CN114299359B - Method, equipment and storage medium for detecting power transmission line faults - Google Patents
Method, equipment and storage medium for detecting power transmission line faults Download PDFInfo
- Publication number
- CN114299359B CN114299359B CN202111577325.2A CN202111577325A CN114299359B CN 114299359 B CN114299359 B CN 114299359B CN 202111577325 A CN202111577325 A CN 202111577325A CN 114299359 B CN114299359 B CN 114299359B
- Authority
- CN
- China
- Prior art keywords
- fault
- fault classification
- determining
- transmission line
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000013145 classification model Methods 0.000 claims abstract description 50
- 238000007689 inspection Methods 0.000 claims abstract description 48
- 238000012549 training Methods 0.000 claims abstract description 34
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000013135 deep learning Methods 0.000 claims description 15
- 238000011176 pooling Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 abstract description 11
- 238000012423 maintenance Methods 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 9
- 238000012545 processing Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000011218 segmentation Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010223 real-time analysis Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002845 discoloration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The method, the device and the storage medium for detecting the power transmission line faults solve the technical problems that the manual detection cannot comprehensively and accurately judge potential safety hazards existing in the line and the overhaul cost is too high. The method comprises the following steps: acquiring a plurality of sample images based on the inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images; determining a fault sample image in the first multi-scale image dataset and performing fault marking to generate a second multi-scale image dataset; determining a first fault classification network, and training the first fault classification network based on a second multi-scale image data set to obtain a first fault classification model; when the power transmission line faults are detected, the images to be analyzed, which are acquired by the inspection equipment, are input into a first fault classification model, and alarming is carried out according to the fault classification result. The embodiment of the application can comprehensively and accurately judge the potential safety hazard of the circuit by the method, thereby greatly reducing the maintenance cost.
Description
Technical Field
The present application relates to the field of transmission line fault detection technologies, and in particular, to a method, an apparatus, and a storage medium for detecting a transmission line fault.
Background
Transmission lines typically span hundreds of kilometers and are exposed to the outside for a long period of time, suffer from various natural environments or man-made damages, etc., and jeopardize the safety of the entire grid system. Therefore, regular inspection of the transmission line is an effective measure for preventing transmission line faults.
In order to solve the problem that the transmission line is damaged by various natural environments or people, and the like, the safety of the whole power grid system is endangered, the current inspection mode still stays in manual inspection of the transmission line. Because the manual detection inevitably causes misjudgment or missed judgment, the potential safety hazard existing in the line cannot be comprehensively and accurately judged, and thus the maintenance cost is greatly increased.
Disclosure of Invention
The embodiment of the application provides a method, equipment and a storage medium for detecting a power transmission line fault, which solve the technical problem that the potential safety hazard existing in the line cannot be comprehensively and accurately judged by manual detection, thereby greatly increasing the maintenance cost.
In a first aspect, an embodiment of the present application provides a method for detecting a power transmission line fault, where the method includes: acquiring a plurality of sample images based on the inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images; determining a fault sample image in the first multi-scale image dataset and performing fault category marking on the fault sample image to generate a second multi-scale image dataset; determining a first fault classification network, and training the first fault classification network based on the second multi-scale image dataset to obtain a converged first fault classification model; under the condition that the power transmission line faults need to be detected, the image to be analyzed, which is acquired by the inspection equipment, is input into a first fault classification model, a fault classification result is determined, and an alarm is given according to the fault classification result.
The method for detecting the power transmission line faults comprises the steps of firstly constructing a plurality of sample images acquired by inspection equipment into a first multi-scale data set; then, fault category marking is carried out on fault sample images in the first multi-scale image data set, and the marked sample images form a second multi-scale image data set; training the first fault classification network based on the second multi-scale data set to obtain a first fault classification model, inputting an image to be analyzed into the first fault classification model when detecting the fault of the power transmission line, and determining a fault classification result so as to give an alarm. The embodiment of the application effectively avoids the situation that the manual detection can generate misjudgment or missed judgment, and can comprehensively and accurately judge the potential safety hazard of the circuit, thereby greatly reducing the overhaul cost.
