CN116707143A - Three-dimensional visual information processing-based intelligent video monitoring management system for power grid - Google Patents
Three-dimensional visual information processing-based intelligent video monitoring management system for power grid Download PDFInfo
- Publication number
- CN116707143A CN116707143A CN202310967511.XA CN202310967511A CN116707143A CN 116707143 A CN116707143 A CN 116707143A CN 202310967511 A CN202310967511 A CN 202310967511A CN 116707143 A CN116707143 A CN 116707143A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- module
- video
- image
- information processing
- 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.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 102
- 230000010365 information processing Effects 0.000 title claims abstract description 23
- 230000000007 visual effect Effects 0.000 title claims abstract description 23
- 238000007689 inspection Methods 0.000 claims abstract description 17
- 238000010191 image analysis Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 24
- 230000011218 segmentation Effects 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 20
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 14
- 230000004927 fusion Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 9
- 230000003213 activating effect Effects 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 238000001931 thermography Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 12
- 230000010354 integration Effects 0.000 abstract description 2
- 238000007619 statistical method Methods 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 abstract 1
- 230000001066 destructive effect Effects 0.000 abstract 1
- 230000009545 invasion Effects 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007654 immersion Methods 0.000 description 1
- 238000000034 method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 231100000862 numbness Toxicity 0.000 description 1
- 231100000614 poison Toxicity 0.000 description 1
- 230000007096 poisonous effect Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06N3/048—Activation functions
-
- 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
- G06N3/09—Supervised learning
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00032—Systems characterised by the controlled or operated power network elements or equipment, the power network elements or equipment not otherwise provided for
- H02J13/00034—Systems characterised by the controlled or operated power network elements or equipment, the power network elements or equipment not otherwise provided for the elements or equipment being or involving an electric power substation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Power Engineering (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a three-dimensional visual information processing-based intelligent video monitoring management system for a power grid, which comprises a monitoring workstation, a monitoring center video master station, a multi-dimensional intelligent remote inspection system host, an environment monitoring module, an image analysis module based on the three-dimensional visual information processing, a field management and control linkage control module and a video working mode selection module. The intelligent video monitoring management system can realize automation, informatization and digitization of all facilities in the substation, and uploads data to the statistical analysis system through an effective data integration and collection system, so that technical support and decision basis are provided for operation management work, anomalies are found in time, and accidents are reduced. By researching the environment and security comprehensive monitoring technology by adopting an image analysis module technology based on three-dimensional visual information processing, active video monitoring of the internal and external environments of the transformer substation can be realized, and invasion and destructive behaviors of the transformer substation can be timely processed.
Description
Technical Field
The invention belongs to the technical field of power grid monitoring, and particularly relates to a power grid intelligent video monitoring management system based on three-dimensional visual information processing.
Background
With the development of computer network technology, digital video communication technology and digital image processing technology, the unattended transformer station has very good technical support for realizing remote image monitoring (commonly called remote vision) and intelligent analysis and early warning. The security monitoring products in the power industry are continuously developed along with the development of the power industry, and like the existing mature comprehensive automatic system, the security monitoring products are also developed under the accumulation of experiences for many years and are continuously perfect systems, but the security monitoring products are too focused on the management of special equipment of a power system, the fire fighting, anti-theft, dynamic data acquisition and post-processing analysis capabilities of the power system are slightly insufficient, and the technological utilization degree of equipment and systems such as the existing video monitoring alarm is also behind the main stream of security monitoring markets. With the continuous progress of the on-site comprehensive automation level of the transformer substation and the continuous increase of the number of unattended transformer substations, the video inspection work of the inspection personnel on the controlled stations is quite challenging, the capability of effectively inspecting video images of each controlled station and the related inspection work quality can be reduced due to information overload, an intelligent remote vision analysis early warning system is needed to assist the inspection personnel to comprehensively control various conditions of the controlled stations, such as environmental conditions, equipment operation, civilized production and the like, the situation of seriously threatening safety production, such as personnel management, security, power environment, comprehensive self-system, facility theft, malicious damage and the like in the stations is enhanced, and the critical instant stimulus type accurate detection is carried out on site abnormal conditions, so that rapid and intelligent decision and real-time response are carried out on events, and the real-time monitoring level and the safety level of the unmanned transformer substation are practically improved.
