CN111597868A - SSD-based substation disconnecting switch state analysis method - Google Patents
SSD-based substation disconnecting switch state analysis method Download PDFInfo
- Publication number
- CN111597868A CN111597868A CN202010016676.5A CN202010016676A CN111597868A CN 111597868 A CN111597868 A CN 111597868A CN 202010016676 A CN202010016676 A CN 202010016676A CN 111597868 A CN111597868 A CN 111597868A
- Authority
- CN
- China
- Prior art keywords
- image
- ssd
- switch
- disconnecting switch
- video
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 16
- 238000001514 detection method Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 14
- 238000012544 monitoring process Methods 0.000 claims abstract description 13
- 238000002955 isolation Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 238000012795 verification Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a substation disconnecting switch state analysis method based on SSD, which comprises the following steps: s1) acquiring a monitoring video of the disconnecting switch of the transformer substation, converting the video into a processable image data format, and forming a sample image library for subsequent network training; s2) image calibration: calibrating an isolation switch image in a sample image library, and marking the position and the opening and closing state of the isolation switch in the image; s3) model training: inputting the calibrated sample image and the corresponding text file into an SSD detection and identification network for training; s4) isolating switch detection identification: the method comprises the steps of obtaining a monitoring video of a disconnecting switch of the transformer substation, extracting a certain frame of image and converting the certain frame of image into a processable image data format, inputting the monitoring video and the image into a trained SSD network respectively, and obtaining the image and the video with the position, the opening and closing state classification result and the accuracy of the disconnecting switch. The method has the characteristics of high identification precision and quick video processing time, and provides technical support for automation of the transformer substation.
Description
Technical Field
The invention relates to the field of computer vision, in particular to a substation disconnecting switch state analysis method based on SSD.
Background
With the rapid increase of electric energy consumption, the scale of the electric power system in China is gradually enlarged year by year and the structure is more and more complex, so that the safe operation of the electric power system is more and more emphasized. In recent years, artificial intelligence technology represented by machine vision is rapidly developed and is applied to each link of a power system, automation of the power system is performed by means of the technical means of machine vision, so that not only can manpower consumption be reduced, but also the safety and efficiency of operation of the power system can be greatly improved.
At present, inspection robots put into operation in substations are mostly used for detecting infrared temperatures of electrical equipment and judging whether the electrical equipment has an overheating defect or not according to the infrared temperatures, but cannot automatically identify the operation states of the electrical equipment such as isolating switches, and the like, so that the application range of the inspection robots is greatly limited. The method for automatically judging the state of the electrical equipment of the isolating switch, which is disclosed at present, still has the characteristics of low analysis speed and low precision, for example, the intelligent analysis of the fitting video of the line segment of the disconnecting switch of the transformer substation, disclosed in the chinese patent document with the application number of 201510549097, judges the on-off state of the disconnecting switch according to the duty ratio and the number of intervals, the requirements on background images are high, the method cannot adapt to the light change conditions at different moments in a day, a transformer substation isolating switch detection and identification method based on Mask RCNN disclosed by Chinese patent document with application number CN201811513584 and a transformer substation isolating switch detection and identification method based on Faster RCNN disclosed by Chinese patent document with application number CN201910235821 respectively adopt Mask RCNN and fast RCNN, compared with the SSD algorithm, the detection rate of the two algorithms is obviously lower than that of the SSD algorithm, the requirements of fast video processing cannot be met, and the accuracy rate of the detection of the monitoring of a large target is obviously lower than that of the SSD algorithm.
Therefore, the SSD algorithm is utilized to analyze the state of the disconnecting switch of the transformer substation, so that the method has important significance for the production and operation of a power grid, and meanwhile, the method has great promotion effect on improving the intelligent level of the state inspection of the electrical equipment of the transformer substation.
Disclosure of Invention
The invention aims to provide a substation disconnecting switch state analysis method based on SSD, which solves the defects in the prior art, effectively judges the state of a disconnecting switch and improves the accuracy and the practicability of detection to the maximum extent.
