Defect positioning method, system and equipment based on vehicle-mounted fusion detection and medium

Abstract

The invention discloses a defect positioning method, system and device based on vehicle-mounted fusion detection and a medium, and belongs to the technical field of distribution line fault detection. The method comprises the steps that a coordinate system is established according to the environment of a detected vehicle; the method comprises the following steps: collecting detection data through a vehicle-mounted probe, and preprocessing the detection data; when the vehicle is detected to be in a denial environment, optimizing the preprocessed detection data by constructing an acoustic reflection path model and a dynamic mask, and extracting infrared partial discharge features and acoustic imaging features; judging the discharge condition of the distribution line by using an acoustic infrared registration fusion mechanism; the infrared partial discharge features and the acoustic imaging features are fused, a fusion detection result is obtained, and a defect part is judged in combination with a line equipment drawing or a three-dimensional model; and combining the defect part with the coordinate system to obtain a defect coordinate, and rechecking the defect part by using the detection equipment according to the defect coordinate. According to the invention, the problem that accurate fault detection and fault positioning of the fault part of the distribution line cannot be realized in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of power distribution line fault detection technology, specifically to a defect location method, system, equipment, and medium based on vehicle-mounted fusion detection. Background Technology

[0002] Overhead distribution lines are a crucial component of urban power systems, and their safe and reliable operation directly impacts the stable development of the regional economy. Because these lines are constantly exposed to the natural environment, they not only bear the internal pressures of normal mechanical and electrical loads but also endure external threats such as pollution, lightning strikes, strong winds, landslides, subsidence, and bird damage. These factors accelerate the aging of various components along the lines, and if not detected and addressed promptly, can develop into various faults, posing a serious threat to the safety and stability of the power system. Therefore, the inspection of distribution lines is a fundamental task for effectively ensuring the safety of distribution lines and their equipment. By inspecting distribution lines, we can understand the operational status of the lines and changes in the surrounding environment, promptly identify equipment defects and potential hazards to line safety, and propose specific maintenance recommendations to eliminate defects in a timely manner, prevent accidents, or limit faults to a minimum, thereby ensuring the safety of transmission and distribution lines and the stability of the power system.

[0003] Currently, the inspection of partial discharge faults in overhead power lines mainly relies on handheld devices, engineering vehicle patrols, or detection devices mounted on drone platforms for line inspections. While these methods have achieved some success in practice, many shortcomings remain in the specific inspection process. There is currently no new and efficient technology or equipment that combines vehicle-mounted and drone-based inspections. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is: how to achieve accurate fault detection and fault location of faulty parts of power distribution lines through a vehicle-based fusion detection defect location method, while using drones to re-inspect the faulty parts.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a defect localization method based on vehicle-mounted fusion detection, comprising the following steps: establishing a coordinate system according to the environment of the detection vehicle; collecting detection data through a vehicle-mounted probe and preprocessing the detection data; when the detection vehicle is in a restricted environment, optimizing the preprocessed detection data by constructing an acoustic reflection path model and a dynamic mask, and extracting infrared partial discharge features and acoustic imaging features; using an acoustic-infrared registration fusion mechanism to determine the actual discharge situation of the power distribution line; fusing the infrared partial discharge features and acoustic imaging features to obtain a fusion detection result; determining the defect location based on the fusion detection result and in conjunction with line equipment drawings or a three-dimensional model; combining the defect location with the coordinate system to obtain defect coordinates; and verifying the defect part using detection equipment based on the defect coordinates.

[0007] As a preferred embodiment of the defect localization method based on vehicle-mounted fusion detection described in this invention, the step of establishing a coordinate system according to the route of the detection vehicle includes: the coordinate system comprising an open-air coordinate system and a rejected coordinate system; establishing the starting coordinates of the detection vehicle as the origin; when the detection vehicle is in the open-air environment, detecting the route of the detection vehicle using satellite positioning technology and establishing the open-air coordinate system; when the detection vehicle is in an environment rejected by satellite positioning technology, scanning using the radar pre-loaded on the detection vehicle to acquire point cloud data, and establishing the rejected coordinate system based on the point cloud data. The beneficial effects of this preferred embodiment are: by subdividing the open-air and rejected coordinate systems, it solves the positioning limitations of a single coordinate system in complex scenarios; by using the starting coordinates of the detection vehicle as the origin, it ensures a unified and continuous coordinate reference; the two coordinate systems can be used independently without fusion, simplifying the operation logic and reducing errors; and ultimately, it provides spatial support for subsequent detection data processing, feature extraction, and defect localization.