In one implementation of the present application, determining a first failure classification network specifically includes: adding a preset space pyramid pooling layer before a full-connection layer of a preset deep learning network to generate a multi-scale image deep learning network; and replacing the standard convolution calculation mode in the multi-scale image deep learning network with a depth separable convolution calculation mode to obtain a first fault classification network.
According to the embodiment of the application, the spatial pyramid pooling layer is added before the full-connection layer of the preset depth network, so that the obtained first fault classification network can process sample images with any scale input into the first fault classification network; the standard convolution calculation mode in the depth network is replaced by the depth separable convolution calculation mode, so that the parameter scale in the obtained first fault classification network can be greatly reduced. The first fault classification network obtained by the method can process sample images of any scale for identification processing, and the required storage space and machine calculation force are also greatly reduced.
In one implementation of the present application, after obtaining the converged first fault classification model, the method further comprises: determining a plurality of first training parameters corresponding to the first fault classification model, and simplifying and adjusting the plurality of first training parameters to obtain a plurality of second training parameters; transmitting a plurality of second training parameters into a second fault classification network to determine a second fault classification model; and setting the second fault classification model in the inspection equipment so that the inspection equipment can analyze the acquired image to be analyzed in real time.
In one implementation of the application, before passing the number of second training parameters into the second failure classification network, the method comprises: determining an adapted second network computing logic based on the type of the inspection device; the first failure classification network is reconfigured based on the second network computing logic to obtain a second failure classification network.
In one implementation of the present application, constructing a first multi-scale image dataset based on a number of sample images, specifically comprises: dividing the plurality of sample images to determine a plurality of sample image groups; the sample image group comprises a preset number of sample images with different scales; a first multi-scale image dataset is generated based on the number of sample image groups.
In one implementation of the present application, fault marking a faulty sample image to generate a second multiscale image dataset specifically includes: determining a location of the fault in the sample image and marking the fault sample image based on the location; determining the fault type of the power transmission line in the fault sample image, and marking the fault sample image based on the fault type; a second multiscale image dataset is constructed and the marked fault sample image is added to the second multiscale image dataset.
In one implementation manner of the application, determining the fault type of the power transmission line in the fault sample image specifically comprises: determining a first fault class and a corresponding first fault class level of the power transmission line in the fault sample image; and determining the fault type of the power transmission line in the fault sample image based on the first fault class and the first fault class level.
In one implementation of the present application, the warning is performed according to the fault classification result, which specifically includes: determining a second fault class and a corresponding second fault class level in the fault classification result; and determining a preset alarming mode to alarm based on the second fault class and the second fault class level.
In a second aspect, an embodiment of the present application further provides an apparatus for detecting a power transmission line fault, where the apparatus includes: a processor; and a memory having executable code stored thereon which, when executed, causes the processor to perform a method as claimed in any one of claims 1 to 8.
In a third aspect, an embodiment of the present application further provides a non-volatile computer storage medium for detecting a power transmission line fault, storing computer executable instructions, where the computer executable instructions are configured to: acquiring a plurality of sample images based on the inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images; determining a fault sample image in the first multi-scale image dataset and fault marking the fault sample image to generate a second multi-scale image dataset; determining a first fault classification network, and training the first fault classification network based on the second multi-scale image dataset to obtain a converged first fault classification model; under the condition that the power transmission line faults need to be detected, the image to be analyzed, which is acquired by the inspection equipment, is input into a first fault classification model, a fault classification result is determined, and an alarm is given according to the fault classification result.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
Fig. 1 is a flowchart of a method for detecting a power transmission line fault according to an embodiment of the present application;
Fig. 2 is a schematic diagram of an internal structure of a device for detecting a fault of a power transmission line according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The embodiment of the application provides a method, equipment and a storage medium for detecting a power transmission line fault, which solve the technical problem that the potential safety hazard existing in the line cannot be comprehensively and accurately judged by manual detection, thereby greatly increasing the maintenance cost.
The following describes the technical scheme provided by the embodiment of the application in detail through the attached drawings.