Disclosure of Invention
The invention aims to provide a three-dimensional visual information processing-based intelligent video monitoring management system for a power grid, which mainly solves the problems of high labor intensity and low working efficiency of the existing power grid inspection personnel.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
an intelligent video monitoring management system of a power grid based on three-dimensional visual information processing, comprising:
the monitoring workstation is used for managing the whole-station monitoring video of the transformer substation;
the monitoring center video master station is used for receiving the remote returned monitoring video and uploading the remote returned monitoring video to the monitoring workstation;
the multidimensional intelligent remote inspection system host is used for managing the intelligent equipment of the transformer substation site and uploading video monitoring information to a monitoring center video master station;
the environment monitoring module is used for monitoring the static environment of the transformer substation and uploading monitoring information to the multi-dimensional intelligent remote inspection system host;
the image analysis module is used for monitoring the dynamic environment of the transformer substation and uploading the monitoring information to the multi-dimensional intelligent remote inspection system host;
the field control linkage control module is used for carrying out linkage control on intelligent equipment on the transformer substation field;
and the video working mode selection module is used for carrying out mode selection control on video monitoring equipment on the transformer substation site.
Further, in the invention, a YOLOv4-tiny network based on three-dimensional visual information processing is adopted as a backbone network for image information extraction in the image analysis module, wherein the YOLOv4-tiny network comprises 4 image feature fusion modules and 4 image feature transmission modules; each feature fusion module comprises a sampling layer, a grouping layer and a feature extraction layer.
Further, in the present invention, the feature extraction layer extracts image features using two-dimensional convolution blocks, each of which is composed of 2 convolution layers of 3×3, 1 batch normalization layer, and 1 correction linear unit activation layer.
Further, in the invention, the sampling layer classifies the acquired images through the depth information of the image coordinates; wherein, the depth information of the image coordinates is the three-dimensional coordinate information (x, y, z) of the original image, and when z is larger than or equal to a set value d, the depth information is regarded as a far point; when z is smaller than a set value d, the near points are regarded as, and then all the far points are reserved, so that the original sparse far points are not removed due to downsampling; for the near points, dividing and scoring the image information through the grouping layer; the segmentation function for segmentation is as follows:
in the formula ,pthree-dimensional coordinate information (x, y, z) for the original image;activating a function for Sigmoid; />For the first convolution layer, the information of the image is increased from 3 dimension to 128 dimension; />For the second convolution layer, reducing the dimension of the information of the image from 128 dimensions to 1 dimension, obtaining the segmentation score of each near point through the calculation of a Sigmoid function, sequencing the segmentation scores of the near points, selecting 2048 near points with the highest score, and randomly sampling the rest near points until the total points reach the preset total number, thereby being used as the input of the first feature fusion layer of the YOLOv4-tiny network.
Further, in the present invention, the segmentation function is trained by adopting an RCNN algorithm, and is supervised by adopting a cross entropy loss function, wherein the loss function is as follows:
wherein ,is the total number of the original image information;Eexpected for cross entropy loss functions; />A predictive segmentation score for the point calculated from the segmentation function; />The true value of the score is partitioned for that point.
Further, in the invention, the environment monitoring module comprises a temperature monitoring module, a humidity monitoring module, a harmful gas monitoring module, a water immersion monitoring module, a switch contact temperature measuring module, a cable temperature measuring module and a security perimeter monitoring module.
Further, in the invention, the image analysis module comprises a personnel video dynamic analysis module, a small animal video dynamic analysis module, a running track analysis module, an infrared thermal imaging analysis module, a working range warning analysis module and a crossing warning module.
Further, in the invention, the field control linkage control module comprises an acousto-optic remote linkage control module, a humidity and dehumidifier linkage control module, a gas and fan linkage control module, a water level and water pump linkage control module, and a network camera and light linkage control module.