[0008] As a preferred embodiment of the vehicle-based fusion detection defect localization method of the present invention, the step of processing the preprocessed detection data when the detection vehicle is in a denied environment includes: constructing an acoustic reflection path model based on the denied environment parameters; filtering interference caused by reflection in the detection data based on the acoustic reflection path model; and constructing a dynamic mask in the detection data to shield high-temperature areas generated by unexpected situations. The beneficial effects of this preferred embodiment are: filtering reflection interference by constructing an acoustic reflection path model reduces the impact of the environment on acoustic data; simultaneously, shielding unexpected high-temperature areas with a dynamic mask ensures the reliability of infrared data; and optimizing the quality of detection data in a denied environment enhances anti-interference capabilities.

[0009] As a preferred embodiment of the defect localization method based on vehicle-mounted fusion detection according to the present invention, the steps of extracting infrared partial discharge features and acoustic imaging features include: the step of extracting infrared partial discharge features includes: identifying temperature anomaly regions from the processed detection data; analyzing the thermal gradient of the distribution parameters by calculating the thermal parameters of the temperature anomaly regions; and summarizing the thermal parameters and thermal gradients into the infrared partial discharge features; the step of extracting acoustic imaging features includes: identifying acoustic signals related to partial discharge from the processed detection data; obtaining the amplitude variation law of the acoustic signals by analyzing the vibration frequency of the acoustic signals; and summarizing the acoustic signals and amplitude variation law into the acoustic imaging features.

[0010] As a preferred embodiment of the defect localization method based on vehicle-mounted fusion detection described in this invention, the step of detecting the actual discharge situation of the power distribution line includes: extracting the corresponding three-dimensional coordinates from the acoustic imaging features, converting the three-dimensional coordinates into pixel coordinates on the infrared image in the detection data through coordinate transformation; and determining the existence of a discharge situation by judging its authenticity, including: judging that the discharge is real when the three-dimensional coordinates are within the conical range of the denial environment dome, the temperature at the pixel coordinates reaches a temperature threshold, and the diameter of the power distribution equipment corresponding to the pixel coordinates is less than a diameter threshold. The beneficial effect of this preferred embodiment is that it achieves spatial alignment of acoustic and infrared features through the conversion between three-dimensional coordinates and infrared pixel coordinates, and, combined with the verification of the conical range of the denial environment dome, temperature threshold, and equipment diameter threshold, it filters interference signals and reduces false judgments.

[0011] As a preferred embodiment of the vehicle-mounted fusion detection defect localization method of the present invention, the step of determining the defect location includes: processing the infrared partial discharge feature and the acoustic imaging feature through weighted fusion to obtain the fusion detection result; comparing the three-dimensional coordinates of the fusion detection result with the line equipment drawings or three-dimensional model; when the thermal parameters and acoustic signals of the fusion detection result exceed the risk threshold of the power distribution equipment, it is determined to be the defect location of the fusion detection result.

[0012] As a preferred embodiment of the vehicle-based fusion detection defect localization method of the present invention, the defect part is verified by using a detection device, including: combining the defect part with the coordinate system to obtain the defect coordinates; setting the flight path of the UAV according to the defect coordinates, and using the UAV to confirm the defect part.

[0013] This invention provides a vehicle-mounted fusion detection-based defect localization system.

[0014] To address the aforementioned technical problems, the present invention further provides the following technical solution: a vehicle-mounted fusion detection defect localization system, comprising: a coordinate system construction module, which establishes a coordinate system based on the travel route of the detection vehicle; a data processing module, which collects detection data through vehicle-mounted probes and preprocesses the detection data; a special data processing module, which, when the detection vehicle is in a denied environment, optimizes the preprocessed detection data by constructing an acoustic reflection path model and a dynamic mask, and extracts infrared partial discharge features and acoustic imaging features; an authenticity determination module, which uses an acoustic-infrared registration fusion mechanism to determine the actual discharge situation of the power distribution line; a defect location determination module, which fuses the infrared partial discharge features and acoustic imaging features to obtain a fusion detection result, and determines the defect location based on the fusion detection result and in conjunction with line equipment drawings or a three-dimensional model; and a verification module, which combines the defect location with the coordinate system to obtain defect coordinates, and verifies the defect part using detection equipment based on the defect coordinates.

[0015] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the defect localization method based on vehicle-mounted fusion detection.

[0016] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the defect localization method based on vehicle-mounted fusion detection.

[0017] The beneficial effects of this invention are as follows: By dividing the open-air coordinate system and the denied coordinate system, and combining the acoustic reflection path model constructed based on environmental parameters in the denied environment with a dynamic mask for unexpected high-temperature areas, it effectively solves the problems of satellite signal loss, detection data interference by reflection, and high-temperature interference in complex environments, ensuring that detection work can be carried out stably in both open-air and denied scenarios; by using the conversion of acoustic imaging feature three-dimensional coordinates to infrared image pixel coordinates, and combining the dome cone range of the denied environment, the temperature threshold at the pixel coordinates, and the diameter threshold of the power distribution equipment for verification, non-discharge interference signals are filtered out, the discharge false judgment rate is reduced, and reliable data support is provided for subsequent positioning; by obtaining the detection results through the weighted fusion of infrared partial discharge features and acoustic imaging features, and combining the line equipment drawings or three-dimensional models to locate the defect location, and then planning the UAV verification path according to the defect coordinates, it not only ensures the accurate positioning of the defect location, but also reduces manual intervention and improves detection efficiency; finally, it can quickly identify and efficiently verify the discharge defects of the power distribution line, providing accurate basis for line operation and maintenance, reducing the risk of failure, and ensuring the safe and stable operation of the power distribution line. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0019] Figure 1 The above is a flowchart of a vehicle-based fusion detection defect localization method provided in one embodiment of the present invention. Detailed Implementation

[0020] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0021] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a defect localization method based on vehicle-mounted fusion detection, including: S100: Establish a coordinate system based on the environment of the vehicle being inspected.

[0022] S200: Collects detection data through vehicle-mounted probes and preprocesses the detection data.

[0023] S300: When the detected vehicle is in a denied environment, the preprocessed detection data is optimized by constructing an acoustic reflection path model and a dynamic mask, and infrared partial discharge features and acoustic imaging features are extracted.

[0024] S400: Utilizes an acoustic-infrared registration and fusion mechanism to determine the actual discharge situation of power distribution lines.

[0025] S500: It fuses infrared partial discharge characteristics with acoustic imaging characteristics to obtain fused detection results. Based on the fused detection results, it determines the location of defects by combining the circuit equipment drawings or 3D models.

[0026] S600: Combine the defect location with the coordinate system to obtain the defect coordinates, and use the inspection equipment to verify the defect based on the defect coordinates.

[0027] It should be noted that overhead power distribution lines are prone to various faults due to long-term exposure to the natural environment. Existing technologies mainly rely on handheld devices, engineering vehicle inspections, or detection devices mounted on drone platforms for line inspections, but these are insufficient in terms of efficiency and reliability to support current needs.

[0028] Therefore, to address the shortcomings of existing technologies in terms of execution efficiency and reliability in supporting current needs, this invention constructs a vehicle-mounted fusion detection defect localization method through steps S100-S600. First, a coordinate system is established based on the environment of the vehicle being inspected. Second, detection data is collected using vehicle-mounted probes and preprocessed. When the vehicle is in a restricted environment, an acoustic reflection path model and a dynamic mask are constructed to optimize the preprocessed detection data, and infrared partial discharge features and acoustic imaging features are extracted. Next, an acoustic-infrared registration fusion mechanism is used to determine the actual discharge situation of the power distribution line. Then, the infrared partial discharge features and acoustic imaging features are fused to obtain the fused detection result. Based on the fused detection result, the defect location is determined in conjunction with the line equipment drawings or a 3D model. Finally, the defect location is combined with the coordinate system to obtain the defect coordinates. Based on the defect coordinates, the defect is verified using detection equipment.

[0029] Example 2, refer to Figure 1 This is the second embodiment of the present invention, which provides a defect localization method based on vehicle-mounted fusion detection.

[0030] In this embodiment of the invention, step S100, which establishes a coordinate system based on the environment of the detected vehicle, includes the following steps A1~A2: A1: Construct an open-air coordinate system in an open-air environment.

[0031] Specifically, the inspection vehicle is first parked in an area along the power line with a clear view and no satellite signal obstruction, and then stopped to ensure a stable satellite positioning signal. The starting coordinates of this parking position are obtained using vehicle-mounted satellite positioning technology, and these starting coordinates are set as the origin of the outdoor coordinate system. An outdoor coordinate system is established with the direction of travel of the inspection vehicle along the power line as the positive X-axis direction of the outdoor coordinate system.

[0032] A2: Construct a rejection coordinate system in scenarios where satellite positioning technology is rejected.

[0033] It should be noted that the environments where the monitoring is denied are mainly tunnels and canyons, and the characteristic is that satellite positioning technology cannot meet the requirements for real-time positioning of the monitoring vehicles.

[0034] Specifically, the system retrieves pre-stored 3D terrain data of tunnels and canyons along the power distribution line to be inspected from the vehicle control module, activates the radar pre-loaded on the vehicle, and performs a comprehensive scan of the current restricted environment, collecting and generating real-time point cloud data containing the environmental spatial structure and the relative position of the equipment. The real-time point cloud data is compared and matched with the feature point cloud in the pre-stored 3D terrain data, and the precise relative position of the vehicle in the current restricted environment is calculated through a point cloud registration algorithm. This relative position is set as the origin of the restricted coordinate system, and the restricted coordinate system is constructed with the direction of travel of the vehicle along the power distribution line as the positive X-axis.

[0035] Specifically, when GPS or BeiDou signals are lost, the relative position of the vehicle within the tunnel is determined by real-time matching of point cloud data with lidar. This is manifested in the following way: In the formula, Let be the rigid body transformation matrix. Let (x, y, z) be the coordinates of the i-th 3D point scanned by the vehicle-mounted LiDAR. The coordinates of the i-th pre-stored point in the DEM (Digital Elevation Model) are... This represents the total number of matching point pairs.

[0036] In this embodiment of the invention, in step S200, detection data is collected by the vehicle-mounted probe and the detection data is preprocessed.

[0037] It should be noted that the preprocessing of the test data includes preprocessing the data from both open-air and rejected environments.

[0038] Specifically, the vehicle-mounted probe synchronously collects infrared thermal imaging data and acoustic signal data of the power distribution equipment along the vehicle's route. During preprocessing, the infrared thermal imaging data is first denoised and temperature calibrated to retain the temperature distribution information of the equipment area. Then, the acoustic signal data is filtered and denoised to retain the acoustic feature information related to the discharge. Finally, the preprocessed data of the open-air environment that can be used for subsequent feature extraction is obtained.

[0039] In one possible implementation, the noise reduction technology for infrared thermal imaging data can also remove infrared image noise and retain the temperature distribution information of power distribution equipment through median filtering. In this solution, median filtering reads the original thermal imaging data collected by the vehicle-mounted infrared thermal imager, traverses the image pixels in a 3×3 sliding window, sorts the temperature values ​​of all pixels in each window, takes the median value to replace the original temperature value of the center pixel of the window, and finally retains the complete temperature distribution of the insulator and tower equipment.

[0040] In another possible implementation, the noise reduction technology for infrared thermal imaging data can also remove infrared image noise and retain the temperature distribution information of power distribution equipment through bilateral filtering. In this solution, bilateral filtering reads the original thermal imaging data collected by the vehicle-mounted infrared thermal imager, sets the spatial domain standard deviation and the value domain standard deviation, calculates the spatial distance weight and temperature difference weight of its neighboring pixels for each pixel, multiplies the two weighted values ​​to obtain the total weight, and replaces the original pixel value with the weighted average temperature value.

[0041] In this embodiment of the invention, when the detected vehicle is in a denied environment in step S300, the preprocessed detection data is optimized by constructing an acoustic reflection path model and a dynamic mask, and infrared partial discharge features and acoustic imaging features are extracted, including the following steps C1~C2: C1: When the detected vehicle is in a denied environment, the preprocessed detection data is optimized by constructing an acoustic reflection path model and a dynamic mask.

[0042] Specifically, optimizing the preprocessed detection data includes the following steps C11~C12: C11: Construct an acoustic reflection path model and filter out interference caused by reflection in the detection data based on the acoustic reflection path model.

[0043] Specifically, the concrete manifestation of constructing an acoustic reflection path model is as follows: In the formula, This is the time delay of the sound wave reflection echo; The speed of sound in air is approximately 343 m / s at room temperature; To detect the vertical distance between the vehicle and the tunnel wall; The pitch angle between the central axis of the microphone array and the horizontal plane; The sound source is specifically represented by the vertical distance between the discharge point of the insulator and the rail surface.

[0044] It should be noted that dynamic filtering in acoustic imaging The reflected signal.

[0045] In one possible implementation, the acoustic reflection path model can also dynamically filter reflected signals by modeling the amplitude attenuation characteristics of sound waves. By analyzing the amplitude difference between reflected waves and direct waves, interference is filtered out. First, the theoretical attenuation coefficient of the sound wave after reflection from the tunnel wall material is calculated based on the acoustic impedance of the tunnel wall material. The amplitude value of the sound wave signal is collected in real time, and the spatial directivity of the microphone array is combined to distinguish sound waves with different propagation paths. Signals with amplitudes lower than the theoretical amplitude threshold of the direct wave are identified as reflected waves and filtered out, while the discharge sound wave signal that conforms to the amplitude characteristics of the direct wave is retained, thereby achieving dynamic filtering of reflection interference.

[0046] In another possible implementation, the acoustic reflection path model can also dynamically filter reflected signals by modeling the geometric parameters of the tunnel cross-section. By extracting the cross-sectional parameters from the pre-stored three-dimensional terrain data of the tunnel based on the rectangular or circular cross-sectional features of the tunnel, and by calculating the theoretical propagation distance of the sound wave from the sound source through the tunnel top and side walls to the microphone based on the real-time position of the detected vehicle and the coordinates of the microphone array, combined with the geometric relationship of the cross-section, the theoretical propagation distance of the sound wave from the sound source through the tunnel top and side walls to the microphone is calculated. The actual sound wave propagation distance is compared with the theoretical reflection distance. If the difference is less than 5%, it is determined to be a reflected wave, which is then filtered out by threshold screening. Sound wave signals with propagation distances that conform to the characteristics of a direct path are retained, thereby achieving dynamic filtering of reflection interference.

[0047] C12: A dynamic mask is constructed based on the detection data to shield high-temperature areas caused by unexpected situations.

[0048] It should be noted that constructing a dynamic mask requires obtaining the real-time coordinates of the pantograph slider collected by the vehicle-mounted pantograph position sensor to determine the spatial position of the pantograph within the denied environment. In addition, it is also necessary to extract the initial position and temperature value of high-temperature areas from the infrared thermal imaging preprocessing data and screen out potential interference areas with temperatures more than 5°C higher than the ambient reference temperature.

[0049] Specifically, the dynamic mask manifests itself as follows: In the formula, Let be the mask value at coordinates (x, y) in the infrared image, where 0 indicates masking and 1 indicates preservation. Spatial coordinates of the pantograph sliding plate The projection position in the infrared image. The radius of the pantograph arc interference. The current operating speed of the train. This is an empirical coefficient of arc radius-speed based on actual measurements on heavy-haul railways.

[0050] In one possible implementation, the dynamic mask can also shield unexpected high-temperature areas by matching the motion trajectory of high-temperature areas. First, the coordinates of high-temperature areas in 10 consecutive frames of infrared thermal imaging preprocessing data are extracted, and the motion speed and direction of each area are calculated. Simultaneously, the real-time motion speed and direction of the detection vehicle are acquired. The motion parameters of the high-temperature area are compared with the motion parameters of the vehicle. If the speed deviation between the two is ≤5% and the directions are consistent, it is determined to be an interference high-temperature area such as pantograph arc moving with the vehicle. Using the center of this area as a reference, a mask is generated by expanding it by 1.2 times according to its actual size. The mask position is updated in real time with the movement of the vehicle to achieve dynamic shielding of unexpected high-temperature areas and retain the true temperature information of the power distribution equipment.

[0051] In another possible implementation, the dynamic mask can also shield unexpected high-temperature areas by matching equipment type with temperature thresholds. The normal operating temperature thresholds of various equipment on the power distribution line are retrieved from the pre-stored line equipment drawings. For high-temperature areas in the infrared thermal imaging preprocessing data, the type of equipment they cover is determined by combining image recognition technology. If the temperature of the high-temperature area exceeds the normal threshold of the corresponding equipment, and the distance between the area and the projection position of the pantograph slide is no more than 0.5 meters, a mask is generated based on the boundary of the area. The coordinates of the mask in the infrared image are adjusted in real time according to the vehicle's travel speed to ensure continuous shielding of unexpected high-temperature areas that do not conform to the normal heating characteristics of the equipment, thus preserving the true equipment temperature distribution information.

[0052] C2: Extraction of infrared partial discharge features and acoustic imaging features Specifically, when extracting infrared partial discharge features, the reserved areas with a mask value of 1 are screened from the infrared image, and temperature anomaly areas with temperatures 2°C or higher than the normal operating reference temperature of the equipment are identified. The thermal parameters of the area are calculated, including the highest temperature, average temperature and temperature gradient, and their spatial coordinates in the rejection coordinate system are correlated to form infrared partial discharge features.

[0053] Furthermore, when extracting acoustic imaging features, acoustic signals related to partial discharge of power distribution lines are identified from the acoustic signal data optimized by the acoustic reflection path model. Then, by analyzing the vibration frequency of the acoustic signal, the amplitude variation law of the acoustic signal over time is obtained. The identified acoustic signals and the obtained amplitude variation law are summarized to form acoustic imaging features.

[0054] In this embodiment of the invention, step S400 utilizes an acoustic-infrared registration and fusion mechanism to determine the actual discharge situation of the power distribution line, including the following steps D1~D2: D1: Extract the corresponding three-dimensional coordinates from the acoustic imaging features, and convert the three-dimensional coordinates into pixel coordinates on the infrared image in the detection data through coordinate transformation.

[0055] Specifically, coordinate transformation is performed through acoustic-infrared terrain registration and fusion, and the specific manifestation is as follows: In the formula, These are the pixel coordinates of the sound source in the infrared image. This is the intrinsic parameter matrix of the infrared camera. Let be a rotation matrix. It is a translation vector; , , ) represents the corresponding three-dimensional coordinates in the acoustic imaging feature.

[0056] D2: To determine the authenticity.

[0057] Specifically, only when the sound source is located within a ±15° cone angle of the tunnel arch, corresponding to the infrared coordinates... Temperature rise When the diameter of the temperature rise area is no more than 3 cm, it is determined to be a true discharge of the insulator.

[0058] In this embodiment of the invention, in step S500, the infrared partial discharge characteristics and acoustic imaging characteristics are fused to obtain the fused detection results. Based on the fused detection results, the defect location is determined in conjunction with the circuit equipment drawings or three-dimensional models.

[0059] Specifically, when fusing the infrared partial discharge features extracted in step C2 with the acoustic imaging features, a weighted fusion algorithm is used. The weights are dynamically assigned according to the reliability of the two types of features. In the infrared partial discharge features, the temperature gradient of the temperature anomaly area is significant, for example, the temperature difference between adjacent pixels is ≥1℃ and the duration is ≥3s, so its weight is set to 0.6. If the peak amplitude of the acoustic signal in the acoustic imaging features exceeds 3 times the background noise and the frequency is stable in the 50kHz-80kHz band, its weight is set to 0.4. The fusion detection result is obtained through weighted calculation.

[0060] Furthermore, the pre-stored power distribution line equipment drawings or 3D models are called up, and the suspected defect coordinates in the fusion detection results are matched with the equipment coordinates in the drawings or models. If the coordinates correspond to equipment components that are prone to discharge, and the structural position of the component in the drawings or models is consistent with the fusion features, then the component is determined to be a defect location, and the defect location is completed.

[0061] In one possible implementation, the fusion of infrared partial discharge characteristics and acoustic imaging characteristics can also be achieved through a decision tree fusion algorithm. The characteristics of both infrared partial discharge and acoustic imaging features are used as input features for the decision tree; decision nodes are constructed, and the defect confidence level is output through the decision tree path to form a fused detection result. Subsequently, the suspected coordinates in the result are compared with the three-dimensional model of the line equipment. If they correspond to insulators or surge arresters and their structural positions match, they are determined to be defective locations.

[0062] In another possible implementation, the fusion of infrared partial discharge features and acoustic imaging features can also be achieved through a neural network. By using infrared partial discharge features and acoustic imaging features as input layer parameters, a fully connected neural network with one hidden layer is constructed. The network is trained with historical discharge samples, and the weights are optimized through a backpropagation algorithm, enabling the network to learn the correlation between the two types of features. During real-time fusion, the extracted features are input into the trained network, and the output layer directly outputs the confidence level of the existence of the defect as the fusion detection result. Subsequently, the result is matched with the three-dimensional model of the line equipment. If the suspected coordinates correspond to easily discharged components and the features match, it is determined to be a defect location.

[0063] In this embodiment of the invention, in step S600, the defect location is combined with the coordinate system to obtain the defect coordinates, and the defect location is checked using a detection device based on the defect coordinates.

[0064] Specifically, when the defect is in an open-air environment, the three-dimensional coordinates of the defect are calculated according to the open-air coordinate system, combined with the relative position of the defective component in the line equipment drawings or three-dimensional model. If the defect is in a restricted environment, the three-dimensional coordinates of the defect are calculated according to the restricted coordinate system and the relative spatial parameters of the defective part in the restricted environment, thus completing the determination of the defect coordinates.

[0065] Furthermore, the flight path of the drone was confirmed based on the coordinates of the defect, and the defect area was hovered and scanned at ±2 meters along the contact line direction to avoid water seepage areas inside the tunnel.

[0066] In summary, this invention effectively solves the problems of satellite signal loss, data interference from reflection, and high temperature interference in complex environments by dividing the open-air coordinate system and the denied coordinate system, combined with an acoustic reflection path model constructed based on environmental parameters in the denied environment and a dynamic mask for unexpected high-temperature areas. This ensures stable progress of detection work in both open-air and denied scenarios. By converting the three-dimensional coordinates of acoustic imaging features to pixel coordinates of infrared images, and verifying the results by combining the cone-shaped range of the dome in the denied environment, the temperature threshold at the pixel coordinates, and the diameter threshold of the power distribution equipment, non-discharge interference signals are filtered out, reducing the discharge false alarm rate and providing reliable data support for subsequent positioning. The detection results are obtained by weighted fusion of infrared partial discharge features and acoustic imaging features. Combined with line equipment drawings or three-dimensional models, the defect location is located, and then a UAV verification path is planned according to the defect coordinates. This ensures accurate defect location and reduces manual intervention, improving detection efficiency. Ultimately, it can quickly identify and efficiently verify discharge defects in power distribution lines, providing accurate basis for line operation and maintenance, reducing the risk of fault occurrence, and ensuring the safe and stable operation of power distribution lines.

[0067] Example 3 is the third embodiment of the present invention, which provides a vehicle-mounted fusion detection defect localization system, including...

[0068] The coordinate system construction module establishes a coordinate system based on the travel route of the detected vehicle; The data processing module collects detection data through the vehicle-mounted probe and preprocesses the detection data; The special data processing module optimizes the pre-processed detection data and extracts infrared partial discharge features and acoustic imaging features when the detected vehicle is in a denied environment by constructing an acoustic reflection path model and a dynamic mask. The authenticity determination module uses an acoustic-infrared registration and fusion mechanism to determine the actual discharge situation of the power distribution line. The defect location determination module fuses infrared partial discharge features with acoustic imaging features to obtain fused detection results. Based on the fused detection results, the defect location is determined in conjunction with the circuit equipment drawings or 3D model. The verification module combines the defect location with the coordinate system to obtain the defect coordinates, and uses detection equipment to verify the defect based on the defect coordinates.

[0069] Example 4, the fourth embodiment of the present invention, differs from the previous three embodiments in that: if the function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0071] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0072] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination of all three. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for defect localization based on in-vehicle fusion detection, characterized in that: The utility model relates to a kind of power distribution line detection method and device, including, A coordinate system is established according to the environment of the detection vehicle; Detection data is collected by a vehicle-mounted probe, and the detection data is preprocessed; When the detection vehicle is in a forbidden environment, the preprocessed detection data is optimized by constructing an acoustic reflection path model and a dynamic mask, and infrared partial discharge features and acoustic imaging features are extracted; The real situation of power distribution line discharge is determined using acoustic-infrared registration fusion mechanism; The infrared partial discharge features and acoustic imaging features are fused to obtain a fusion detection result, and the defect position is determined based on the fusion detection result in combination with the line equipment drawing or three-dimensional model; The defect position is combined with the coordinate system to obtain defect coordinates, and the defect part is reviewed using detection equipment based on the defect coordinates.

2. The vehicle-based fusion-detected defect localization method of claim 1, wherein: The step of establishing a coordinate system according to the route of the detection vehicle includes: The coordinate system includes an open-air coordinate system and a forbidden coordinate system; The starting coordinates of the detection vehicle are established as the coordinate origin; When the detection vehicle is in the open-air environment, the route of the detection vehicle is detected by satellite positioning technology and the open-air coordinate system is established; When the detection vehicle is in an environment where satellite positioning technology is forbidden, the detection vehicle preloaded radar is scanned to obtain point cloud data, and the forbidden coordinate system is established based on the point cloud data.

3. The vehicle-based fusion-detected defect localization method of claim 2, wherein: When the detection vehicle is in a forbidden environment, the step of processing the preprocessed detection data includes: An acoustic reflection path model is constructed based on the forbidden environment parameters, and the detection data is filtered based on the acoustic reflection path model to remove interference caused by reflection; A dynamic mask is constructed based on the detection data, and the dynamic mask is used to shield high-temperature areas caused by unexpected situations.

4. The vehicle-based fusion-detected defect localization method of claim 3, wherein: The step of extracting infrared partial discharge features and acoustic imaging features includes: The step of extracting infrared partial discharge features includes: Temperature abnormal areas are identified from the processed detection data; The thermal parameters of the temperature abnormal areas are calculated to analyze the thermal gradient of the distribution parameters; The thermal parameters and thermal gradient are summarized as the infrared partial discharge features. The step of extracting acoustic imaging features includes: Sound signals related to partial discharge are identified from the processed detection data; The amplitude variation law of the sound signals is obtained by analyzing the vibration frequency of the sound signals; The sound signals and amplitude variation law are summarized as the acoustic imaging features.

5. The vehicle-based fusion-detected defect localization method of claim 4, wherein: The step of detecting the real situation of power distribution line discharge includes: The corresponding three-dimensional coordinates in the acoustic imaging features are extracted, and the three-dimensional coordinates are converted to pixel coordinates on the infrared image in the detection data through coordinate conversion; The existence of discharge is determined by judging the authenticity, including: When the three-dimensional coordinates are within the conical range of the vault in the forbidden environment, the temperature at the pixel coordinates reaches the temperature threshold, and the diameter of the power distribution line equipment corresponding to the pixel coordinates is less than the diameter threshold, it is determined that the discharge is real.

6. The vehicle-based fusion-detection based defect localization method of claim 5, wherein: The step of determining the defect position includes: The infrared arcing feature and the acoustic imaging feature are processed by weighted fusion to obtain a fusion detection result; According to the three-dimensional coordinates of the fusion detection result, comparison is made with the line equipment drawing or three-dimensional model; When the thermal parameter and the acoustic wave signal of the fusion detection result exceed the risk threshold of the power distribution equipment, it is determined that the defect site of the fusion detection result.

7. The vehicle-based fusion-detected defect localization method of claim 6, wherein: The defect part is reviewed using a detection device, including: The defect site is combined with the coordinate system to obtain the defect coordinates; According to the defect coordinates, the flight route of the unmanned aerial vehicle is set, and the defect site is confirmed by the unmanned aerial vehicle.

8. A vehicle-mounted fusion detection-based defect positioning system applying the vehicle-mounted fusion detection-based defect positioning method according to any one of claims 1 to 7, characterized in that, It includes: The coordinate system construction module establishes a coordinate system according to the travel route of the detection vehicle; The data processing module collects detection data through a vehicle-mounted probe and pre-processes the detection data; When the detection vehicle is in a forbidden environment, the special data processing module optimizes the pre-processed detection data by constructing an acoustic reflection path model and a dynamic mask, and extracts infrared arcing features and acoustic imaging features; The authenticity determination module determines the discharge authenticity of the power distribution line using an acoustic-infrared registration fusion mechanism; The defect site judgment module fuses the infrared arcing feature and the acoustic imaging feature to obtain a fusion detection result, and judges the defect site according to the fusion detection result in combination with the line equipment drawing or three-dimensional model; The review module combines the defect site with the coordinate system to obtain the defect coordinates, and reviews the defect part using the detection device according to the defect coordinates. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the vehicle-based fusion detection defect positioning method in any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the vehicle-based fusion detection defect positioning method in any one of claims 1 to 7.

Images