Fig. 1 is a flowchart of a method for detecting a power transmission line fault according to an embodiment of the present application. As shown in fig. 1, the method for detecting a power transmission line fault provided by the embodiment of the application mainly includes the following steps:
step 101, acquiring a plurality of sample images based on inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images.
It should be noted that the inspection device of the present application includes, but is not limited to, an inspection helicopter or an inspection unmanned plane, and may be selected according to a specific application scenario. The sample image is based on the image acquisition equipment arranged on the inspection equipment, and the acquired image of the area to be inspected. It can be understood that the image of the area to be inspected includes the power transmission line to be inspected.
Further, the acquired plurality of sample images are constructed into a first multi-scale image dataset. The images contained in the first multi-scale image data set are a plurality of sample images, and each sample image can be of different scales; for example, the first multiscale image dataset contains both sample images having a scale of 300 x 300 and sample images having a scale of 400 x 400.
In one embodiment of the present application, the present application may further perform a segmentation process on the acquired sample image to determine a sample image group corresponding to the sample image; the sample image group comprises a preset number of sample images with different scales.
It should be noted that the following segmentation method may be used to segment the sample image: 1. dividing the sample image according to different scales aiming at a fault center point on the power transmission line; 2. and dividing the sample image according to different scales aiming at the corresponding region with the faults in the sample image. The first division method is used for dividing a certain fault feature on the sample image, and the second division method is used for dividing a plurality of fault features on the sample image.
For example, if a certain part of the power transmission line in the obtained sample image is damaged and other parts are intact, then only the damaged part is required to be extracted for information at the moment, the sample image is continuously segmented by taking the part with the fault as the center according to the sequence of the scale from small to large until the information affecting the damage of the part can be reflected, and the segmented sample images with different scales form a plurality of sample image groups.
Secondly, a plurality of transmission lines are damaged in different sizes in the acquired sample images, the damaged parts are scattered at different positions of the sample images, at this time, the sample images are required to be divided in different sizes according to the sizes of the damaged scales, and the divided sample images in different scales form a plurality of sample image groups.
It can be understood that the two modes can be combined, namely, under the condition that a plurality of transmission lines with different scales are damaged in an acquired sample image, a second segmentation mode is adopted for the sample image, and then a first segmentation mode is adopted for each segmented part.
According to the application, through the segmentation processing of the sample images, the sample image groups with different visual depths aiming at a certain sample image can be obtained, so that the subsequent training of the first fault classification network can be more accurate.
In one embodiment of the application, the sample image group can be complete sample images with different dimensions and different information, namely, the plurality of sample images acquired by the inspection equipment are not subjected to any segmentation processing, so that the plurality of sample image groups can be directly obtained, or part of sample images in the plurality of sample images acquired by the inspection equipment can be subjected to segmentation processing, other sample images keep the original dimensions, and thus, the segmented sample images with different dimensions and the complete sample images form a first multi-scale image data set.
Step 102, determining a failure sample image in the first multi-scale image dataset, and performing failure marking on the failure sample image to generate a second multi-scale image dataset.
In one embodiment of the application, marking the failure sample image in the first multi-scale dataset includes location marking of failures in the failure sample image and failure type marking of the failure sample image.
In one embodiment of the application, the fault type of the transmission line in the fault sample image is determined, in particular: determining a first fault class and a first fault class grade of the power transmission line in the fault sample image, wherein the first fault class comprises but is not limited to damage, pollution, discoloration and other problems, and the first fault class grade can be set according to the damage degree of the first fault to the power transmission line, and can be divided into a first grade, a second grade, a third grade and the like. Determining a fault type of the power transmission line in the fault sample image according to the first fault type and the first fault type level, for example: a broken first grade, a broken second grade, a broken third grade, a dirty first grade, a dirty second grade, a dirty third grade, a color-changing first grade, a color-changing second grade, a color-changing third grade, etc.
In one embodiment of the present application, after marking the failure sample image in the first multi-scale data set, a second multi-scale image data set is constructed (it is understood that the newly constructed set is an empty set), and then the failure sample image marked with the specific location of the failure and the type of the failure of the transmission line is added to the second multi-scale image data set.