Further, in the invention, the video working mode selection module comprises a working monitoring mode selection module, an automatic cruising mode selection module and a security mode selection module.
Compared with the prior art, the invention has the following beneficial effects:
(1) The intelligent video monitoring management system can realize automation, informatization and digitization of all facilities in the substation, and uploads data to the statistical analysis system through an effective data integration and collection system, so that technical support and decision basis are provided for operation management work, anomalies are found in time, and accidents are reduced.
(2) According to the invention, the environment and security comprehensive monitoring technology is researched by adopting an image analysis module technology based on three-dimensional visual information processing, so that dynamic environment and security data of a transformer substation can be mastered, abnormal environment of the transformer substation can be timely found and timely processed, and power failure accidents caused by the environment of the transformer substation are reduced; the active video monitoring can timely process the application of the intrusion and damage actions to the active video analysis technology, so that the active video monitoring on the internal and external environments of the transformer substation can be realized, and the intrusion and damage actions on the transformer substation can be timely processed.
Drawings
Fig. 1 is a schematic structural diagram of a management system according to the present invention.
Detailed Description
The invention will be further illustrated by the following description and examples, which include but are not limited to the following examples.
As shown in fig. 1, the intelligent video monitoring management system for the power grid based on three-dimensional visual information processing disclosed by the invention comprises a monitoring workstation for managing the total station monitoring video of a transformer substation, and a monitoring center video master station for receiving the monitoring video transmitted back remotely and uploading the monitoring video to the monitoring workstation; the multidimensional intelligent remote patrol system host is used for managing the intelligent field equipment of the transformer substation and uploading video monitoring information to the video master station of the monitoring center; the environment monitoring module is used for monitoring the static environment of the transformer substation and uploading monitoring information to a multi-dimensional intelligent remote inspection system host; the image analysis module is used for monitoring the dynamic environment of the transformer substation and uploading monitoring information to a multi-dimensional intelligent remote inspection system host machine based on three-dimensional visual information processing; the field control linkage control module is used for carrying out linkage control on intelligent equipment on the transformer substation field; and a video working mode selection module for mode selection control of the video monitoring equipment of the transformer substation site.
In this embodiment, the environment monitoring module includes a temperature monitoring module, a humidity monitoring module, a harmful gas monitoring module, a water logging monitoring module, a switch contact temperature measuring module, a cable temperature measuring module, and a security perimeter monitoring module. All the monitoring modules can be obtained through direct purchasing, the temperature monitoring module is mainly used for monitoring the environmental temperature of operation equipment of the transformer substation, and the temperature monitoring module is matched with the gas and fan linkage control module for use to realize the temperature regulation of the transformer substation; the humidity monitoring module is mainly used for monitoring the environmental humidity of operation equipment of the transformer substation, and is matched with the humidity and dehumidifier linkage control module for use, so that the humidity adjustment of the transformer substation is realized; the switch contact temperature measuring module is mainly used for monitoring the temperature of the switch contact in the transformer substation; the security perimeter monitoring module is mainly used for monitoring the environment of the surrounding limit of the transformer substation.
The cable temperature measuring module adopts an online infrared bus temperature measuring technology and a wireless temperature measuring technology for the cable temperature. The problem that the temperature of the bus can be measured without contacting the bus is solved, and the temperature of the cable is detected by a wireless communication technology. The temperature of the bus and the cable is mastered in real time, and the fault early warning effect is achieved.
Ozone O of harmful gas monitoring module for corrosion cable 3 Performing gas detection to prevent the cable from being corroded; to switch room SF 6 and O2 And detecting leakage conditions in real time, and ensuring personnel safety. H in cable pit 2 S gas detection and real-time mastering of poisonous gas H in cable pit 2 S concentration, the life safety of cable pit constructors is guaranteed.
The water logging monitoring module is linked with the water level and water pump linkage control module to realize the starting and stopping of the unmanned pump, so that the manual inspection frequency is reduced, the underground water leakage detection and the water pump linkage technology are applied, the starting and stopping of the unmanned pump can be realized, and the manual inspection frequency can be reduced.
In this embodiment, the on-site control linkage control module includes an acousto-optic remote linkage control module, a humidity and dehumidifier linkage control module, a gas and fan linkage control module, a water level and water pump linkage control module, and a network camera and light linkage control module.
In this embodiment, the video working mode selection module includes a working monitoring mode selection module, an auto-cruise mode selection module, and a security mode selection module. In daily management, a manager can choose one of the modes in which the remote inspection system operates through monitoring work.
In this embodiment, the image analysis module includes a personnel video dynamic analysis module, a small animal video dynamic analysis module, a running track analysis module, an infrared thermal imaging analysis module, a working range warning analysis module and a boundary crossing warning module. In the image analysis module, a Yolov4-tiny network based on three-dimensional visual information processing is adopted as a backbone network for image information extraction, the network can effectively learn local multi-scale characteristics of image information, and the Yolov4-tiny network comprises 4 image characteristic fusion modules and 4 image characteristic transmission modules; each feature fusion module comprises a sampling layer, a grouping layer and a feature extraction layer. In this embodiment, RGB images captured by a camera are used as input, and feature maps of different sizes including rich semantic information are output. The feature extraction layer extracts image features by adopting two-dimensional convolution blocks, and each convolution block consists of 2 convolution layers with the size of 3 multiplied by 3, 1 batch normalization layer and 1 correction linear unit activation layer. And the step length of the 2 nd convolution layer is set to 2, so that the receptive field is enlarged, and the calculated amount of the network model is reduced.
In the embodiment, during image processing, firstly, the features of an original image are extracted and fused, the image points in a scene are divided into foreground points and background points through the extracted features, and an initial gray feature map is generated for each foreground point; and then, finely adjusting the generated gray feature images, deleting isolated points of the images, and finally, for each image picture, only keeping one gray feature image with highest confidence coefficient as an output result.
The plurality of gray feature maps that are then generated are refined and fine tuned. The refinement network comprises 3 feature fusion modules, 512 pieces of image information are randomly sampled by the generated multiple gray feature images to serve as input of the refinement network (the number of the image information is less than 512 and the image information is randomly resampled), the images are respectively downsampled to 128, 32 and 1 after passing through the feature fusion modules, and finally a one-dimensional vector is output to represent the classification confidence of the images.
In video image processing, in order to make the image processing information quicker and more accurate, the sampling layer classifies the acquired images through the depth information of the image coordinates; wherein, the depth information of the image coordinates is the three-dimensional coordinate information (x, y, z) of the original image, and when z is larger than or equal to a set value d, the depth information is regarded as a far point; when z is smaller than a set value d, the near points are regarded as, and then all the far points are reserved, so that the original sparse far points are not removed due to downsampling; in this embodiment, the set value is 50. For near points, the image information is segmented and scored by the grouping layer; the segmentation function for segmentation is as follows:
in the formula ,pthree-dimensional coordinate information (x, y, z) for the original image;activating a function for Sigmoid; />For the first convolution layer, the information of the image is increased from 3 dimension to 128 dimension; />For the second convolution layer, reducing the dimension of the information of the image from 128 dimensions to 1 dimension, obtaining the segmentation score of each near point through the calculation of a Sigmoid function, sequencing the segmentation scores of the near points, selecting 2048 near points with the highest score, and randomly sampling the rest near points until the total points reach the preset total number, thereby being used as the input of the first feature fusion layer of the YOLOv4-tiny network.
The embodiment also adopts RCNN algorithm to train the segmentation function, adopts cross entropy loss function to monitor, and the loss function is:
wherein ,to the total number of the original image information;EExpected for cross entropy loss functions; />A predictive segmentation score for the point calculated from the segmentation function; />The true value of the score is partitioned for that point. Through model training, originally sparse gray image characteristic information cannot be removed due to downsampling, the number of far points in input data is guaranteed, and then the image processing capacity of the model on a far object is improved. Meanwhile, more foreground points are sent to the network for feature learning, so that abundant foreground information is extracted, and the network is also helpful to image processing of foreground objects.
Through the design, the invention realizes the working state of autonomous analysis and autonomous coping of the unattended transformer substation through the dynamic video system, various data acquisition systems and corresponding response control treatment systems, monitors and analyzes in real time in normal state, monitors the abnormality, carries out treatment under the condition that human intervention cannot be in place in time, achieves the purpose of autonomously eliminating the abnormality or reducing the loss to the maximum extent, ensures that the transformer substation is not just numbness ' seen ', but is required to be alive, gives the transformer substation certain thinking and treatment capability, and realizes the virtuous circle of ' feel ', ' seeing ' thinking ' and ' treatment '.
The above embodiment is only one of the preferred embodiments of the present invention, and should not be used to limit the scope of the present invention, but all the insubstantial modifications or color changes made in the main design concept and spirit of the present invention are still consistent with the present invention, and all the technical problems to be solved are included in the scope of the present invention.
Claims (9)
1. The utility model provides a power grid intelligent video monitoring management system based on three-dimensional visual information processing which characterized in that includes:
the monitoring workstation is used for managing the whole-station monitoring video of the transformer substation;
the monitoring center video master station is used for receiving the remote returned monitoring video and uploading the remote returned monitoring video to the monitoring workstation;
the multidimensional intelligent remote inspection system host is used for managing the intelligent equipment of the transformer substation site and uploading video monitoring information to a monitoring center video master station;
the environment monitoring module is used for monitoring the static environment of the transformer substation and uploading monitoring information to the multi-dimensional intelligent remote inspection system host;
the image analysis module is used for monitoring the dynamic environment of the transformer substation and uploading the monitoring information to the multi-dimensional intelligent remote inspection system host;
the field control linkage control module is used for carrying out linkage control on intelligent equipment on the transformer substation field;
and the video working mode selection module is used for carrying out mode selection control on video monitoring equipment on the transformer substation site.
2. The intelligent video monitoring and management system for the power grid based on three-dimensional visual information processing according to claim 1, wherein a YOLOv4-tiny network based on three-dimensional visual information processing is adopted as a backbone network for image information extraction in the image analysis module, and the YOLOv4-tiny network comprises 4 image feature fusion modules and 4 image feature transmission modules; each feature fusion module comprises a sampling layer, a grouping layer and a feature extraction layer.
3. The intelligent video monitoring and management system for the power grid based on three-dimensional visual information processing according to claim 2, wherein the feature extraction layer extracts image features by adopting two-dimensional convolution blocks, and each convolution block consists of 2 convolution layers of 3×3, 1 batch normalization layer and 1 correction linear unit activation layer.
4. The intelligent video monitoring and management system for the power grid based on three-dimensional visual information processing according to claim 3, wherein the sampling layer classifies the acquired images through depth information of image coordinates; wherein, the depth information of the image coordinates is the three-dimensional coordinate information (x, y, z) of the original image, and when z is larger than or equal to a set value d, the depth information is regarded as a far point; when z is smaller than a set value d, the near points are regarded as, and then all the far points are reserved, so that the original sparse far points are not removed due to downsampling; for the near points, dividing and scoring the image information through the grouping layer; the segmentation function for segmentation is as follows:
in the formula ,pthree-dimensional coordinate information (x, y, z) for the original image;activating a function for Sigmoid; />For the first convolution layer, the information of the image is increased from 3 dimension to 128 dimension; />For the second convolution layer, reducing the dimension of the information of the image from 128 dimensions to 1 dimension, obtaining the segmentation score of each near point through the calculation of a Sigmoid function, sequencing the segmentation scores of the near points, selecting 2048 near points with the highest score, and randomly sampling the rest near points until the total points reach the preset total number, thereby being used as the input of the first feature fusion layer of the YOLOv4-tiny network.
5. The three-dimensional visual information processing-based intelligent video monitoring and management system for a power grid of claim 4, wherein the segmentation function is trained by adopting an RCNN algorithm and supervised by adopting a cross entropy loss function, and the loss function is as follows:
wherein ,is the total number of the original image information;Eexpected for cross entropy loss functions; />A predictive segmentation score for the point calculated from the segmentation function; />The true value of the score is partitioned for that point.
6. The intelligent video monitoring management system of the power grid based on three-dimensional visual information processing according to claim 1, wherein the environment monitoring module comprises a temperature monitoring module, a humidity monitoring module, a harmful gas monitoring module, a water logging monitoring module, a switch contact temperature measuring module, a cable temperature measuring module and a security perimeter monitoring module.
7. The intelligent video monitoring and management system for the power grid based on three-dimensional visual information processing according to claim 1, wherein the image analysis module comprises a personnel video dynamic analysis module, a small animal video dynamic analysis module, a running track analysis module, an infrared thermal imaging analysis module, a working range warning analysis module and a crossing warning module.
8. The intelligent video monitoring management system of the power grid based on three-dimensional visual information processing according to claim 1, wherein the on-site control linkage control module comprises an acousto-optic remote linkage control module, a humidity and dehumidifier linkage control module, a gas and fan linkage control module, a water level and water pump linkage control module and a network camera and light linkage control module.
9. The intelligent video monitoring management system of the power grid based on three-dimensional visual information processing according to claim 1, wherein the video working mode selection module comprises a working monitoring mode selection module, an automatic cruising mode selection module and a security mode selection module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310967511.XA CN116707143B (en) | 2023-08-03 | 2023-08-03 | Three-dimensional visual information processing-based intelligent video monitoring management system for power grid |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310967511.XA CN116707143B (en) | 2023-08-03 | 2023-08-03 | Three-dimensional visual information processing-based intelligent video monitoring management system for power grid |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116707143A true CN116707143A (en) | 2023-09-05 |
CN116707143B CN116707143B (en) | 2024-01-12 |
Family
ID=87831467
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310967511.XA Active CN116707143B (en) | 2023-08-03 | 2023-08-03 | Three-dimensional visual information processing-based intelligent video monitoring management system for power grid |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116707143B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110707823A (en) * | 2019-11-08 | 2020-01-17 | 国网上海市电力公司 | Multi-dimensional intelligent remote inspection system for transformer substation |
CN111784026A (en) * | 2020-05-28 | 2020-10-16 | 国网信通亿力科技有限责任公司 | Cloud-side cooperative sensing-based all-dimensional physical examination system for electrical equipment of transformer substation |
CN112737121A (en) * | 2020-12-28 | 2021-04-30 | 内蒙古电力(集团)有限责任公司包头供电局 | Intelligent video monitoring, analyzing, controlling and managing system for power grid |
CN113991837A (en) * | 2021-09-30 | 2022-01-28 | 国网江苏省电力有限公司 | Protection device intelligence inspection system based on many first perception techniques |
CN113989939A (en) * | 2021-11-16 | 2022-01-28 | 河北工业大学 | Small-target pedestrian detection system based on improved YOLO algorithm |
CN114373099A (en) * | 2022-01-05 | 2022-04-19 | 上海交通大学 | Three-dimensional point cloud classification method based on sparse graph convolution |
CN115516516A (en) * | 2020-03-04 | 2022-12-23 | 奇跃公司 | System and method for efficient floor plan generation from 3D scanning of indoor scenes |
-
2023
- 2023-08-03 CN CN202310967511.XA patent/CN116707143B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110707823A (en) * | 2019-11-08 | 2020-01-17 | 国网上海市电力公司 | Multi-dimensional intelligent remote inspection system for transformer substation |
CN115516516A (en) * | 2020-03-04 | 2022-12-23 | 奇跃公司 | System and method for efficient floor plan generation from 3D scanning of indoor scenes |
CN111784026A (en) * | 2020-05-28 | 2020-10-16 | 国网信通亿力科技有限责任公司 | Cloud-side cooperative sensing-based all-dimensional physical examination system for electrical equipment of transformer substation |
CN112737121A (en) * | 2020-12-28 | 2021-04-30 | 内蒙古电力(集团)有限责任公司包头供电局 | Intelligent video monitoring, analyzing, controlling and managing system for power grid |
CN113991837A (en) * | 2021-09-30 | 2022-01-28 | 国网江苏省电力有限公司 | Protection device intelligence inspection system based on many first perception techniques |
CN113989939A (en) * | 2021-11-16 | 2022-01-28 | 河北工业大学 | Small-target pedestrian detection system based on improved YOLO algorithm |
CN114373099A (en) * | 2022-01-05 | 2022-04-19 | 上海交通大学 | Three-dimensional point cloud classification method based on sparse graph convolution |
Non-Patent Citations (1)
Title |
---|
岳驰涛: ""基于RGB-D图像的室内场景3D目标检测"", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 8, pages 138 - 686 * |
Also Published As
Publication number | Publication date |
---|---|
CN116707143B (en) | 2024-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110996067A (en) | Personnel safety real-time intelligent video monitoring system under high-risk operation environment based on deep learning | |
CN112257632A (en) | Transformer substation monitoring system based on edge calculation | |
CN111814742A (en) | Knife switch state identification method based on deep learning | |
CN113569801B (en) | Distribution construction site live equipment and live area identification method and device thereof | |
CN211184122U (en) | Intelligent video analysis system for linkage of railway operation safety prevention and control and large passenger flow early warning | |
CN116052082A (en) | Power distribution station room anomaly detection method and device based on deep learning algorithm | |
CN115297305A (en) | Multi-video-stream security risk early warning system and method based on edge calculation | |
CN116707143B (en) | Three-dimensional visual information processing-based intelligent video monitoring management system for power grid | |
CN112702570A (en) | Security protection management system based on multi-dimensional behavior recognition | |
CN110852174A (en) | Early smoke detection method based on video monitoring | |
CN113284103B (en) | Substation equipment defect online detection method based on space transformation fast R-CNN model | |
CN115546121A (en) | Communication line optical cable defect identification method based on multi-scale target detection | |
Feng et al. | Identification of Key Links in Electric Power Operation Based-Spatiotemporal Mixing Convolution Neural Network. | |
Pan | Intelligent image recognition research on status of power transmission lines | |
CN111524158A (en) | Method for detecting foreground target in complex scene of hydraulic engineering | |
Fu et al. | Intelligent video analysis platform based on edge artificial intelligence technology in power industry | |
Yang et al. | Abnormal Scene Image Recognition Method for Intelligent Operation and Maintenance of Electrical Equipment in Substations | |
CN113724247B (en) | Intelligent substation inspection method based on image discrimination technology | |
CN117811218B (en) | Unmanned aerial vehicle transmission and transformation and distribution integrated collaborative inspection method | |
Murthy et al. | Distribution system insulator monitoring using video surveillance and support vector machines for complex background images | |
Liu | Research on Hot Operation of a Petrochemical Plant Based on Compound Edge Operator | |
CN111885349A (en) | Pipe rack abnormity detection system and method | |
Chen et al. | Vision-Based UAV Distribution Line Inspection Using Deep Learning | |
Wang et al. | Research on Application and Development of Intelligent Video Recognition Technology in Power Transmission, Transformer and Distribution System | |
Zhong et al. | Research on Optimization of Boundary Detection and Dangerous Area Warning Algorithms Based on Deep Learning in Campus Security System |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20231211 Address after: 014000 No. 4, Building 40, Urban Garden, Block 13, Ziyou Road, Qingshan District, Baotou City, Inner Mongolia Autonomous Region Applicant after: Li Zhibo Address before: Room 2609, 26th Floor, Unit 2, Building 1, No. 10 Jianshe Road, Chenghua District, Chengdu City, Sichuan Province, 610051 Applicant before: Sichuan Runzheng Technology Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |