CN117649374A - Intelligent detection method and device for pin loss of transformer substation - Google Patents
Intelligent detection method and device for pin loss of transformer substation Download PDFInfo
- Publication number
- CN117649374A CN117649374A CN202311367918.5A CN202311367918A CN117649374A CN 117649374 A CN117649374 A CN 117649374A CN 202311367918 A CN202311367918 A CN 202311367918A CN 117649374 A CN117649374 A CN 117649374A
- Authority
- CN
- China
- Prior art keywords
- pin
- image
- different
- images
- training sample
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 65
- 238000012549 training Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000007689 inspection Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Abstract
The invention provides an intelligent detection method and device for pin loss of a transformer substation, which are used for acquiring image data of pins at multiple time points, multiple angles and multiple shooting distances; preprocessing image data; marking the pin positions in each preprocessed image to be used as a training sample data set for pin detection; training the built pin detection network model by using a training sample data set, acquiring feature images of different receptive fields of a training sample image by using downsampling operation with different multiples of the pin detection network model, fusing the feature images, retaining the features of the pin under the different receptive fields, and optimizing model parameters; and (3) processing the target pin image by using the trained pin detection network model, judging whether a pin exists in the image, and sending out an alarm signal if the pin is not detected. The invention can improve the intelligence and accuracy of the missing of the transformer station pins.
Description
Technical Field
The invention belongs to the technical field of power equipment target detection, and particularly relates to an intelligent detection method and device for pin loss of a transformer substation.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
During operation of substation equipment, the integrity of the pins is critical to the safe operation of the equipment. However, due to the numerous pieces of substation equipment, the traditional manual detection method is time-consuming and error-prone, and is difficult to meet the actual requirements.
In addition, the pin has small volume, the pixel area difference is large in a large-scene inspection picture and a close-up picture, and certain difference exists in the pin appearance used on different transformer substation equipment, so that the problem of low accuracy exists when the existing neural network model is utilized for image processing.
Disclosure of Invention
In order to solve the problems, the invention provides an intelligent detection method and device for the pin loss of a transformer substation.
According to some embodiments, the present invention employs the following technical solutions:
an intelligent detection method for the pin loss of a transformer substation comprises the following steps:
acquiring image data comprising multiple time points, multiple angles and multiple shooting distances of pins;
preprocessing image data;
marking the pin positions in each preprocessed image to be used as a training sample data set for pin detection;
training the built pin detection network model by using the training sample data set, acquiring feature images of different receptive fields of the training sample image by using downsampling operation with different multiples of the pin detection network model, fusing the feature images, retaining the features of the pin under the different receptive fields, and optimizing model parameters;
and (3) processing the target pin image by using the trained pin detection network model, judging whether a pin exists in the image, and sending out an alarm signal if the pin is not detected.
As an alternative embodiment, the image data includes multiple sets of identical pin images captured at different time points, different angles, and different capturing distances, and pins captured by each set of images are different.
Further, the pin images shot at different shooting distances have different picture ratios of the pins in the images.
As an alternative embodiment, the preprocessing process includes image denoising, rotation, brightness adjustment, and image enhancement.
As an alternative implementation manner, the pin detection network model uses a convolution check training image with a size of 3*3 and a step size of 2 to perform a convolution operation, and obtains a convolution characteristic map of a training sample, where the operation is downsampling.
As an alternative embodiment, the feature map receptive field with smaller downsampling multiple is smaller, which is suitable for processing pin images with larger frames, and the feature map receptive field with larger downsampling multiple is larger, which is suitable for processing pin images with larger frames.
As a further embodiment, the high-power downsampled feature map is upsampled to remain the same size as the shallow feature map.
Alternatively, the specific process of optimizing the model parameters includes training the model parameters through a loss function constraint and back propagation network until an iteration condition is met.
Intelligent detection device that transformer substation's pin was absent includes:
the image acquisition module is configured to acquire image data of multiple time points, multiple angles and multiple shooting distances of the pin;
a preprocessing module configured to preprocess image data;
the training sample data set construction module is configured to acquire marking information of pin positions in each preprocessed image to serve as a training sample data set for pin detection;
the pin detection network model optimizing module is configured to train the constructed pin detection network model by utilizing the training sample data set, acquire feature graphs of different receptive fields of the training sample image by utilizing downsampling operation with different multiples, fuse the feature graphs, reserve the features of the pin under the different receptive fields, and optimize model parameters;
the target detection module is configured to utilize the trained pin detection network model to process target pin images, judge whether pins exist in the images, and send out alarm signals if the pins are not detected.
As an alternative embodiment, an image capturing apparatus is further included, the resolution of which exceeds a set value.
As an alternative embodiment, the device further comprises an output module configured to display the target detection result.