Step 103, determining a first fault classification network, and training the first fault classification network based on a second multi-scale image dataset to obtain a converged first fault classification model;
In one embodiment of the application, in order to enable the depth network to process sample images of any scale for identification processing, the storage space and the machine calculation force are small, and the application constructs a first fault classification network. The first fault classification network construction includes: adding a spatial pyramid pooling layer before a full connection layer of a preset deep learning network and replacing a standard convolution calculation mode in the preset deep learning network with a depth separable convolution calculation mode.
In one embodiment of the application, since the full connection layer of the traditional deep learning network needs to have fixed input, and the spatial pyramid pooling layer can generate the input sample image with any scale into the output with fixed scale, the application adds a layer of spatial pyramid pooling layer before the full connection layer of the deep learning network, thereby enabling the deep learning network to realize the input of the multi-scale sample image.
In one embodiment of the present application, since the model trained by the server usually contains a large number of parameters, and a complex deep learning environment needs to be deployed, the model is limited by hardware facilities such as memory, and cannot be directly deployed to the device side. In order to consider that the fault classification model can be carried by using the subsequent inspection equipment such as an unmanned aerial vehicle system, the capability of real-time analysis is achieved. In the process of training the network, the depth separable convolution is utilized to replace the standard convolution, so that the parameter scale is reduced. The depth separable convolution divides the volume into the depth convolution and the point-by-point convolution, so that the calculation cost can be effectively reduced, and the memory occupation can be reduced.
In one embodiment of the application, after determining the first fault classification network, the first fault classification network is trained using the second multi-scale image dataset to obtain a converged first fault classification model.
And 104, under the condition that the power transmission line faults need to be detected, inputting the image to be analyzed obtained by the inspection equipment into a first fault classification model, determining a fault classification result, and performing pre-alarming according to the fault classification result.
In one embodiment of the application, after the first fault classification model is obtained, in the case that the transmission line needs to be detected, the inspection equipment acquires an image to be analyzed, and then returns to the first fault classification model arranged on the server to determine a fault classification result. It can be understood that the determined fault classification result includes a second fault class corresponding to the image to be analyzed and a corresponding second fault class level.
In one embodiment of the application, the alarm is performed according to a preset alarm mode after the fault classification result is determined. The alert mode may be selected based on a specific application environment, and the present application is not limited herein.
It can be understood that the inspection device acquires the image to be analyzed and then returns to the first fault classification model set in the server for detection, so that the effect of real-time detection cannot be achieved. Therefore, the embodiment of the application also sets a fault classification model in the inspection equipment. So that the inspection equipment can analyze the acquired image to be analyzed in real time.
The first fault classification model is a fault classification model adapted to the server, and if the fault classification model is to be installed in the inspection apparatus, the first fault classification model needs to be adjusted.
Specifically, first, a plurality of first training parameters corresponding to the first fault classification model are required to be adjusted; it can be understood that the plurality of first training parameters are parameters for training convergence. Because the unmanned aerial vehicle system needs a relatively simple fault classification model to realize real-time analysis of the power transmission line, a large number of parameters in the first fault classification model obtained by training the first fault classification network need to be simplified and adjusted so as to obtain a plurality of second training parameters. For example: some first training parameter is 0.15963456, and the corresponding second training parameter after simplification is 0.16.
In addition, a second fault classification network is also required to be constructed for adapting the inspection equipment. The construction of the second fault classification network firstly needs to determine an adaptive second network calculation logic based on the type of the inspection equipment, namely, one network calculation logic needs to be rewritten, and the first fault classification network is reconstructed based on the rewritten network calculation logic so as to obtain the second fault classification network.
After the second fault classification network is constructed, the obtained plurality of second training parameters are transmitted into the second fault classification network to determine a second fault classification model.
In one embodiment of the present application, after the second fault classification model is obtained, the fault classification model may be applied to the inspection apparatus, so that the inspection apparatus performs real-time analysis on the obtained image to be analyzed.