In order to achieve the purpose, the invention is realized by the following technical scheme:
a transformer substation disconnecting switch state analysis method based on SSD comprises the following steps:
s1) acquiring a monitoring video of the substation disconnecting switch, converting the video into a processable image data format, and forming a sample image library for subsequent network training;
s2) image calibration: calibrating an isolation switch image in a sample image library, and marking the position and the opening and closing state of the isolation switch in the image;
s3) model training: inputting the calibrated sample image and the corresponding text file into an SSD (Single shot MultiBox) detection and recognition network for training;
s4) isolating switch detection identification: the method comprises the steps of obtaining a monitoring video of a disconnecting switch of the transformer substation, extracting a certain frame of image and converting the certain frame of image into a processable image data format, respectively inputting the monitoring video and the image into a trained SSD (Single shot Multi Box) network, and obtaining a classification result and accuracy of the position and the opening and closing state of the disconnecting switch.
Further, in step S2), considering the output result of the ssd (single shot multi-box detector) network, when calibrating the image, calibrating the boundary frame of the disconnecting switch blade and classifying the states "on/off" of the disconnecting switch at the same time;
further, in step S3), the ssd (single shot multi-box detector) detects that the recognition network model has the following four features:
s31) the SSD detection recognition network comprises a convolutional layer, an excitation layer and a pooling layer, and the convolutional layer is added to the basic model to obtain more feature maps for detection by taking VGG-16 as the basic model;
s32) the SSD detects and identifies that each newly added convolution layer is connected with an output layer in the network to obtain characteristic diagrams with different scales;
s33) the SSD detects and identifies the network and sets a priori frame with different length-width ratios for each unit in each feature map, and the priori frame is used as a reference of a prediction boundary frame to carry out position prediction;
s34) the SSD detects the recognition network and takes the confidence error and the position error of the bounding box into consideration to calculate the loss function;
further, in steps S4) and S5), the state determination results of the disconnecting switch are as follows:
s41) for the isolating switch image output result is an isolating switch image containing the class and confidence of the isolating switch classification and the boundary frame of the isolating switch knife, and for the isolating switch video output result is an isolating switch video containing the class of the isolating switch classification and the boundary frame of the isolating switch knife;
s42) for different classification results, are represented in the image to be detected and the video in different color bounding boxes.
According to the SSD-based substation disconnecting switch state analysis method provided by the invention, the on-site monitoring video of the variable electric field high-voltage disconnecting switch can be directly processed, the real-time switch state of the knife switch is obtained, and the operation safety and efficiency of a power system are greatly improved.
Drawings
The invention is further described with reference to the following drawings and detailed description:
fig. 1 is a flow chart of a method for analyzing the state of a substation disconnecting switch based on SSD according to the present invention.
Fig. 2 is a schematic diagram of an SSD network structure of the SSD-based substation disconnecting switch state analysis method of the present invention.
Fig. 3 is a process processing image of the SSD-based substation disconnecting switch state analysis method of the present invention.
Detailed Description
Embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
The invention relates to a substation disconnecting switch state analysis method based on SSD, which utilizes image acquisition equipment to acquire the video of a disconnecting switch, can effectively process images and videos of the disconnecting switch, and obtains the opening and closing state classification result of the images and the videos of the disconnecting switch and the position boundary frame of a switch blade. The method of the invention comprises the following steps:
s1) video acquisition and data format conversion
Erecting image acquisition equipment on the site of the disconnecting switch equipment, and acquiring video images of the opening and closing motion state of the disconnecting switch equipment to obtain real-time video data of the disconnecting switch motion (considering the robustness of the result, and performing video acquisition on the disconnecting switch from multiple angles);
considering that the image calibration is carried out aiming at the image, the acquired video data is decomposed by using ffmpeg software to obtain image data of each frame, and the image data is converted into a processable data format which is uniformly named as' 00XXXX.
S2) image calibration
Adopting labelImg software to calibrate the image, and specifically comprising the following steps:
s21) opening a path where the calibration image is located;
s22) establishing a bounding box for the target (isolator blade);
s23) labeling (opening and closing) the target object in the boundary frame;
s24), after the calibration of all the target objects in the pictures is finished, storing the generated xml files into a preset folder, and after the calibration of all the pictures is finished, exiting the software.
And after the generation of the image label of the isolating switch is finished, generating a training set and a verification set according to the proportion of 7: 3. Because the image file is large, the image file occupies a large memory when being directly input into a network for training, so that the image file and the label file are converted into tfrecrds files, the training is convenient, and the training sample is prepared.
S3) model training
In this embodiment, a tensrflow1.8.0 framework is mainly used for network training, and ssd (single shot multi box detector) detection and recognition network model training has the following four features:
s31) the SSD detection recognition network comprises a convolutional layer, an excitation layer and a pooling layer, and the convolutional layer is added to the basic model to obtain more feature maps for detection by taking VGG-16 as the basic model;
s32) connecting each newly added convolution layer with an output layer in the SSD detection identification network to obtain characteristic graphs of different scales, so that the accuracy of target detection is improved;
s33) setting a prior box: the SSD detection and identification network sets prior frames with different scales or length-width ratios for each unit, and the predicted boundary frame takes the prior frames as a reference, so that the training intensity can be reduced to a certain extent.
S34) loss function calculation: since the SSD detects and identifies the network while obtaining the bounding box and the confidence of the target, when performing the loss function calculation, the confidence error and the position error of the bounding box also need to be considered:
wherein L isconf(x, c) represents confidence error, Lloc(x, l, g) represents the position error of the bounding box.
Training the isolation switch tfrecrds training set and the verification set obtained in the step S2) as inputs of the SSD detection identification network, wherein the training adopts a small batch random gradient descent method, that is, each training selects a certain number (4 samples are selected in this embodiment) from the training set to calculate the average loss value of all the small batches, and sets the initial learning rate to be 0.001, the final learning rate to be 0.0001, and the learning rate attenuation rate in the iterative process to be 0.94, and performs back propagation to calculate the gradient training network. The training result can be obtained by predicting the verification set, and the training parameters in the SSD, such as the learning rate and the learning rate attenuation rate, are adjusted by checking the change of the error of the verification set.
S4) isolating switch detection identification
Acquiring a monitoring video of a substation disconnecting switch, extracting a certain frame of image and converting the certain frame of image into a processable image data format, and respectively inputting the monitoring video and the image data into a trained SSD (Single shot Multibx detector) network, wherein the identification result is as follows:
s41) for image input, the detection network output result is an isolating switch image containing the classification and confidence of the isolating switch classification and the boundary frame of the isolating switch knife; for the video, the detection network outputs the result as an isolating switch video containing the classification of the isolating switch and the boundary frame of the isolating switch knife;
s42) for different classification results, are represented in the image with different color bounding boxes.
And therefore, the judgment of the state of the disconnecting switch of the transformer substation is completed.
The present invention is well implemented according to the above embodiments, and it should be noted that, based on the above structural design, a plurality of improvements and modifications can be made without departing from the concept of the present invention, and these improvements and modifications should also be considered as within the protection scope of the present invention.
Claims (4)
1. A transformer substation disconnecting switch state analysis method based on SSD is characterized by comprising the following steps:
s1) acquiring a monitoring video of the substation disconnecting switch, converting the video into a processable image data format, and forming a sample image library for subsequent network training;
s2) image calibration: calibrating an isolation switch image in a sample image library, and marking the position and the opening and closing state of the isolation switch in the image;
s3) model training: inputting the calibrated sample image and the corresponding text file into an SSD (Single shot MultiBox) detection and recognition network for training;
s4) isolating switch detection identification: the method comprises the steps of obtaining a monitoring video of a disconnecting switch of the transformer substation, extracting a certain frame of image and converting the certain frame of image into a processable image data format, respectively inputting the monitoring video and the processed image data into a trained SSD (Single shot Multi-Box) network, and obtaining a classification result and accuracy of the position and the opening and closing state of the disconnecting switch.
2. The SSD-based substation disconnector state analysis method of claim 1, characterized in that:
step S2), when the image is calibrated, the calibration of the boundary frame of the disconnecting switch blade and the classification of the state "on/off" of the disconnecting switch are performed at the same time.
3. The SSD-based substation disconnector state analysis method of claim 1, characterized in that: in step S3), the ssd (single shot multi box detector) detects and recognizes that the network model training has the following features;
s31) the SSD detection recognition network comprises a convolutional layer, an excitation layer and a pooling layer, and the convolutional layer is added to the basic model to obtain more feature maps for detection by taking VGG-16 as the basic model;
s32) the SSD detects and identifies that each newly added convolution layer is connected with an output layer in the network to obtain characteristic diagrams with different scales;
s33) the SSD detects and identifies the network and sets a priori frame with different length-width ratios for each unit in each feature map, and the priori frame is used as a reference of a prediction boundary frame to carry out position prediction;
s34) the SSD detection recognition network performs the loss function calculation while considering the confidence error and the position error of the bounding box.
4. The SSD-based substation disconnector state analysis method of claim 1, characterized in that: step S4), the state determination result of the disconnector is as follows,
(1) the isolating switch image output result is an isolating switch image containing the class and confidence of the isolating switch classification and the boundary frame of the isolating switch knife, and the isolating switch video output result is an isolating switch video containing the class of the isolating switch classification and the boundary frame of the isolating switch knife;
(2) and representing the images to be detected and the videos by different color bounding boxes for different classification results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010016676.5A CN111597868A (en) | 2020-01-08 | 2020-01-08 | SSD-based substation disconnecting switch state analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010016676.5A CN111597868A (en) | 2020-01-08 | 2020-01-08 | SSD-based substation disconnecting switch state analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111597868A true CN111597868A (en) | 2020-08-28 |
Family
ID=72184880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010016676.5A Pending CN111597868A (en) | 2020-01-08 | 2020-01-08 | SSD-based substation disconnecting switch state analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111597868A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826577A (en) * | 2019-11-06 | 2020-02-21 | 国网新疆电力有限公司电力科学研究院 | High-voltage isolating switch state tracking identification method based on target tracking |
| CN112508019A (en) * | 2020-12-16 | 2021-03-16 | 国网江苏省电力有限公司检修分公司 | GIS isolation/grounding switch state detection method and system based on image recognition |
| CN113743355A (en) * | 2021-09-15 | 2021-12-03 | 中国南方电网有限责任公司超高压输电公司大理局 | Method, device and system for checking state of switch device and computer equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446869A (en) * | 2016-10-20 | 2017-02-22 | 国家电网公司 | Automatic detection method for state of isolating switch based on image intelligent recognition algorithm |
| CN108564065A (en) * | 2018-04-28 | 2018-09-21 | 广东电网有限责任公司 | A kind of cable tunnel open fire recognition methods based on SSD |
| CN108710913A (en) * | 2018-05-21 | 2018-10-26 | 国网上海市电力公司 | A kind of switchgear presentation switch state automatic identification method based on deep learning |
| CN109063764A (en) * | 2018-07-26 | 2018-12-21 | 福建和盛高科技产业有限公司 | A kind of judgment method of disconnecting switch closing operation in place based on machine vision |
| CN109101906A (en) * | 2018-07-27 | 2018-12-28 | 中国南方电网有限责任公司超高压输电公司贵阳局 | A kind of converting station electric power equipment infrared image exception real-time detection method and device |
| CN109684967A (en) * | 2018-12-17 | 2019-04-26 | 东北农业大学 | A kind of soybean plant strain stem pod recognition methods based on SSD convolutional network |
| US20190130580A1 (en) * | 2017-10-26 | 2019-05-02 | Qualcomm Incorporated | Methods and systems for applying complex object detection in a video analytics system |
| CN109712118A (en) * | 2018-12-11 | 2019-05-03 | 武汉三江中电科技有限责任公司 | A kind of substation isolating-switch detection recognition method based on Mask RCNN |
| CN109902629A (en) * | 2019-03-01 | 2019-06-18 | 成都康乔电子有限责任公司 | A kind of real-time vehicle target detection model under vehicles in complex traffic scene |
| CN110008877A (en) * | 2019-03-27 | 2019-07-12 | 国网内蒙古东部电力有限公司 | A kind of substation isolating-switch detection recognition method based on Faster RCNN |
-
2020
- 2020-01-08 CN CN202010016676.5A patent/CN111597868A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446869A (en) * | 2016-10-20 | 2017-02-22 | 国家电网公司 | Automatic detection method for state of isolating switch based on image intelligent recognition algorithm |
| US20190130580A1 (en) * | 2017-10-26 | 2019-05-02 | Qualcomm Incorporated | Methods and systems for applying complex object detection in a video analytics system |
| CN108564065A (en) * | 2018-04-28 | 2018-09-21 | 广东电网有限责任公司 | A kind of cable tunnel open fire recognition methods based on SSD |
| CN108710913A (en) * | 2018-05-21 | 2018-10-26 | 国网上海市电力公司 | A kind of switchgear presentation switch state automatic identification method based on deep learning |
| CN109063764A (en) * | 2018-07-26 | 2018-12-21 | 福建和盛高科技产业有限公司 | A kind of judgment method of disconnecting switch closing operation in place based on machine vision |
| CN109101906A (en) * | 2018-07-27 | 2018-12-28 | 中国南方电网有限责任公司超高压输电公司贵阳局 | A kind of converting station electric power equipment infrared image exception real-time detection method and device |
| CN109712118A (en) * | 2018-12-11 | 2019-05-03 | 武汉三江中电科技有限责任公司 | A kind of substation isolating-switch detection recognition method based on Mask RCNN |
| CN109684967A (en) * | 2018-12-17 | 2019-04-26 | 东北农业大学 | A kind of soybean plant strain stem pod recognition methods based on SSD convolutional network |
| CN109902629A (en) * | 2019-03-01 | 2019-06-18 | 成都康乔电子有限责任公司 | A kind of real-time vehicle target detection model under vehicles in complex traffic scene |
| CN110008877A (en) * | 2019-03-27 | 2019-07-12 | 国网内蒙古东部电力有限公司 | A kind of substation isolating-switch detection recognition method based on Faster RCNN |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826577A (en) * | 2019-11-06 | 2020-02-21 | 国网新疆电力有限公司电力科学研究院 | High-voltage isolating switch state tracking identification method based on target tracking |
| CN112508019A (en) * | 2020-12-16 | 2021-03-16 | 国网江苏省电力有限公司检修分公司 | GIS isolation/grounding switch state detection method and system based on image recognition |
| CN112508019B (en) * | 2020-12-16 | 2024-02-27 | 国网江苏省电力有限公司检修分公司 | GIS isolation/grounding switch state detection method and system based on image recognition |
| CN113743355A (en) * | 2021-09-15 | 2021-12-03 | 中国南方电网有限责任公司超高压输电公司大理局 | Method, device and system for checking state of switch device and computer equipment |
| CN113743355B (en) * | 2021-09-15 | 2024-01-09 | 中国南方电网有限责任公司超高压输电公司大理局 | Switch device state checking method, device, system and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Automatic fault diagnosis of infrared insulator images based on image instance segmentation and temperature analysis | |
| CN111597868A (en) | SSD-based substation disconnecting switch state analysis method | |
| CN112200178B (en) | Transformer substation insulator infrared image detection method based on artificial intelligence | |
| CN113436184B (en) | Power equipment image defect discriminating method and system based on improved twin network | |
| CN111652835A (en) | Method for detecting insulator loss of power transmission line based on deep learning and clustering | |
| Wang et al. | Insulator defect recognition based on faster R-CNN | |
| CN115346083A (en) | Temperature anomaly detection model training method, device, equipment and medium | |
| CN116681962A (en) | Power equipment thermal image detection method and system based on improved YOLOv5 | |
| CN108537792B (en) | Power defect image identification method based on convolutional neural network | |
| Li | Design of infrared anomaly detection for power equipment based on YOLOv3 | |
| CN116363536B (en) | Unmanned aerial vehicle inspection data-based power grid infrastructure equipment defect archiving method | |
| CN107204741B (en) | Method and device for determining environmental parameters | |
| CN115147591A (en) | Transformer equipment infrared image voltage heating type defect diagnosis method and system | |
| CN113284103B (en) | Substation equipment defect online detection method based on space transformation fast R-CNN model | |
| CN115311509A (en) | Power system transient stability evaluation method and system based on imaging data driving | |
| CN115224795A (en) | Intelligent substation equipment operation monitoring and early warning system and method | |
| CN112255141B (en) | Thermal imaging gas monitoring system | |
| Sheng et al. | A Method and Implementation of Transmission Line's Key Components and Defects Identification Based on YOLO | |
| Di et al. | Research on Real-Time Power Line Damage Detection Method Based on YOLO Algorithm | |
| Xudong et al. | Research of YOLOv5s Model Acceleration Strategy in AI Chip | |
| Wang et al. | Substation Equipment Defect Detection based on Temporal-spatial Similarity Calculation | |
| CN112036472A (en) | Visual image classification method and system for power system | |
| Shen et al. | Research on transmission equipment defect detection based on edge intelligent analysis | |
| Feng et al. | A Power Grid Equipment Fault Prediction Model Based on Faster RCNN and Video Streaming | |
| CN113487550B (en) | Target detection method and device based on improved activation function |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200828 |