the transformer.

[0055] The steps for the inspection drone 11 to remove the ultrasonic acquisition device from the transformer casing 21 include: the inspection drone 11 uses flight control and attitude control to attract the magnetic head 13 to the back magnet of the ultrasonic acquisition device; then controls the electromagnet to generate a magnetic field in the opposite direction of the magnet's magnetic poles, so that the two magnetic fields basically cancel each other out; the inspection drone 11 flies backward, so that the magnetic head 13 drives the ultrasonic acquisition device out of the transformer casing 21, and then controls the electromagnet to cut off power; the inspection drone 11 uses flight control and attitude control to transport the ultrasonic acquisition device to a place close to the placed steel plate. When it is close to a certain distance from the steel plate, the electromagnet is controlled to generate a magnetic field in the opposite direction of the magnet's magnetic poles. After the position is completed, the electromagnet is cut off and the inspection drone 11 can fly away.

[0056] Step S103 ) Control the inspection drone 11 to collect sound data from the marked locations of the transformer housing 21 in the distribution network one by one along a preset first inspection route.

[0057] Please see the attached Figure 3 , the method for presetting the first inspection path includes:

[0058] Step S201) Mark the first marking point on the housing 21 of each transformer in each substation. Select a prominent and easily identifiable position on the transformer housing 21 as the first marking point, such as a fixed position on the front of the housing. Figure 4 Use physical markings, such as labels or spray paint, or virtual markings, such as using coordinate points based on a GIS map. Marking points should be unified to ensure that the first marking point on all transformers is in the same location.

[0059] Step S202) Obtain the coordinates of all substation transformers in the distribution network, as well as the coordinates of the first marker point. Collect the spatial location information of all substation transformers and their first marker points to provide basic data for subsequent path planning. Use a GPS device or GIS system to obtain the geographic coordinates of the substation transformers. For the first marker point, its specific coordinates can be calculated through relative position measurement or directly based on the offset of the transformer coordinates.

[0060] Step S203) Generate an approach path based on the coordinates of all first marked points. The approach path has a preset length and is oriented perpendicular to the transformer housing 21. A fixed approach path length, such as 0.5-0.8 meters, is preset. The approach path is oriented perpendicular to the surface of the transformer housing 21 to ensure that the inspection equipment can approach the target point at an optimal angle.

[0061] Step S204) Generate a first inspection path based on the free ends of all the approach paths. The free end of the approach path is the starting point of the path away from the transformer end. A path planning algorithm known in the art is used to connect all free ends into a shortest path or an optimal path.

[0062] The marking location includes a first marking point and a plurality of geometric marking points 22. The method for collecting sound data from the marking location of the transformer housing 21 in the distribution network includes:

[0063] Controlling the inspection drone 11 along the approach path to couple the ultrasonic acquisition device with the transformer housing 21 of the substation area;

[0064] After waiting for a preset time, the ultrasonic collection device is removed to obtain the sound data collected by the ultrasonic collection device.

[0065] Step S104 ) After the inspection drone 11 returns, an electromagnetic wave collection device is installed on the bracket 12 of the inspection drone 11 , and a communication connection is established between the electromagnetic wave collection device and the inspection drone 11 .

[0066] A short-term electrical breakdown phenomenon occurs when the electric field strength in a local area exceeds the breakdown strength of the medium in that area. Partial discharge is usually accompanied by a variety of physical phenomena, including electromagnetic radiation. The electromagnetic radiation generated by partial discharge has a certain directionality, which is related to the specific location of the discharge source and the surrounding geometric structure. Partial discharge is usually in the form of a transient pulse with an extremely short duration, generally in the nanosecond to microsecond level. These pulse signals contain a large amount of high-frequency components, which enables the electromagnetic radiation to radiate outward in the form of electromagnetic waves. The amplitude and frequency of partial discharge may vary with time and operating conditions. The frequency range of the electromagnetic waves generated is quite wide, extending from several thousand Hertz (kHz) to several gigahertz (GHz). The specific frequency band depends on many factors, including but not limited to the characteristics of the partial discharge source, the properties of the surrounding medium, and the discharge location. In the ultra-high frequency part, in the frequency band from several hundred MHz to several GHz, the electromagnetic waves generated by partial discharge have stronger penetration and less attenuation, which makes ultra-high frequency very suitable for detecting partial discharge phenomena in closed equipment, such as attached Figure 5 Detection of an electrical cabinet 30 in a power distribution network is shown. Typically, 2-4 antennas are used to capture electromagnetic radiation in different frequency bands.

