Contact net pillar identification method and device

Abstract

The invention discloses a contact network pillar identification method and device, and the method comprises the steps: recognizing the position of a pillar in a pull-out value detection waveform according to a preset pull-out value feature of the pillar, and obtaining the rod position detection information; according to a preset anchor section joint pull-out value feature, the anchor section joint is identified in the pull-out value detection waveform, and anchor section detection information is obtained according to an anchor section joint identification result; according to the anchor section detection information and the rod position detection information, determining the number of struts and the sequence of the struts in the range of each anchor section; correcting the number of the struts and the sequence of the struts in the range of each anchor section by taking the machine account data of the power supply equipment as a reference to obtain a rod position identification result; and matching a pole number corresponding to the pole position identification result in the machine account data of the power supply equipment, and determining the pole position identification result and the matched pole number as an overhead line system strut identification result. The method can improve the pole position and pole number identification precision of the contact network strut, reduces the calculation cost, and has universality.

Description

Technical Field

[0001] This invention relates to the field of infrastructure inspection technology, and in particular to a method and apparatus for identifying overhead contact line supports. Background Technology

[0002] In the dynamic inspection and evaluation method for overhead contact lines, the span is used as the statistical step size for dynamic defect diagnosis. Therefore, accurately identifying the pole positions and defining the spans during dynamic inspection is crucial for subsequent quality evaluation.

[0003] Based on different implementation principles, pole position identification methods are currently mainly divided into three types: First, laser ranging technology is used. The laser ranging device is installed on the roof of the vehicle. During the detection process, high-frequency laser pulses are continuously emitted upwards perpendicular to the roof, and identification is achieved by measuring distance characteristics. This method requires high real-time processing, but is prone to misidentification in tunnels. Second, video camera technology is used. First, a camera with a fixed shooting angle is used to acquire photos of both sides of the track. Then, image processing technologies such as optical character recognition are used to identify the pole position. Pole number identification can be achieved simultaneously, but the computational load is large, and the pole number markings on conventional railways are not standardized, thus lacking universality. Third, identification is based on real-time detected contact network geometric parameter data, using the inflection point characteristics of the pulled-out value data. This method does not require additional hardware, requires less data, and has high computational efficiency, but it is prone to misidentification at curves and where inflection point characteristics are not obvious.

[0004] During the maintenance of overhead contact line equipment, railway workers locate defects by indexing pole numbers, making pole number identification a crucial task. Depending on the implementation method, there are three main pole number identification methods: First, combining pole position identification with manually initialized pole numbers for cumulative addition or subtraction. This method is simple and has been used for a long time, but it suffers from significant errors due to issues such as the accuracy of pole position identification and line section switching, and these errors tend to accumulate. Second, using digital image processing and neural network technologies to identify pole numbers from images. This method involves a large amount of computation and lacks universality due to the complex environment of the overhead contact line poles and the significant differences in pole number plate installation locations. Third, directly identifying pole numbers based on mileage records. This method is simple to implement, but it is greatly affected by the accuracy of the records and mileage deviations, easily leading to incorrect pole identification. Summary of the Invention

[0005] This invention provides a method for identifying overhead contact line supports, which improves the accuracy of identifying the position and number of overhead contact line supports, reduces computational costs, and has universality. The method includes:

[0006] Based on the preset pull-out value characteristics at the support, the position of the support is identified in the pull-out value detection waveform to obtain the pole position detection information; the pole position detection information includes: the number of supports, the order of the supports, and the mileage information of each support;

[0007] Based on the pre-defined pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform, and the anchor joint detection information is obtained based on the anchor joint identification results.

[0008] Based on the anchor section detection information and pole position detection information, the number of supports, the order of supports, and the mileage information of each support are obtained, and the number and order of supports within each anchor section are determined.

[0009] Based on the power supply equipment ledger data, the number and sequence of pillars within each anchor section are corrected according to the anchor section information in the power supply equipment ledger data. The mileage information of each pillar, as well as the corrected number and sequence of pillars, are determined as the pole position identification result.

[0010] Based on the mileage information of each support pole in the pole position identification results, the pole number corresponding to the pole position identification results is matched in the power supply equipment ledger data, and the pole position identification results and the matched pole number are determined as the contact network support pole identification results.

[0011] This invention also provides a contact wire support post identification device to improve the accuracy of contact wire support post position and post number identification, reduce computational costs, and has versatility. The device includes:

[0012] The pole position recognition module is used to: identify the position of the pole in the pull-out value detection waveform based on the preset pull-out value characteristics of the pole, and obtain pole position detection information; the pole position detection information includes: the number of poles, the order of the poles, and the mileage information of each pole;

[0013] The anchor segment identification module is used to: identify the anchor segment joint in the pull-out value detection waveform based on the preset pull-out value characteristics at the anchor segment joint, and obtain the anchor segment detection information based on the anchor segment joint identification result;

[0014] The pole position analysis module is used to determine the number and order of poles within each anchor section based on the anchor section detection information and the pole position detection information, including the number and order of poles and the mileage information of each pole.

[0015] The pole position correction module is used to: use the power supply equipment ledger data as a benchmark, and correct the number and order of the poles within each anchor section based on the anchor section information in the power supply equipment ledger data, and determine the mileage information of each pole and the corrected number and order of the poles as the pole position identification result.

[0016] The pole number matching module is used to: match the pole number corresponding to the pole position identification result in the power supply equipment ledger data based on the mileage information of each pole in the pole position identification result, and determine the pole position identification result and the matched pole number as the contact network pole identification result.

[0017] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described contact wire support identification method.

[0018] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described contact wire support identification method.

[0019] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described contact wire support identification method.

[0020] Compared with existing technologies that have high error rates or lack universality, this invention, in its embodiment, identifies the position of the support pillar in the pull-out value detection waveform based on preset pull-out value characteristics at the support pillar, obtaining pole position detection information including the number of pillars, the order of pillars, and the mileage information of each pillar; identifies the anchor joint in the pull-out value detection waveform based on preset anchor joint characteristics, and obtains anchor joint detection information based on the anchor joint identification result; determines the number and order of pillars within each anchor segment range based on the number of pillars, the order of pillars, and the mileage information of each pillar in the anchor segment detection information and pole position detection information; and, using power supply equipment ledger data as a benchmark, further determines the number and order of pillars within each anchor segment range based on the anchor segment information in the power supply equipment ledger data. The calibration process involves determining the pole position identification result by identifying the mileage information of each pole, as well as the corrected number and sequence of poles. During this calibration, the pole position detection information is divided into multiple intervals, treating each anchor segment as an independent section. This breaks down a complex problem into manageable smaller problems, allowing the calibration process to focus on individual anchor segments rather than the entire line, thus significantly improving the accuracy, efficiency, and reliability of the calibration and enhancing the precision of the contact network pole position identification result. Furthermore, the process matches the pole number corresponding to the pole position identification result with the power supply equipment ledger data, combining the pole position identification result and the matched pole number to determine the contact network pole identification result. This further improves the accuracy of contact network pole position and number identification, and features low computational cost and high versatility. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0022] Figure 1 This is a flowchart of the contact wire support identification method in an embodiment of the present invention;

[0023] Figure 2 This is a schematic diagram illustrating the training of the anchor segment recognition model in an embodiment of the present invention;

[0024] Figure 3 This is an example diagram showing the pole position and pole number identification results of the contact wire support identification method in this embodiment of the invention;

[0025] Figure 4 This is an example diagram showing the pole position and pole number identification results of the traditional contact wire support identification method;

[0026] Figure 5 This is a schematic diagram of the contact wire support identification device in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0028] Contact line supports are a crucial component of the contact line system, and their presence in contact line inspection data is primarily represented by two elements: support position and support number. The support position indicates the location of the support within the inspection data and serves as the basis for span division. The support number is the identification code for the contact line support, facilitating the management of the contact line equipment.

[0029] Existing methods for pole position and pole number identification typically suffer from limitations such as high false positive rates, large computational loads, lack of universality, or large cumulative errors. To address at least some of these problems, the inventors propose a contact wire support identification method based on power supply equipment ledgers and pull-out value parameters, which can improve the identification effect of contact wire pole positions and pole numbers.

[0030] Figure 1 This is a flowchart of the overhead contact line support identification method in an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0031] Step 101: Based on the preset pull-out value characteristics at the support, identify the position of the support in the pull-out value detection waveform to obtain the pole position detection information; the pole position detection information includes: the number of supports, the order of supports, and the mileage information of each support.

[0032] Step 102: Based on the preset pull-out value characteristics at the anchor joint, identify the anchor joint in the pull-out value detection waveform, and obtain the anchor joint detection information based on the anchor joint identification results;

[0033] Step 103: Based on the anchor section detection information and the pole position detection information, determine the number of poles and the order of poles within each anchor section.

[0034] Step 104: Based on the power supply equipment ledger data, and according to the anchor section information in the power supply equipment ledger data, correct the number and order of the pillars within each anchor section, and determine the mileage information of each pillar and the corrected number and order of the pillars as the pole position identification result.

[0035] Step 105: Based on the mileage information of each support in the pole position identification results, match the pole number corresponding to the pole position identification results in the power supply equipment ledger data, and determine the pole position identification results and the matched pole number as the contact network support identification results.

[0036] Compared with existing technologies that have high error rates or lack universality, this invention, in its embodiment, identifies the position of the support pillar in the pull-out value detection waveform based on preset pull-out value characteristics at the support pillar, obtaining pole position detection information including the number of pillars, the order of pillars, and the mileage information of each pillar; identifies the anchor joint in the pull-out value detection waveform based on preset anchor joint characteristics, and obtains anchor joint detection information based on the anchor joint identification result; determines the number and order of pillars within each anchor segment range based on the number of pillars, the order of pillars, and the mileage information of each pillar in the anchor segment detection information and pole position detection information; and, using power supply equipment ledger data as a benchmark, further determines the number and order of pillars within each anchor segment range based on the anchor segment information in the power supply equipment ledger data. The calibration process involves determining the pole position identification result by identifying the mileage information of each pole, as well as the corrected number and sequence of poles. During this calibration, the pole position detection information is divided into multiple intervals, treating each anchor segment as an independent section. This breaks down a complex problem into manageable smaller problems, allowing the calibration process to focus on individual anchor segments rather than the entire line, thus significantly improving the accuracy, efficiency, and reliability of the calibration and enhancing the precision of the contact network pole position identification result. Furthermore, the process matches the pole number corresponding to the pole position identification result with the power supply equipment ledger data, combining the pole position identification result and the matched pole number to determine the contact network pole identification result. This further improves the accuracy of contact network pole position and number identification, and features low computational cost and high versatility.

[0037] The alternating positive and negative inflection points of the pull-out values ​​at the support (pole position) are an inherent physical law of the overhead contact line suspension structure. Compared to pole position identification methods using laser ranging or video camera technology, the pole position identification method based on the inflection point characteristics of the pull-out value waveform does not rely on prior data and is insensitive to systematic deviations.

[0038] In this embodiment of the invention, the position of the support is identified in the pull-out value detection waveform based on the preset pull-out value characteristics at the support, and pole position detection information is obtained; the pole position detection information includes: the number of supports, the order of supports, and the mileage information of each support.

[0039] While pole position identification methods based on the inflection point features of the pull-out waveform are not affected by the accuracy of prior data, they also have inherent limitations: the reliability of pole position identification decreases when the waveform inflection point features are not obvious. To overcome this deficiency, this embodiment of the invention uses ledger data associated with anchor segment information in the pole position identification results to correct the pole position identification results.

[0040] First, it is necessary to obtain anchor segment detection information that originates from the same pull-out value detection waveform as the pole position detection information. In this embodiment of the invention, the anchor segment joint is identified in the pull-out value detection waveform based on the preset pull-out value characteristics at the anchor segment joint, and the anchor segment detection information is obtained based on the anchor segment joint identification result.

[0041] In the regular section, the pull-out value undergoes a simple alternating positive and negative jump at each positioning point (support), and the waveform is a regular "sawtooth" shape. However, in the anchor joint area (e.g., four-span anchor joint, five-span anchor joint), in order to achieve a smooth transition of the contact line, multiple positioning points (such as two transition posts and two center posts) will appear continuously in a relatively short interval. The pull-out value will undergo multiple directional changes and amplitude adjustments, thus forming a multi-peak "W" or "M" shaped waveform envelope, which is the most intuitive sign for identifying the anchor joint.

[0042] In one embodiment, the anchor joint is identified in the pull-out value detection waveform based on the preset pull-out value characteristics at the anchor joint, and the anchor joint identification result is obtained. The anchor joint identification result includes: the mileage information of the anchor joint, the type of the anchor joint, and the distribution of the anchor joint. Based on the mileage information, the type of the anchor joint, and the distribution of the anchor joint, the anchor detection information is obtained.

[0043] For example, two adjacent five-span non-insulated anchor joints are identified within a certain interval. The first anchor joint starts at kilometer K100+050, with center post kilometers of K100+150 and K100+200, and ends at k100+300. The second anchor joint starts at kilometer K101+800, with center post kilometers of K101+900 and K101+950, ​​and ends at k102+050. Based on the kilometers and distribution of these two anchor joints, a complete anchor segment can be defined between them. The starting boundary of the complete anchor segment is the ending kilometer K100+300 of the first anchor joint, and the ending boundary is the starting kilometer K101+800 of the second anchor joint. The length of the anchor segment is 1500 meters, and because both ends are non-insulated anchor joints, this complete anchor segment is identified as a non-insulated anchor segment. Thus, anchor segment detection information including the starting and ending kilometers, length, and type of the anchor segment is obtained.

[0044] While anchor joint positions can be directly identified using preset thresholds and logical rules, this approach lacks flexibility when dealing with complex or noisy waveforms. By using the pulled-out waveform as input, a deep learning model (classification or object detection model) can be trained, allowing the model to automatically learn the complex features of anchor joints and adapt to various complex line conditions.

[0045] In one embodiment, the pull-out value detection waveform is input into the anchor joint recognition model to obtain the anchor joint recognition result output by the anchor joint recognition model; wherein, the anchor joint recognition model is obtained by training a deep learning model using historical pull-out value detection waveforms labeled with anchor joint position information.

[0046] Figure 2 This is a schematic diagram illustrating the training of the anchor segment recognition model in an embodiment of the present invention. Figure 2 As shown, the pre-selected waveform data of the pull-out value at the anchor joint and the waveform data of the catenary pull-out value are input into the deep learning model together, so that the model learns the complex features of the waveform of the pull-out value at the anchor joint and outputs the identification result of the anchor joint.

[0047] For example, to ensure the generalization ability and robustness of the catenary joint recognition model, catenary anchor joint pull-out value data are collected under different line environments (high-speed, conventional, station, tunnel, etc.), different lighting conditions (daytime, nighttime, cloudy, etc.), and different weather conditions (rain, snow, fog, etc.). Deep learning technology is used to learn from the sample pull-out value data of the catenary anchor joint positions, automatically extracting the feature information of the anchor joints. Through continuous optimization of the algorithm and network, the real-time performance of the algorithm is improved, achieving accurate identification of anchor joints from the catenary pull-out value data.

[0048] In this embodiment of the invention, the number and order of pillars within each anchor section are determined based on the anchor section detection information and the pole position detection information, including the number of pillars, the pillar order, and the mileage information of each pillar.

[0049] The initially identified scattered support points are accurately classified and organized based on the macroscopic anchor segment structure, thus laying a solid foundation for subsequent fine correction based on anchor segments and ledgers. This solves the problem of uncertainty in the number and order of supports when relying solely on waveform recognition, and improves the reliability of pole position identification results.

[0050] In one embodiment, the start mileage information and end mileage information of each anchor segment in the anchor segment detection information are determined as the center column detection mileage information; based on the number of columns, column order and mileage information of each column in the pole position detection information, the number of columns and column order in the anchor segment detection information, and the center column detection mileage information, the number of columns and column order within the range of each anchor segment are determined.

[0051] The mileage of the supports recorded in the ledger may deviate from the actual location of the supports due to factors such as construction errors and historical changes. Within an anchor section, logical information such as the number and order of supports is extremely stable and accurate in the ledger. Once a support is recorded, there will generally be no omissions or over-recordings in subsequent ledger changes, and the order of the supports will not change.

[0052] Instead of relying on absolute mileage in the ledger, the inventors utilize reliable logical information (number and order of pillars) in the ledger, combined with the anchor segment information reflected by the pull-out value waveform and the correspondence in the ledger, to finely supplement and delete pole position detection information within each anchor segment range, in order to overcome the problem that pole position identification results based on pull-out value waveform may have "over-identification and under-identification".

[0053] In this embodiment of the invention, based on the power supply equipment ledger data, the number and order of pillars within each anchor section are corrected according to the anchor section information in the power supply equipment ledger data, and the mileage information of each pillar and the corrected number and order of pillars are determined as the pole position identification result.

[0054] For example, based on the anchor segment detection information (including anchor segment mileage, anchor segment type, anchor segment distribution statistics, etc.) obtained from the pull-out value detection waveform, and combined with the anchor segment information (including anchor segment mileage, anchor segment type, anchor segment distribution statistics, etc.) in the power supply equipment ledger data, the correspondence between the anchor segment joints in the pull-out value detection waveform and the anchor segment information in the ledger can be realized.

[0055] Currently, the functional requirements for high-speed pantograph-catenary integrated inspection devices stipulate that "the statistical data of inspection parameters should use pole number and mileage as a reference coordinate system." Railway workers locate defects using pole numbers during catenary equipment maintenance. Therefore, without accurate pole number identification, even the most precise inspection data cannot effectively guide on-site maintenance.

[0056] In this embodiment of the invention, based on the mileage information of each support in the pole position identification result, the pole number corresponding to the pole position identification result is matched in the power supply equipment ledger data, and the pole position identification result and the matched pole number are determined as the contact network support identification result.

[0057] Figure 3 This is an example diagram showing the pole position and pole number identification results of the contact wire support identification method in an embodiment of the present invention. Figure 3 As shown, the lower part of the horizontal axis represents mileage information, the upper part of the horizontal axis represents pole number, and the vertical axis records the guide height, pull-out value, and travel speed from top to bottom in the contact wire detection data. The vertical dashed line in the figure represents the support position identified based on the pull-out value detection waveform. Each support corresponds to unique mileage information and pole number.

[0058] Figure 4 This is an example diagram showing the pole position and pole number identification results of the traditional contact wire support identification method. Figure 4 Continue Figure 3 The meanings of the icons in the text, such as Figure 4 As shown, the support column identification results cannot establish a clear correspondence with the pole number information in the ledger.

[0059] This invention also provides a contact wire support identification device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to that of the contact wire support identification method, the implementation of this device can refer to the implementation of the contact wire support identification method; repeated details will not be elaborated further.

[0060] Figure 5 This is a schematic diagram of the overhead contact line support identification device in an embodiment of the present invention. Figure 5 As shown, the device includes:

[0061] The pole position recognition module 501 is used to: identify the position of the pole in the pull-out value detection waveform according to the preset pull-out value characteristics at the pole position, and obtain pole position detection information; the pole position detection information includes: the number of poles, the order of the poles, and the mileage information of each pole;

[0062] The anchor segment identification module 502 is used to: identify the anchor segment joint in the pull-out value detection waveform according to the preset pull-out value characteristics at the anchor segment joint, and obtain anchor segment detection information based on the anchor segment joint identification result;

[0063] The pole position analysis module 503 is used to: determine the number and order of poles within each anchor section based on the anchor section detection information and the pole position detection information, including the number and order of poles and the mileage information of each pole.

[0064] The pole position correction module 504 is used to: use the power supply equipment ledger data as a benchmark, and correct the number and order of the poles within each anchor section based on the anchor section information in the power supply equipment ledger data, and determine the mileage information of each pole and the corrected number and order of the poles as the pole position identification result.

[0065] The pole number matching module 505 is used to: match the pole number corresponding to the pole position identification result in the power supply equipment ledger data according to the mileage information of each pole in the pole position identification result, and determine the pole position identification result and the matched pole number as the contact network pole identification result.

[0066] In one embodiment, the anchor segment identification module 502 is specifically used for:

[0067] Based on the pre-defined pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform to obtain the anchor joint identification result; the anchor joint identification result includes: the mileage information of the anchor joint, the type of the anchor joint, and the distribution of the anchor joint;

[0068] Anchor segment detection information is obtained based on the mileage information, type, and distribution of the anchor segment joints.

[0069] In one embodiment, the anchor segment identification module 502 is specifically used for:

[0070] The pull-out value detection waveform is input into the anchor joint recognition model to obtain the anchor joint recognition result output by the anchor joint recognition model; the anchor joint recognition model is obtained by training the deep learning model using historical pull-out value detection waveforms labeled with anchor joint position information.

[0071] In one embodiment, the pole position analysis module 503 is specifically used for:

[0072] The start and end mileage information of each anchor segment in the anchor segment detection information is determined as the center column detection mileage information.

[0073] Based on the number and sequence of supports and the mileage information of each support in the pole position detection information, the number and sequence of supports in the anchor section detection information, and the mileage information of the center column detection, the number and sequence of supports within the range of each anchor section are determined.

[0074] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described contact wire support identification method.

[0075] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described contact wire support identification method.

[0076] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described contact wire support identification method.

[0077] Compared with existing technologies that have high error rates or lack universality, this invention, in its embodiment, identifies the position of the support pillar in the pull-out value detection waveform based on preset pull-out value characteristics at the support pillar, obtaining pole position detection information including the number of pillars, the order of pillars, and the mileage information of each pillar; identifies the anchor joint in the pull-out value detection waveform based on preset anchor joint characteristics, and obtains anchor joint detection information based on the anchor joint identification result; determines the number and order of pillars within each anchor segment range based on the number of pillars, the order of pillars, and the mileage information of each pillar in the anchor segment detection information and pole position detection information; and, using power supply equipment ledger data as a benchmark, further determines the number and order of pillars within each anchor segment range based on the anchor segment information in the power supply equipment ledger data. The calibration process involves determining the pole position identification result by identifying the mileage information of each pole, as well as the corrected number and sequence of poles. During this calibration, the pole position detection information is divided into multiple intervals, treating each anchor segment as an independent section. This breaks down a complex problem into manageable smaller problems, allowing the calibration process to focus on individual anchor segments rather than the entire line, thus significantly improving the accuracy, efficiency, and reliability of the calibration and enhancing the precision of the contact network pole position identification result. Furthermore, the process matches the pole number corresponding to the pole position identification result with the power supply equipment ledger data, combining the pole position identification result and the matched pole number to determine the contact network pole identification result. This further improves the accuracy of contact network pole position and number identification, and features low computational cost and high versatility.

[0078] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0080] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0082] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for identifying overhead contact line supports, characterized in that, include: Based on the preset pull-out value characteristics at the support, the position of the support is identified in the pull-out value detection waveform to obtain the pole position detection information; The pole position detection information includes: the number of poles, the order of the poles, and the mileage information of each pole; Based on the pre-defined pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform, and the anchor joint detection information is obtained based on the anchor joint identification results. Based on the anchor section detection information and pole position detection information, the number of supports, the order of supports, and the mileage information of each support are obtained, and the number and order of supports within each anchor section are determined. Based on the power supply equipment ledger data, the number and sequence of pillars within each anchor section are corrected according to the anchor section information in the power supply equipment ledger data. The mileage information of each pillar, as well as the corrected number and sequence of pillars, are determined as the pole position identification result. Based on the mileage information of each support pole in the pole position identification results, the pole number corresponding to the pole position identification results is matched in the power supply equipment ledger data, and the pole position identification results and the matched pole number are determined as the contact network support pole identification results.

2. The method as described in claim 1, characterized in that, Based on the preset pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform. Based on the anchor joint identification results, anchor detection information is obtained, including: Based on the pre-defined pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform to obtain the anchor joint identification result; the anchor joint identification result includes: the mileage information of the anchor joint, the type of the anchor joint, and the distribution of the anchor joint; Anchor segment detection information is obtained based on the mileage information, type, and distribution of the anchor segment joints.

3. The method as described in claim 2, characterized in that, Based on the preset pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform, and the anchor joint identification result is obtained, including: The pull-out value detection waveform is input into the anchor joint recognition model to obtain the anchor joint recognition result output by the anchor joint recognition model; the anchor joint recognition model is obtained by training the deep learning model using historical pull-out value detection waveforms labeled with anchor joint position information.

4. The method as described in claim 1, characterized in that, Based on the anchor section detection information and pole position detection information, including the number and sequence of supports, as well as the mileage information of each support, the number and sequence of supports within each anchor section are determined, including: The start and end mileage information of each anchor segment in the anchor segment detection information is determined as the center column detection mileage information. Based on the number and sequence of supports and the mileage information of each support in the pole position detection information, the number and sequence of supports in the anchor section detection information, and the mileage information of the center column detection, the number and sequence of supports within the range of each anchor section are determined.

5. A contact wire support identification device, characterized in that, include: The pole position recognition module is used to: identify the position of the pole in the pull-out value detection waveform based on the preset pull-out value characteristics of the pole, and obtain pole position detection information; the pole position detection information includes: the number of poles, the order of the poles, and the mileage information of each pole; The anchor segment identification module is used to: identify the anchor segment joint in the pull-out value detection waveform based on the preset pull-out value characteristics at the anchor segment joint, and obtain the anchor segment detection information based on the anchor segment joint identification result; The pole position analysis module is used to determine the number and order of poles within each anchor section based on the anchor section detection information and the pole position detection information, including the number and order of poles and the mileage information of each pole. The pole position correction module is used to: use the power supply equipment ledger data as a benchmark, and correct the number and order of the poles within each anchor section based on the anchor section information in the power supply equipment ledger data, and determine the mileage information of each pole and the corrected number and order of the poles as the pole position identification result. The pole number matching module is used to: match the pole number corresponding to the pole position identification result in the power supply equipment ledger data based on the mileage information of each pole in the pole position identification result, and determine the pole position identification result and the matched pole number as the contact network pole identification result.

6. The apparatus as claimed in claim 5, characterized in that, The anchor segment identification module is specifically used for: Based on the pre-defined pull-out value characteristics at the anchor joint, the anchor joint is identified in the pull-out value detection waveform, and the anchor joint identification result is obtained. The anchor joint identification results include: anchor joint mileage information, anchor joint type, and anchor joint distribution; Anchor segment detection information is obtained based on the mileage information, type, and distribution of the anchor segment joints.

7. The apparatus as claimed in claim 6, characterized in that, The anchor segment identification module is specifically used for: The pull-out value detection waveform is input into the anchor joint recognition model to obtain the anchor joint recognition result output by the anchor joint recognition model; the anchor joint recognition model is obtained by training the deep learning model using historical pull-out value detection waveforms labeled with anchor joint position information.

8. The apparatus as claimed in claim 5, characterized in that, The pole position analysis module is specifically used for: The start and end mileage information of each anchor segment in the anchor segment detection information is determined as the center column detection mileage information. Based on the number and sequence of supports and the mileage information of each support in the pole position detection information, the number and sequence of supports in the anchor section detection information, and the mileage information of the center column detection, the number and sequence of supports within the range of each anchor section are determined.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 4.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 4.

Images