Compared with the prior art, the invention has the beneficial effects that:
the invention innovatively provides an intelligent detection method and device for realizing pin loss of a transformer substation by utilizing artificial intelligence and a target detection technology, and the method and device can acquire pin images with different picture occupation ratios under different states by utilizing image data comprising multiple time points, multiple angles and multiple shooting distances of pins, so that the accuracy of subsequent detection is ensured; and by utilizing multiple downsampling and utilizing characteristic diagrams of different receptive fields, different images are processed in a self-adaptive mode, and the pin detection rate of different sizes in a picture is effectively improved. The invention can realize the real-time monitoring of the integrity of the pin, and timely find out abnormal conditions, thereby saving manpower resources and cost.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
Fig. 1 is a flow chart of the present embodiment.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
Example 1
An intelligent detection method for the pin loss of a transformer substation comprises the following steps:
step 1, data acquisition
Because the volume of the pin is smaller, the pixel area difference between the large scene inspection picture and the close-up picture is large; there is a certain difference in the pin shapes used on different substation equipment. And shooting the pin installation part of the substation equipment by using high-resolution camera equipment, so that the details and characteristics of the pin installation part can be comprehensively shot. A series of images are acquired through shooting at multiple time points and multiple angles by the image pickup device, so that pin image data in different states are ensured to be acquired. And through multi-distance shooting, pin images with different picture occupation ratios are ensured to be acquired.
Step 2, data preprocessing
Preprocessing the acquired image data, including image denoising, rotation, brightness adjustment, mosaic image enhancement and the like, so as to improve the accuracy of subsequent target detection.
Step 3, training the target detection neural network
And marking the pin positions in the preprocessed images to serve as a pin detection algorithm training data set. The pin detection network model is trained.
Further, the convolution check training image with the size of 3*3 and the step length of 2 is used for convolution operation, so that a convolution characteristic diagram of the training sample is obtained, and the operation is downsampling.
Further, feature images of different receptive fields of the image are obtained through multiple downsampling, the receptive fields of the feature images with small downsampling multiple are small, the geometric detail information representation capability is strong, the feature images with large downsampling multiple are suitable for processing small targets, the receptive fields of the feature images with large downsampling multiple are large, the semantic information representation capability is strong, but the resolution of the feature images is low, the geometric information representation capability is weak, and the feature images with large downsampling multiple are suitable for processing large targets. The feature map of the high-power downsampling is upsampled to keep the same size with the feature map of the shallow layer, then feature map fusion is carried out, and the features of the pin under different receptive fields are reserved; the pin detection rate of different sizes in the picture is effectively improved.
Further, model parameters are trained through loss function constraint and back propagation network to obtain an optimal pin detection model.
Step 4, judging the missing of the pin
And inputting a pin part image of the substation equipment, and judging whether a pin exists in the image through the prediction of the pin detection model. If no pin is detected, an alarm signal is issued.
Step 5, outputting the result
And outputting the detection result to a display device or a storage device for the operator to check and analyze.
Example two
Intelligent detection device that transformer substation's pin was absent includes:
and the image pickup equipment is used for collecting image data of the substation equipment.
The image acquisition module is configured to acquire image data of multiple time points, multiple angles and multiple shooting distances of the pin;
a preprocessing module configured to preprocess image data;
the training sample data set construction module is configured to acquire marking information of pin positions in each preprocessed image to serve as a training sample data set for pin detection;
the pin detection network model optimizing module is configured to train the constructed pin detection network model by utilizing the training sample data set, acquire feature graphs of different receptive fields of the training sample image by utilizing downsampling operation with different multiples, fuse the feature graphs, reserve the features of the pin under the different receptive fields, and optimize model parameters;
the target detection module is configured to utilize the trained pin detection network model to process target pin images, judge whether pins exist in the images, and send out alarm signals if the pins are not detected.
And the result output module is used for outputting the detection result to the display device or the storage device.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each 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.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which do not require the inventive effort by those skilled in the art, are intended to be included within the scope of the present invention.
Claims (10)
1. The intelligent detection method for the pin loss of the transformer substation is characterized by comprising the following steps of:
acquiring image data of pins at multiple time points, multiple angles and multiple shooting distances;
preprocessing image data;
marking the pin positions in each preprocessed image to be used as a training sample data set for pin detection;
training the built pin detection network model by using the training sample data set, acquiring feature images of different receptive fields of the training sample image by using downsampling operation with different multiples of the pin detection network model, fusing the feature images, retaining the features of the pin under the different receptive fields, and optimizing model parameters;
and (3) processing the target pin image by using the trained pin detection network model, judging whether a pin exists in the image, and sending out an alarm signal if the pin is not detected.
2. The intelligent detection method for pin missing of transformer substation according to claim 1, wherein the image data comprises a plurality of groups of identical pin images shot at different time points, different angles and different shooting distances, and pins shot by the images in each group are different.
3. The intelligent detection method for pin missing of transformer substation according to claim 2, wherein the pin images shot at different shooting distances have different picture ratios in the images.