[0067] Step S105 ) Control the inspection drone 11 to fly over the electrical cabinets 30 in the distribution network along a preset second inspection path to collect electromagnetic radiation data.

[0068] The method for presetting the second inspection path includes: obtaining the positions of all electrical cabinets 30 in the distribution network, setting a flyover point for each electrical cabinet 30, and obtaining the coordinates of the flyover point; and generating the second inspection path according to the coordinates of all the flyover points.

[0069] Step S106 ) The server receives and obtains a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data.

[0070] For details, please refer to the attached Figure 6 The method for obtaining a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data includes:

[0071] Step S301) using a preset feature extraction model to extract features of the mutual inductance current data;

[0072] Step S302) comparing the characteristics with pre-stored reference characteristics. If the comparison is consistent, it is determined that no partial discharge exists in the corresponding neutral line, trunk line or branch line. If the comparison is inconsistent, it is determined that partial discharge exists in the corresponding neutral line, trunk line or branch line.

[0073] Step S303) extracting features of the sound data and comparing them with reference sound features stored in the inspection drone 11. If the features match, it is determined that there is no partial discharge.

[0074] Step S304) When the comparison does not match, the inspection drone 11 is controlled to carry multiple ultrasonic acquisition devices one by one from the designated location and couple to multiple geometric marking points 22 pre-marked on the target transformer housing 21;

[0075] Step S305) After waiting for a preset time, the plurality of ultrasonic collection devices are removed one by one, and the sound data collected by the plurality of ultrasonic collection devices are acquired;

[0076] Step S306) obtaining a partial discharge position prediction result based on the plurality of sound data;

[0077] Step S307) extracting the frequency domain features of the electromagnetic radiation data to obtain frequency components;

[0078] Step S308 ) Compare the frequency composition with a preset reference frequency composition. If the comparison is consistent, it is determined that no partial discharge exists in the corresponding electrical cabinet 30 . If the comparison is inconsistent, it is determined that partial discharge exists in the corresponding electrical cabinet 30 .

[0079] Among them, the method for presetting the feature extraction model includes: reading multiple mutual inductance current data as sample data; establishing an autoencoding model and using the sample data to train the autoencoding model; and obtaining a feature extraction model based on the first half of the autoencoding model.

[0080] Collect a large amount of mutual inductance current data, including signals from both normal operation and partial discharge conditions, to provide sufficient data support for subsequent training. The data volume must be large enough to ensure that the model can capture the data's latent characteristics. An autoencoder is an unsupervised learning model that compresses input data into a low-dimensional space and reconstructs the original data from this space. By training the autoencoder, the latent characteristics of the mutual inductance current data can be learned.

[0081] The reference sound features stored in the inspection drone 11 include the frequency composition and amplitude ratio of the sound data collected at the same predetermined position corresponding to a plurality of transformer load intervals. The method of extracting the features of the sound data and comparing them with the reference sound features stored in the inspection drone 11 includes:

[0082] Selecting a frequency component of the sound data within a preset frequency range, and constructing a vector of the selected frequency component and its amplitude ratio, which is recorded as a first vector;

[0083] According to the frequency components included in the first vector, select corresponding frequency components and their amplitude proportions from the frequency components of the sound data corresponding to each transformer load interval to obtain multiple second vectors;

[0084] The similarity between the first vector and each of the second vectors is calculated separately. If any similarity value exceeds a preset reference similarity threshold, the comparison is determined to be a match; if no similarity value exceeds the preset reference similarity threshold, the comparison is determined to be a mismatch. By dividing the load intervals and comparing them using amplitude ratios, changes in the sound data caused by load differences can be offset, improving the accuracy of the comparison. Distance calculation between vectors can be performed using techniques disclosed in the art.

[0085] If a mismatch is detected, the inspection drone 11 carries each ultrasonic acquisition device, sets time synchronization, acquisition duration, and acquisition frequency, and couples the device to multiple pre-marked locations on the target transformer housing 21. After waiting for a preset period of time, the ultrasonic acquisition devices are removed one by one, and the sound data collected by the ultrasonic acquisition devices is acquired via the wireless communication connection.

[0086] The method for obtaining a location prediction result of a partial discharge based on a plurality of the sound data includes:

[0087] After aligning the plurality of sound data along the time axis, selecting the sound data corresponding to the designated marked point to calibrate a discharge pulse, and obtaining the sound propagation delay generated by the discharge pulse at the remaining calibrated points based on the remaining sound data to obtain a delay vector;

[0088] The position recognition result of the partial discharge is obtained according to the delay vector and the delay positioning reference table pre-stored in the inspection drone 11. The delay positioning reference table is an association table of the position recognition results of the same model and the same marked point obtained under laboratory conditions and the delay vector.

[0089] Ultrasonic waves from a partial discharge fault within the transformer are transmitted via a multi-media path to an ultrasonic sensor coupled to the wall. The ultrasonic sensors are mounted on the transformer oil tank housing 21 according to a specific geometric pattern. The time difference between the partial discharge time measured by each sensor and the reference time, as well as the coordinates of each sensor, is determined. The measured delay is used as the time from the discharge point to the sensor. Multiplying the delay by the equivalent wave velocity v yields the spatial position from the discharge point to each sensor. Substituting this into a nonlinear overdetermined equation yields the location of the discharge source. This involves establishing the spherical equation:

[0090] (x0-x i ) 2 +(y0-y i ) 2+ (z0-z i ) 2 =(vt i ) 2

[0091] Where i=1,2,3,…,n, represents the serial number of the ultrasonic acquisition device, x i 、y i 、z i represents the coordinates of the ith ultrasonic acquisition device, and x0, y0, and z0 represent the location of the partial discharge point. i represents the delay corresponding to each ultrasonic acquisition device, and v is the equivalent wave velocity. Because ultrasonic wave propagation within the transformer is not a constant but rather has a certain range, the wave velocity is considered a variable, namely the equivalent wave velocity v. Adding the coordinates of the desired partial discharge point, there are four variables. Using at least four ultrasonic acquisition devices, the coordinates of the partial discharge point and the equivalent wave velocity can be obtained. The spherical equation can be solved using the least squares method.

[0092] However, the least squares method is a nonlinear algorithm with long computation time and high memory usage. Furthermore, the results are significantly affected by initial values ​​and step size. Improper selection can result in an unsolvable system of equations or significant errors. Therefore, this embodiment provides a time-delayed location reference table that allows for rapid determination of the approximate location of a partial discharge (PD). Based on the structure at the approximate location, the structure with the PD fault can be identified for subsequent reference.

[0093] Please see the attached Figure 7 , the method for obtaining the delayed positioning reference table includes:

[0094] Step S401) Dividing the internal space of the selected model of transformer into a plurality of grids, grouping the grids capable of undergoing pulse discharge tests under laboratory conditions into a first group, and grouping the remaining grids into a second group;

[0095] Step S402) For the cells in the first group, pulse discharge tests are performed within the cell range under laboratory conditions;

[0096] Step S403) obtaining sound data collected by multiple ultrasonic collection devices coupled to the transformer housing 21 according to the marked points and aligning the sound data along the time axis;

[0097] Step S404) Identify the time corresponding to the discharge pulse in the sound data corresponding to the designated mark point, and use this time as a reference to obtain the sound propagation delay generated by the discharge pulse in the remaining ultrasonic acquisition devices;

[0098] Step S405) establishing a delay vector corresponding to the grid based on the sound propagation delays corresponding to all ultrasonic acquisition devices;

[0099] Step S406) traverse the grids in the second group and find the grid in the first group that is closest to the grid on the ultrasonic wave propagation path from the grid to each marked point, and record it as a reference grid;

[0100] Step S407) Calculating the sound propagation time between the grid and each reference grid, and obtaining the time it takes for the grid to reach the ultrasonic collection device at each marked point based on the sound propagation time and the time it takes for the reference grid to reach the ultrasonic collection device at the corresponding marked point during the pulse discharge test;

[0101] Step S408) Obtaining a delay vector corresponding to the grid according to the time it takes for the grid to reach the ultrasonic collection device at each marked point;

[0102] Step S409) Obtain the delay positioning reference table according to the delay vectors of all grids.

[0103] The delayed positioning reference table allows the inspection drone 11 to quickly determine the preliminary location of a partial discharge (PD) by looking up the table, eliminating the need for extensive calculations. The accuracy of the preliminary PD location is determined by the process of obtaining the delayed positioning reference table. The preliminary PD location is generally sufficient to assist in determining the structure where the PD fault has occurred.

[0104] The method for obtaining the position identification result of the partial discharge according to the delay vector and the delay positioning reference table pre-stored by the inspection drone 11 includes:

[0105] respectively calculating the similarity between the delay vector and each delay vector in the delay positioning reference table;

[0106] The grid corresponding to the delay vector with the highest similarity is used as the position recognition result of the partial discharge.

[0107] On the other hand, this manual provides a prediction system for partial discharge faults in distribution networks. Figure 8 ,include:

[0108] The mutual inductance prediction module 100 is configured to collect mutual inductance current data by installing a radio frequency current transformer at the neutral line of the transformer in the distribution network and at the trunk and branch lines in the distribution network. The radio frequency current transformer has a wireless communication module that establishes a communication connection with a preset server.

[0109] The ultrasonic prediction module 200 installs a bracket 12 on the inspection drone 11, and installs an ultrasonic acquisition device on the bracket 12, wherein the ultrasonic acquisition device establishes a communication connection with the inspection drone 11, and the inspection drone 11 establishes a communication connection with the server;

[0110] The first collection module 300 controls the inspection drone 11 to collect sound data from the marked locations of the transformer housings 21 in the distribution network along a preset first inspection route.

[0111] The electromagnetic wave prediction module 400 installs an electromagnetic wave collection device on the bracket 12 of the inspection drone 11 after the inspection drone 11 returns, and the electromagnetic wave collection device establishes a communication connection with the inspection drone 11;

[0112] The second collection module 500 controls the inspection drone 11 to fly over the electrical cabinets 30 in the distribution network along a preset second inspection path to collect electromagnetic radiation data;

[0113] The fault prediction module 600 receives and obtains a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data.

[0114] See also Figure 9 A schematic structural diagram of an electronic device provided in an embodiment of this specification is shown.

[0115] like Figure 9 As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105, and at least one communication bus 1102. The communication bus 1102 may be used to implement communication between the aforementioned components. The user interface 1103 may include buttons, and optionally may also include a standard wired interface or a wireless interface. The network interface 1104 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc. The processor 1101 may include one or more processing cores. The processor 1101 utilizes various interfaces and circuits to connect the various components within the entire electronic device 1100. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 1105 and accessing data stored in the memory 1105, it performs various functions of the routing device 1100 and processes data. Optionally, the processor 1101 may be implemented in hardware using at least one of a DSP, an FPGA, and a PLA. The processor 1101 may integrate one or a combination of a CPU, a GPU, and a modem. The CPU primarily processes the operating system, user interface, and applications; the GPU is responsible for rendering and drawing content displayed on the display; and the modem handles wireless communications.

[0116] It is understandable that the above-mentioned modem may not be integrated into the processor 1101, but may be implemented by a separate chip.

[0117] Memory 1105 may include either RAM or ROM. Optionally, memory 1105 may include non-transitory computer-readable media. Memory 1105 may be used to store instructions, programs, codes, code sets, or instruction sets. Memory 1105 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, sound playback function, image playback function, etc.), instructions for implementing the aforementioned method embodiments, etc.; the data storage area may store data related to the aforementioned method embodiments, etc. Memory 1105 may also optionally be at least one storage device located remotely from the aforementioned processor 1101. Memory 1105, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and application programs. Processor 1101 may be configured to invoke the application programs stored in memory 1105 and execute the methods described in the aforementioned embodiments.

[0118] The embodiments of this specification also provide a computer-readable storage medium having instructions stored therein that, when executed on a computer or processor, cause the computer or processor to perform the steps of the aforementioned embodiments. If the components of the aforementioned electronic device are implemented as software functional units and sold or used as independent products, they may be stored in the computer-readable storage medium.

[0119] The embodiments of this specification also provide a computer program product, including a computer program, which implements multiple steps in the above embodiments when executed by a processor.

[0120] In the absence of conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0121] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product comprises multiple computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium accessible by a computer or a data storage device such as a server or data center that integrates multiple available media. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state drive (SSD)).

[0122] When implemented via hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and realize the corresponding function. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit, whose logical function is determined by the user's device programming. Designers can "integrate" a digital system on a PLD through self-programming, eliminating the need for chip manufacturers to design and manufacture dedicated integrated circuit chips. Moreover, today, instead of manually manufacturing integrated circuit chips, this programming is often performed using "logic compiler" software. This is similar to the software compiler used in program development. Before compilation, the original code must also be written in a specific programming language, called a hardware description language (HDL). There are not just one HDL, but many. Those skilled in the art will also understand that simply by programming the method flow in one of the aforementioned hardware description languages ​​and programming it into the integrated circuit, a hardware circuit that implements the logical method flow can be easily obtained.

[0123] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A method for predicting partial discharge faults in a distribution network, characterized in that: Including steps: A radio frequency current transformer is set at the neutral line of the transformer in the distribution network and the trunk line and branch line in the distribution network to collect mutual induction current data. The radio frequency current transformer has a wireless communication module, and the wireless communication module establishes a communication connection with a preset server; A bracket is installed on the inspection drone, and an ultrasonic acquisition device is installed on the bracket, wherein the ultrasonic acquisition device establishes a communication connection with the inspection drone, and the inspection drone establishes a communication connection with the server; Controlling the inspection drone to collect sound data from marked locations on transformer housings in the distribution network along a preset first inspection route; After the inspection drone returns, an electromagnetic wave collection device is installed on the bracket of the inspection drone, and the electromagnetic wave collection device establishes a communication connection with the inspection drone; Controlling the inspection drone to fly over electrical cabinets in the distribution network along a preset second inspection path to collect electromagnetic radiation data; The server receives and obtains a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data; The method for obtaining a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data includes: Using a preset feature extraction model, extracting features of the mutual inductance current data; Comparing the features with pre-stored reference features, and if the comparison is consistent, determining that no partial discharge exists in the corresponding neutral line, trunk line, or branch line; and if the comparison is inconsistent, determining that partial discharge exists in the corresponding neutral line, trunk line, or branch line; Extracting features from the sound data and comparing them with reference sound features stored in the inspection drone, and determining that there is no partial discharge if the comparison matches; When the comparison does not match, the inspection drone is controlled to carry multiple ultrasonic acquisition devices one by one from the designated position and couple to multiple geometric marking points pre-marked on the target transformer housing; After waiting for a preset time, removing the plurality of ultrasonic collection devices one by one and acquiring the sound data collected by the plurality of ultrasonic collection devices; Obtaining a position prediction result of a partial discharge based on the plurality of sound data; Extracting frequency domain features of the electromagnetic radiation data to obtain frequency components; Comparing the frequency composition with a preset reference frequency composition, and when the comparison is consistent, determining that no partial discharge exists in the corresponding electrical cabinet; and when the comparison is inconsistent, determining that partial discharge exists in the corresponding electrical cabinet; The method for obtaining a location prediction result of a partial discharge based on a plurality of the sound data includes: After aligning the plurality of sound data along the time axis, selecting the sound data corresponding to the designated marked point to calibrate a discharge pulse, and obtaining the sound propagation delay generated by the discharge pulse at the remaining calibrated points based on the remaining sound data to obtain a delay vector; Obtaining a position recognition result of the partial discharge based on the delay vector and a delay positioning reference table pre-stored by the inspection drone, wherein the delay positioning reference table is an association table between position recognition results of the same model and the same marked point obtained under laboratory conditions and the delay vector; Methods for obtaining a delayed positioning reference table include: The internal space of the selected transformer model is divided into multiple grids, and the grids that can be used for pulse discharge tests under laboratory conditions are classified into the first group, and the remaining grids are classified into the second group; For the grids in the first group, pulse discharge tests are carried out in the grid range under laboratory conditions in turn; Acquire sound data collected by multiple ultrasonic collection devices coupled to the transformer housing according to the marked points and align them according to the time axis; Identify the time corresponding to the discharge pulse in the sound data corresponding to the designated mark point, and use this time as a reference to obtain the sound propagation delay generated by the discharge pulse in other ultrasonic acquisition devices; According to the sound propagation delays corresponding to all ultrasonic acquisition devices, a delay vector corresponding to the grid is established; Traversing the grids in the second group, finding the grid in the first group that is closest to the grid on the ultrasonic wave propagation path from the grid to each marked point, and recording it as the reference grid; Calculating the sound propagation time between the grid and each reference grid, and obtaining the time it takes for the grid to reach the ultrasonic collection device at each marked point based on the sound propagation time and the time it takes for the reference grid to reach the ultrasonic collection device at the corresponding marked point during the pulse discharge test; Obtaining a delay vector corresponding to the grid according to the time it takes for the grid to reach the ultrasonic acquisition device at each marked point; The delay positioning reference table is obtained according to the delay vectors of all grids.

2. A method for predicting partial discharge faults in a distribution network according to claim 1, characterized in that: The method of presetting the first inspection path includes: Mark the first marking point on the outer shell of each transformer in each substation; Obtaining the coordinates of all transformers in the distribution network and the coordinates of the first marking point; Generate an approach path according to the coordinates of all the first marking points, wherein the length of the approach path is a preset length and the direction is perpendicular to the transformer housing of the substation; A first inspection path is generated according to the free ends of all the approach paths.

3. The method for predicting partial discharge faults in a distribution network according to claim 2, wherein: The marking location includes a first marking point and a plurality of geometric marking points. The method for collecting sound data from the markings on the transformer casing in the distribution network includes: Controlling the inspection drone to couple the ultrasonic acquisition device with the transformer housing in the substation area along the approach path; After waiting for a preset time, the ultrasonic collection device is removed to obtain the sound data collected by the ultrasonic collection device.

4. The method for predicting partial discharge faults in a distribution network according to claim 1, wherein: The method of presetting the second inspection path includes: Obtain the locations of all electrical cabinets in the distribution network, set a flyover point for each electrical cabinet, and obtain the coordinates of the flyover point; A second inspection path is generated according to the coordinates of all the flying points.

5. The method for predicting partial discharge faults in a distribution network according to claim 1, wherein: Methods for presetting feature extraction models include: Reading a plurality of mutual inductance current data as sample data; Establishing an autoencoding model, and using the sample data to train the autoencoding model; According to the first half of the autoencoder model, a feature extraction model is obtained.

6. The method for predicting partial discharge faults in a distribution network according to claim 1, characterized in that: The reference sound features stored in the inspection drone include the frequency composition and amplitude ratio of the sound data collected at the same predetermined position corresponding to several transformer load intervals. The method of extracting the features of the sound data and comparing them with the reference sound features stored in the inspection drone includes: Selecting a frequency component of the sound data within a preset frequency range, and constructing a vector of the selected frequency component and its amplitude ratio, which is recorded as a first vector; According to the frequency components included in the first vector, select corresponding frequency components and their amplitude proportions from the frequency components of the sound data corresponding to each transformer load interval to obtain multiple second vectors; The similarity between the first vector and each of the second vectors is calculated respectively. When a similarity value is higher than a preset reference similarity threshold, the comparison is determined to be matched; when no similarity value is higher than the preset reference similarity threshold, the comparison is determined to be mismatched.

7. The method for predicting partial discharge faults in a distribution network according to claim 1, characterized in that: The method for obtaining a position identification result of a partial discharge according to the delay vector and the delay positioning reference table pre-stored by the inspection drone includes: respectively calculating the similarity between the delay vector and each delay vector in the delay positioning reference table; The grid corresponding to the delay vector with the highest similarity is used as the position recognition result of the partial discharge.

8. A prediction system for partial discharge faults in a distribution network, characterized in that: include: A mutual inductance prediction module is configured to collect mutual inductance current data by installing a radio frequency current transformer at the neutral line of the transformer in the distribution network and at the trunk and branch lines in the distribution network. The radio frequency current transformer has a wireless communication module that establishes a communication connection with a preset server. Ultrasonic prediction module, installing a bracket on the inspection drone, installing an ultrasonic acquisition device on the bracket, the ultrasonic acquisition device establishes a communication connection with the inspection drone, and the inspection drone establishes a communication connection with the server; A first acquisition module controls the inspection drone to collect sound data from marked locations on transformer housings in the distribution network along a preset first inspection route; An electromagnetic wave prediction module, after the inspection drone returns, an electromagnetic wave collection device is installed on the bracket of the inspection drone, and the electromagnetic wave collection device establishes a communication connection with the inspection drone; A second collection module controls the inspection drone to fly over electrical cabinets in the distribution network along a preset second inspection path to collect electromagnetic radiation data; a fault prediction module, wherein the server receives and obtains a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data; The method for obtaining a prediction result of a partial discharge fault based on the mutual inductance current data, the sound data, and the electromagnetic radiation data includes: Using a preset feature extraction model, extracting features of the mutual inductance current data; Comparing the features with pre-stored reference features, and if the comparison is consistent, determining that no partial discharge exists in the corresponding neutral line, trunk line, or branch line; and if the comparison is inconsistent, determining that partial discharge exists in the corresponding neutral line, trunk line, or branch line; Extracting features from the sound data and comparing them with reference sound features stored in the inspection drone, and determining that there is no partial discharge if the comparison matches; When the comparison does not match, the inspection drone is controlled to carry multiple ultrasonic acquisition devices one by one from the designated position and couple to multiple geometric marking points pre-marked on the target transformer housing; After waiting for a preset time, removing the plurality of ultrasonic collection devices one by one and acquiring the sound data collected by the plurality of ultrasonic collection devices; Obtaining a position prediction result of a partial discharge based on the plurality of sound data; Extracting frequency domain features of the electromagnetic radiation data to obtain frequency components; Comparing the frequency composition with a preset reference frequency composition, and when the comparison is consistent, determining that no partial discharge exists in the corresponding electrical cabinet; and when the comparison is inconsistent, determining that partial discharge exists in the corresponding electrical cabinet; The method for obtaining a location prediction result of a partial discharge based on a plurality of the sound data includes: After aligning the plurality of sound data along the time axis, selecting the sound data corresponding to the designated marked point to calibrate a discharge pulse, and obtaining the sound propagation delay generated by the discharge pulse at the remaining calibrated points based on the remaining sound data to obtain a delay vector; Obtaining a position recognition result of the partial discharge based on the delay vector and a delay positioning reference table pre-stored by the inspection drone, wherein the delay positioning reference table is an association table between position recognition results of the same model and the same marked point obtained under laboratory conditions and the delay vector; Methods for obtaining a delayed positioning reference table include: The internal space of the selected transformer model is divided into multiple grids, and the grids that can be used for pulse discharge tests under laboratory conditions are classified into the first group, and the remaining grids are classified into the second group; For the grids in the first group, pulse discharge tests are carried out in the grid range under laboratory conditions in turn; Acquire sound data collected by multiple ultrasonic collection devices coupled to the transformer housing according to the marked points and align them according to the time axis; Identify the time corresponding to the discharge pulse in the sound data corresponding to the designated mark point, and use this time as a reference to obtain the sound propagation delay generated by the discharge pulse in other ultrasonic acquisition devices; According to the sound propagation delays corresponding to all ultrasonic acquisition devices, a delay vector corresponding to the grid is established; Traversing the grids in the second group, finding the grid in the first group that is closest to the grid on the ultrasonic wave propagation path from the grid to each marked point, and recording it as the reference grid; Calculating the sound propagation time between the grid and each reference grid, and obtaining the time it takes for the grid to reach the ultrasonic collection device at each marked point based on the sound propagation time and the time it takes for the reference grid to reach the ultrasonic collection device at the corresponding marked point during the pulse discharge test; Obtaining a delay vector corresponding to the grid according to the time it takes for the grid to reach the ultrasonic acquisition device at each marked point; The delay positioning reference table is obtained according to the delay vectors of all grids.

9. An electronic device, characterized in that: including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

11. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images