It should be noted that, after the second fault classification model is applied to the inspection device, each image to be analyzed acquired by the inspection device is not required to be uploaded to the server for detection. Also, since the second fault classification model has limited accuracy, it is preferable to use the first fault classification model in combination with the second fault classification model. And a certain image to be analyzed obtained by the inspection equipment is uploaded to the server for detection at regular intervals, so that the detection precision can be ensured, and the real-time performance of detection can be ensured.
Based on the same inventive concept, the embodiment of the application also provides equipment for detecting the power transmission line fault, and the internal structure of the equipment is shown in fig. 2.
Fig. 2 is a schematic diagram of an internal structure of a device for detecting a fault of a power transmission line according to an embodiment of the present application. As shown in fig. 2, the apparatus includes: a processor 201; the memory 202 has stored thereon executable instructions that, when executed, cause the processor 201 to perform a method of detecting a transmission line fault as described above.
In one embodiment of the present application, the processor 201 is configured to obtain a plurality of sample images based on the inspection device, and construct a first multi-scale image dataset based on the plurality of sample images; determining a fault sample image in the first multi-scale image dataset and fault marking the fault sample image to generate a second multi-scale image dataset; determining a first fault classification network, and training the first fault classification network based on the second multi-scale image dataset to obtain a converged first fault classification model; under the condition that the power transmission line faults need to be detected, the image to be analyzed, which is acquired by the inspection equipment, is input into a first fault classification model, a fault classification result is determined, and an alarm is given according to the fault classification result.
Some embodiments of the application provide a non-volatile computer storage medium corresponding to one of fig. 1 for detecting a transmission line fault, storing computer executable instructions configured to:
Acquiring a plurality of sample images based on the inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images;
determining a fault sample image in the first multi-scale image dataset and fault marking the fault sample image to generate a second multi-scale image dataset;
Determining a first fault classification network, and training the first fault classification network based on the second multi-scale image dataset to obtain a converged first fault classification model;
Under the condition that the power transmission line faults need to be detected, the image to be analyzed, which is acquired by the inspection equipment, is input into a first fault classification model, a fault classification result is determined, and an alarm is given according to the fault classification result.
The embodiments of the present application are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the internet of things device and the medium embodiment, since they are substantially similar to the method embodiment, the description is relatively simple, and the relevant points are referred to in the description of the method embodiment.
The system, the medium and the method provided by the embodiment of the application are in one-to-one correspondence, so that the system and the medium also have similar beneficial technical effects to the corresponding method, and the beneficial technical effects of the method are explained in detail above, so that the beneficial technical effects of the system and the medium are not repeated here.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may 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, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may 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, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (7)
1. A method of detecting a transmission line fault, the method comprising:
acquiring a plurality of sample images based on inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images;
Determining a fault sample image in the first multi-scale image dataset and fault marking the fault sample image to generate a second multi-scale image dataset;
Determining a first fault classification network, and training the first fault classification network based on the second multi-scale image dataset to obtain a converged first fault classification model;
Under the condition that the power transmission line faults need to be detected, inputting an image to be analyzed obtained by the inspection equipment into the first fault classification model, determining a fault classification result, and alarming according to the fault classification result;
Determining a first fault classification network, specifically comprising:
Adding a preset space pyramid pooling layer before a full-connection layer of a preset deep learning network to generate a multi-scale image deep learning network;
Replacing a standard convolution calculation mode in the multi-scale image deep learning network with a depth separable convolution calculation mode to obtain a first fault classification network;
after obtaining the converged first fault classification model, the method further comprises:
Determining a plurality of first training parameters corresponding to the first fault classification model, and simplifying and adjusting the plurality of first training parameters to obtain a plurality of second training parameters;
Transmitting the plurality of second training parameters into a second fault classification network to determine a second fault classification model;
setting the second fault classification model in the inspection equipment so that the inspection equipment can analyze the acquired image to be analyzed in real time;
Before passing the number of second training parameters into the second fault classification network, the method includes: determining an adapted second network computing logic based on the type of the inspection device;
And reconstructing the first fault classification network based on the second network computing logic to obtain a second fault classification network.
2. The method for detecting a transmission line fault according to claim 1, wherein constructing a first multiscale image dataset based on the plurality of sample images, in particular comprises:
dividing the plurality of sample images to determine a plurality of sample image groups; the sample image group comprises a preset number of sample images with different scales;
the first multi-scale image dataset is generated based on the number of sample image groups.
3. A method of detecting a transmission line fault according to claim 1, characterized in that fault marking the fault sample image to generate a second multiscale image dataset, in particular comprises:
determining a location of a fault in the image and marking the fault sample image based on the location; and
Determining a fault type of the power transmission line in the fault sample image, and marking the fault sample image based on the fault type;
a second multiscale image dataset is constructed and the marked faulty sample image is added to the second multiscale image dataset.
4. A method of detecting a transmission line fault as claimed in claim 3, wherein determining the type of fault of the transmission line in the fault sample image comprises:
Determining a first fault class and a corresponding first fault class level of the power transmission line in the fault sample image;
And determining the fault type of the power transmission line in the fault sample image based on the first fault class and the first fault class level.
5. The method for detecting a fault of a power transmission line according to claim 1, wherein the alarming is performed according to the fault classification result, specifically comprising:
determining a second fault class and a corresponding second fault class level in the fault classification result;
And determining a preset alarming mode to alarm based on the second fault class and the second fault class level.
6. An apparatus for detecting a transmission line fault, the apparatus comprising:
A processor;
And a memory having executable code stored thereon that, when executed, causes the processor to perform a method as claimed in any one of claims 1 to 5.
7. A non-volatile computer storage medium storing computer executable instructions for detecting a transmission line fault, the computer executable instructions configured to:
acquiring a plurality of sample images based on inspection equipment, and constructing a first multi-scale image data set based on the plurality of sample images;
Determining a fault sample image in the first multi-scale image dataset and fault marking the fault sample image to generate a second multi-scale image dataset;
Determining a first fault classification network, and training the first fault classification network based on the second multi-scale image dataset to obtain a converged first fault classification model;
Under the condition that the power transmission line faults need to be detected, inputting an image to be analyzed obtained by the inspection equipment into the first fault classification model, determining a fault classification result, and alarming according to the fault classification result;
Determining a first fault classification network, specifically comprising:
Adding a preset space pyramid pooling layer before a full-connection layer of a preset deep learning network to generate a multi-scale image deep learning network;
Replacing a standard convolution calculation mode in the multi-scale image deep learning network with a depth separable convolution calculation mode to obtain a first fault classification network;
after obtaining the converged first fault classification model, the method further comprises:
Determining a plurality of first training parameters corresponding to the first fault classification model, and simplifying and adjusting the plurality of first training parameters to obtain a plurality of second training parameters;
Transmitting the plurality of second training parameters into a second fault classification network to determine a second fault classification model;
setting the second fault classification model in the inspection equipment so that the inspection equipment can analyze the acquired image to be analyzed in real time;
Before passing the number of second training parameters into the second fault classification network, the method includes: determining an adapted second network computing logic based on the type of the inspection device;
And reconstructing the first fault classification network based on the second network computing logic to obtain a second fault classification network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111577325.2A CN114299359B (en) | 2021-12-22 | 2021-12-22 | Method, equipment and storage medium for detecting power transmission line faults |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111577325.2A CN114299359B (en) | 2021-12-22 | 2021-12-22 | Method, equipment and storage medium for detecting power transmission line faults |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114299359A CN114299359A (en) | 2022-04-08 |
| CN114299359B true CN114299359B (en) | 2024-05-10 |
Family
ID=80969182