4. The intelligent detection method for pin loss of transformer substation according to claim 1, wherein the preprocessing process comprises image denoising, rotation, brightness adjustment and image enhancement.
5. The intelligent detection method for pin missing of transformer substation according to claim 1, wherein the pin detection network model uses a convolution check training image with a size of 3*3 and a step length of 2 to perform convolution operation, and a convolution feature map of a training sample is obtained, and the operation is downsampling.
6. The intelligent detection method for pin missing of transformer substation according to claim 1, wherein the feature map with smaller downsampling multiple is smaller in receptive field and suitable for processing pin images with larger frames, and the feature map with larger downsampling multiple is larger in receptive field and suitable for processing pin images with larger frames.
7. An intelligent detection method for pin loss of transformer substation according to claim 1 or 6, wherein the high-power downsampled feature map is upsampled to be the same size as the shallow feature map.
8. The intelligent detection method for pin loss of transformer substation according to claim 1, wherein the specific process of optimizing the model parameters comprises training the model parameters through loss function constraint and back propagation network until the iteration condition is satisfied.
9. Intelligent detection device that transformer substation's pin was absent, characterized by includes:
the image acquisition module is configured to acquire image data of multiple time points, multiple angles and multiple shooting distances of the pin;
a preprocessing module configured to preprocess image data;
the training sample data set construction module is configured to acquire marking information of pin positions in each preprocessed image to serve as a training sample data set for pin detection;
the pin detection network model optimizing module is configured to train the constructed pin detection network model by utilizing the training sample data set, acquire feature graphs of different receptive fields of the training sample image by utilizing downsampling operation with different multiples, fuse the feature graphs, reserve the features of the pin under the different receptive fields, and optimize model parameters;
the target detection module is configured to utilize the trained pin detection network model to process target pin images, judge whether pins exist in the images, and send out alarm signals if the pins are not detected.
10. The intelligent substation pin missing detection device according to claim 9, further comprising a camera device, wherein the resolution of the camera device exceeds a set value;
or/and, the device further comprises an output module configured to display the target detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311367918.5A CN117649374A (en) | 2023-10-20 | 2023-10-20 | Intelligent detection method and device for pin loss of transformer substation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311367918.5A CN117649374A (en) | 2023-10-20 | 2023-10-20 | Intelligent detection method and device for pin loss of transformer substation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117649374A true CN117649374A (en) | 2024-03-05 |
Family
ID=90042266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311367918.5A Pending CN117649374A (en) | 2023-10-20 | 2023-10-20 | Intelligent detection method and device for pin loss of transformer substation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117649374A (en) |
-
2023
- 2023-10-20 CN CN202311367918.5A patent/CN117649374A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110650316A (en) | Intelligent patrol and early warning processing method and device, electronic equipment and storage medium | |
| CN112528971B (en) | Power transmission line abnormal target detection method and system based on deep learning | |
| CN106851229B (en) | Security and protection intelligent decision method and system based on image recognition | |
| CN111222478A (en) | Construction site safety protection detection method and system | |
| CN111008961A (en) | Transmission line equipment defect detection method and system, equipment and medium thereof | |
| CN111815623A (en) | Power transmission line cotter pin missing identification method | |
| CN109297978B (en) | Binocular imaging-based power line unmanned aerial vehicle inspection and defect intelligent diagnosis system | |
| CN109326125B (en) | Picture quality diagnosis system and method based on embedded system | |
| CN111695493A (en) | Method and system for detecting hidden danger of power transmission line | |
| CN113299073A (en) | Method, device, equipment and storage medium for identifying illegal parking of vehicle | |
| CN115082813A (en) | Detection method, unmanned aerial vehicle, detection system and medium | |
| CN114863311A (en) | Automatic tracking method and system for inspection target of transformer substation robot | |
| CN111147815A (en) | Video monitoring system | |
| CN114565952A (en) | Pedestrian trajectory generation method, device, equipment and storage medium | |
| CN117649374A (en) | Intelligent detection method and device for pin loss of transformer substation | |
| CN111369591A (en) | Method, device and equipment for tracking moving object | |
| CN115984672A (en) | Method and device for detecting small target in high-definition image based on deep learning | |
| CN114494931A (en) | Intelligent classification processing method and system for video image faults | |
| CN110765932B (en) | Scene change sensing method | |
| CN114549628A (en) | Power pole inclination detection method, device, equipment and storage medium | |
| CN114360064A (en) | Office place personnel behavior lightweight target detection method based on deep learning | |
| CN109325441B (en) | Method for identifying insulator object of power transmission line | |
| CN113239931A (en) | Logistics station license plate recognition method | |
| CN113033355A (en) | Abnormal target identification method and device based on intensive power transmission channel | |
| KR102520218B1 (en) | System and Method for improving hardware usage in control server using artificial intelligence image processing |
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 |