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111577325.2A Active CN114299359B (en) | 2021-12-22 | 2021-12-22 | Method, equipment and storage medium for detecting power transmission line faults |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114299359B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114926723B (en) * | 2022-06-10 | 2024-05-28 | 国网江苏省电力有限公司电力科学研究院 | Method, terminal and storage medium for identifying and alarming and triggering peripheral interferents of power transmission line |
| CN115620496B (en) * | 2022-09-30 | 2024-04-12 | 北京国电通网络技术有限公司 | Fault alarm method, device, equipment and medium applied to power transmission line |
| CN116029537A (en) * | 2023-03-29 | 2023-04-28 | 中国南方电网有限责任公司超高压输电公司广州局 | Power transmission line inspection method, device, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109765462A (en) * | 2019-03-05 | 2019-05-17 | 国家电网有限公司 | Fault detection method, device and the terminal device of transmission line of electricity |
| CN110378221A (en) * | 2019-06-14 | 2019-10-25 | 安徽南瑞继远电网技术有限公司 | A kind of power grid wire clamp detects and defect identification method and device automatically |
| CN111582323A (en) * | 2020-04-17 | 2020-08-25 | 山东信通电子股份有限公司 | Power transmission line channel detection method, device and medium |
| WO2021056630A1 (en) * | 2019-09-26 | 2021-04-01 | 北京国网富达科技发展有限责任公司 | Defect detection method and device for transmission line tower structure |
| CN113569672A (en) * | 2021-07-16 | 2021-10-29 | 国网电力科学研究院有限公司 | Lightweight target detection and fault identification method, device and system |
-
2021
- 2021-12-22 CN CN202111577325.2A patent/CN114299359B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109765462A (en) * | 2019-03-05 | 2019-05-17 | 国家电网有限公司 | Fault detection method, device and the terminal device of transmission line of electricity |
| CN110378221A (en) * | 2019-06-14 | 2019-10-25 | 安徽南瑞继远电网技术有限公司 | A kind of power grid wire clamp detects and defect identification method and device automatically |
| WO2021056630A1 (en) * | 2019-09-26 | 2021-04-01 | 北京国网富达科技发展有限责任公司 | Defect detection method and device for transmission line tower structure |
| CN111582323A (en) * | 2020-04-17 | 2020-08-25 | 山东信通电子股份有限公司 | Power transmission line channel detection method, device and medium |
| CN113569672A (en) * | 2021-07-16 | 2021-10-29 | 国网电力科学研究院有限公司 | Lightweight target detection and fault identification method, device and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114299359A (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114299359B (en) | Method, equipment and storage medium for detecting power transmission line faults | |
| WO2021027213A1 (en) | Detection method and apparatus, electronic device and computer-readable medium | |
| US10771994B2 (en) | Method and test system for mobile network testing as well as a network testing system | |
| Rai et al. | Empirical terrain-based topographic modification factors for use in ground motion prediction | |
| CN114722965A (en) | Method and device for determining mechanical equipment fault | |
| CN113313716B (en) | Training method and device for automatic driving semantic segmentation model | |
| CN112926636A (en) | Method and device for detecting abnormal temperature of traction converter cabinet body | |
| CN115047064A (en) | Pipeline defect quantification method, processor and pipeline defect quantification device | |
| CN114547822A (en) | Pipeline safety early warning method, device and medium | |
| CN108875184B (en) | Shale organic carbon content prediction method and device based on digital outcrop model | |
| CN114965328A (en) | Ultraviolet multispectral high-voltage line fault detection and positioning method and device | |
| CN117521313A (en) | Method and device for leak detection of a pipeline network | |
| CN115311574B (en) | Building monitoring method, equipment and medium | |
| CN114062656B (en) | Geological disaster prediction method and system | |
| CN116484468A (en) | Risk assessment method, device and equipment for reservoir dam and storage medium | |
| CN105159826B (en) | A kind of method and apparatus of wrong sentence in positioning target program | |
| Bonada et al. | Practical-oriented pressure sensor placement for model-based leakage location in water distribution networks | |
| CN115797877A (en) | Intelligent monitoring method, system and medium for power transmission equipment | |
| CN115660917A (en) | Early warning method, device and medium for abnormal grain temperature in one-storey house | |
| CN111045027B (en) | Method and device for determining tree lodging defect and method and device for determining tree lodging area | |
| Mei | Crowdsensing-based monitoring of transportation infrastructure using moving vehicles | |
| CN116306223B (en) | Hydraulic engineering monitoring method, equipment and medium | |
| Dermadi | Tsunami Impact Prediction System Based on TsunAWI Inundation Data. | |
| CN117218488A (en) | Landslide recognition model training method, landslide recognition method and device | |
| EP4088218A1 (en) | Method and system for analyzing geo-spatial data to derive spatial object representations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |