CN115601707B - On-line monitoring method and system for power transmission line of power system - Google Patents
On-line monitoring method and system for power transmission line of power system Download PDFInfo
- Publication number
- CN115601707B CN115601707B CN202211373997.6A CN202211373997A CN115601707B CN 115601707 B CN115601707 B CN 115601707B CN 202211373997 A CN202211373997 A CN 202211373997A CN 115601707 B CN115601707 B CN 115601707B
- Authority
- CN
- China
- Prior art keywords
- image
- video
- video frame
- camera
- wind speed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 68
- 230000005540 biological transmission Effects 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000009545 invasion Effects 0.000 claims abstract description 45
- 230000003068 static effect Effects 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 238000011176 pooling Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 claims description 3
- 229910052744 lithium Inorganic materials 0.000 claims description 3
- 238000010248 power generation Methods 0.000 claims description 2
- 230000005764 inhibitory process Effects 0.000 claims 1
- 238000005265 energy consumption Methods 0.000 abstract description 11
- 238000013527 convolutional neural network Methods 0.000 description 18
- 230000008569 process Effects 0.000 description 12
- 238000009434 installation Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 6
- 238000010276 construction Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 238000013500 data storage Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- 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
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- 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
- G06V20/40—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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Computational Linguistics (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 relates to the field of online monitoring of electric power lines, in particular to an online monitoring method and an online monitoring system of an electric power system electric power line, which are used for monitoring foreign matter invasion of the electric power line, wherein the environment is divided into a relatively static environment and a relatively dynamic environment according to the ambient wind speed, in the relatively static environment, an image similarity calculation program is operated by a processor of a camera and is used for judging the foreign matter invasion condition, in the relatively dynamic environment, a monitoring video is uploaded to a server, and the foreign matter invasion condition is judged by an image recognition model; in a relatively static environment, uploading operation of images and operation of an image recognition model are not needed, so that electric energy consumption is reduced, and solar energy is better applied to each module of the power transmission line online monitoring system.
Description
Technical Field
The invention belongs to the field of online monitoring of power lines, and particularly relates to an online monitoring method and an online monitoring system of a power transmission line of a power system.
Background
In recent years, with the continuous perfection of basic design construction and the high-speed development of economic level in China, the construction of the power transmission line also enters a high-speed development period, but the safe operation of the power transmission line also encounters various challenges. According to statistics, among the reasons of the tripping accident of the power transmission line in the country in the last ten years, the foreign matter invasion causes a large proportion, and the foreign matter invasion refers to the region where dangerous sources enter the safe distance of the power transmission line, wherein the foreign matter invasion of the power transmission line is caused by artificial reasons such as irregular construction of a line passage and the foreign matter invasion caused by natural reasons such as swaying of the tree of the line passage along with wind power; therefore, foreign matter invasion accidents of the power transmission line must be effectively prevented and suppressed so as to ensure safe, stable and reliable power transmission.
In the prior art, image information is generally collected in real time through a camera, and images are identified so as to identify dangerous sources, for example, chinese patent (CN 211509200U) discloses an anti-external damage monitoring system of a power transmission line based on an artificial intelligence technology, which is shown in figure 1 and comprises a video monitoring camera and a monitoring terminal connected with the video monitoring camera in a signal manner, wherein the monitoring terminal comprises ground equipment, a graphic processor and a data storage, the video monitoring camera sends video signals to the graphic processor, the processed video data are stored in the data storage, and the graphic processor is electrically connected with the data storage and the ground equipment; by automatically identifying the video, operators do not need to monitor in real time, danger can be found in time, warning information can be pushed in real time, and the safety benefit is remarkable; however, the above-mentioned technical solution needs to upload real-time images to the graphics processor, and the graphics processor is executing the image recognition program all the time, resulting in data transmission and running of the image recognition program all the time, and the data uploading and the processor running all the time need to consume a large amount of electric energy, in the prior art, a solar panel is generally adopted as the energy supply device of the camera in the online monitoring system, which is a great test for uninterrupted energy supply of the energy supply device of the camera in the online monitoring system.
Disclosure of Invention
Aiming at the defects of the technical scheme, the invention provides an online monitoring method and an online monitoring system for a power transmission line of a power system, which are used for monitoring foreign matter invasion of the power transmission line, wherein the environment is divided into a relatively static environment and a relatively dynamic environment according to the condition of the wind speed of the environment, an image similarity calculation program is operated by a processor of a camera in the relatively static environment to judge the foreign matter invasion condition, and a monitoring video is uploaded to a server in the relatively dynamic environment to judge the foreign matter invasion condition by an image recognition model; in this way, in a relatively static environment, the uploading operation of the image and the running operation of the image recognition model are not needed, and the energy consumption is reduced.
In order to achieve the above object, according to one aspect of the present invention, an on-line monitoring method for a power transmission line of an electric power system is characterized by comprising the steps of:
step 1: collecting a monitoring field video through a camera of an online monitoring system;
the camera is a high-pixel monitoring snapshot integrated machine, in order to enable devices such as the camera to normally work in an outdoor independent environment of a power transmission line, the embodiment selects a solar cell panel for power generation, and a lithium battery with proper capacity is required to be provided as a standby battery in consideration of the fact that the solar cell panel cannot generate power at night and generates less power in overcast and rainy weather.
Furthermore, in order to enable the camera to have a better monitoring effect, specific consideration should be given to the installation position of the camera, on one hand, the voltage level of the power transmission line is different, and the requirements on the safe distance of the power transmission line are different, so that the safe distance between the camera and the power transmission line needs to be determined, and the installation of the camera cannot influence the normal working operation of the power transmission line; meanwhile, the definition of the image shot by the camera is considered, the installation position of the camera cannot be too high, and the camera is too high in installation, so that the monitored object distance is too far, and the difficulty is increased for image identification.
Step 2: the camera performs preprocessing operation on the collected video images;
due to the influence of field environment, light, weather and the like, some noise is mixed in the collected field video, so that the video image needs to be preprocessed; the preprocessing operation comprises color conversion, image enhancement and image filtering; the color conversion mainly completes the conversion from a color image to a gray image, so as to be beneficial to further image recognition; the image enhancement is to carry out image enhancement on the converted image, adjust the image contrast, highlight the target detail characteristics, and the image filtering is to eliminate noise interference through filtering, so as to improve the accuracy of feature extraction; in a word, the purpose of preprocessing an image is to remove noise in the image, highlight useful information in the image, solve exposure and focusing problems during image acquisition, and improve image contrast.
Step 3: the wind speed sensor detects the wind speed level;
the model of the wind speed sensor is Siemens wind speed sensor QVM 62.1.1, has the characteristics of small volume, simple and direct installation, high measurement accuracy, wide range, good stability, strong anti-interference capability and the like, and has good application in a national power grid;
step 4: judging the wind speed grade, and if the wind speed grade is smaller than 6 grades, performing step 5; if the wind speed level is greater than 6 levels, executing the step 6;
when the wind speed is less than 6 levels, at the moment, the main factor of the risk of foreign matter invasion to the transmission line channel is the phenomenon of nonstandard construction of the line channel; the foreign matter invasion is in a relatively static environment, so that the examination of an image recognition system is small, at the moment, whether the foreign matter invasion condition exists can be judged through the image similarity comparison, and the running similarity comparison program has low requirements on hardware, so that the processor of the camera can be used for running the similarity comparison program, the step of uploading video data to a data transmission of a server is avoided, and the energy is saved; when the wind speed is greater than 6 levels, the transmission line and the surrounding environment, particularly surrounding trees, are in a dynamic process, the foreign matter invasion is in the dynamic environment, and the result error obtained by comparing the pattern similarity is larger, so that whether the foreign matter invasion exists is intelligently judged through an image recognition model, and the image recognition model needs to run in a large processor, and therefore, video data need to be uploaded to a server in the dynamic environment to judge the foreign matter invasion;
step 5: converting the video image subjected to preprocessing operation into a video frame picture by using a processor of a video camera, and performing similarity calculation on the video frame picture at the current moment and the video frame picture at the previous 10s moment; judging the invasion condition of foreign matters;
specifically, the similarity calculation is performed through a perceptual hash algorithm, the perceptual hash algorithm can generate a fingerprint character string for each picture, and then fingerprints are compared to judge the similarity of the two pictures;
further, the step of similarity determination includes:
(1) And reducing the size of the video frame picture. The pictures to be compared are reduced to a size of 10 x 10, and the scaling is because the resolution of original pictures is generally very high, the number of pixels is too large, and the energy consumption during model operation is increased, so that the video frame pictures need to be scaled to be very small;
(2) And carrying out gray scale processing on the video frame picture. The details of the zoomed video frame picture are hidden, but the zoomed video frame picture is not enough, because the zoomed video frame picture is colorful, if RGB values are directly used for contrasting color intensity differences, calculation is still complex, so that an original image is converted into a gray image, and the comparison dimension of a three-dimensional space is reduced to one-dimensional comparison;
(3) Performing discrete cosine transform, wherein the discrete cosine transform can decompose the video frame picture to obtain a transform coefficient matrix of 32x 32, and the principle of the discrete cosine transform is as follows:
wherein F (i, j) is an original video frame picture, F (u, v) is a result after discrete cosine transform, N is a pixel point of the video frame picture, and c (u) and c (v) are compensation coefficients;
(4) Calculating to obtain the average value of the transformation coefficient matrix;
(5) A hash value is calculated. Carrying out hash operation on elements in the transformation coefficient matrix to form a 64-bit binary number, wherein the element value in the matrix is greater than or equal to the average value calculated in the step 4 and is 1, and the element value is 0 when the element value is smaller than the average value calculated in the step 4, and the calculated element value is the fingerprint of the video frame picture;
(6) And (5) comparing the similarity of the video frame pictures through the fingerprints in the step 5.
When converting the preprocessed image into video frame pictures, converting the preprocessed image into a plurality of video frame pictures in a one-second one-frame mode for facilitating similarity calculation of subsequent video frame pictures; of course, if more detailed conversion is performed to improve the monitoring accuracy, for example, the preprocessed image is converted into a plurality of video frame pictures in the form of n frames of one second, n is a natural number greater than or equal to 2;
according to the foregoing description, in this embodiment, the processor of the video machine is used to run the similarity comparison program to perform foreign matter intrusion determination, so as to reduce energy consumption, and in this embodiment, similarity calculation is performed between the video frame picture at the current time and the video frame picture at the previous 10s time; therefore, for the purpose of improving the monitoring precision, the similarity calculation can be performed between the video frame picture at the current moment and the video frame picture at the previous n seconds (n is a positive integer smaller than 10), but the energy consumption is increased, so that the energy consumption condition of the camera is considered when the n value is determined;
for example, the video frame picture at 9 am for 20 minutes and 20s is p1, at this time, the processor of the camera will find the video frame picture p2 at 9 am for 10 seconds, and calculate the similarity between p1 and p2, so as to make the subsequent judgment;
further, the process of judging the foreign matter invasion condition specifically comprises the following steps: if the similarity of the two video frame pictures is less than 80%, judging that foreign matter invasion situation possibly exists, and triggering an alarm device by an online monitoring system to remind an operator on duty;
step 6: uploading the video image subjected to the preprocessing operation to a server, running an image recognition model, and judging the invasion condition of foreign matters;
the image recognition model is specifically an R-CNN image recognition model, wherein the R-CNN is an area image recognition intelligent algorithm architecture based on deep learning, and the architecture comprises three components, namely a feature extraction part, a region to be selected generation part and a target classification part; the feature extraction part and the candidate region generation part function to form a candidate region generation network of R-CNN, and the feature extraction part and the target classification part together form an R-CNN detector; that is, the two modules of the candidate region generation network and the R-CNN detector together form an R-CNN architecture, and the two modules share a feature extraction convolution layer; the feature extraction of the image data is realized through a convolutional neural network, and the convolutional neural network comprises a convolutional layer, a pooling layer and a normalization layer; the convolution layer processes the input image data by utilizing a convolution kernel, so that the expansion of the data dimension can be realized, and the characteristic with higher robustness can be learned; in the deep learning process, after the input original data is convolved, a characteristic response diagram is further obtained by performing nonlinear processing on an activation function; when a convolution kernel is considered to have a certain characteristic, then its convolution result is to input a response corresponding to that characteristic; although the local connection and parameter sharing features in the convolution process can greatly reduce the number of connections between the input and the convolution kernels, the dimension of the feature map is still large and needs to be further reduced in the pooling layer. The space dimension reduction of the pooling layer is realized through downsampling, statistical information is extracted in the downsampling process, the space dimension between layers is reduced, and the calculated amount is simplified;
it is worth emphasizing that in the target classification part, an effective object frame is formed by adopting a non-maximum suppression method, then the characteristics of the area where the effective object frame is positioned are extracted through pooling Chi Chihua treatment, and a prediction function is adopted to predict a target class, namely a boundary frame;
specifically, there are a lot of repeated areas inevitably existing between the various candidate areas formed by the candidate area generation network, and in order to remove these repeated areas, the embodiment removes the repeated areas according to the score of the cross ratio; the intersection ratio refers to the degree of overlap between the target candidate region and the real target candidate region generated by the image recognition algorithm, namely the ratio of the intersection area of the two regions to the union area of the two regions; expressed by a mathematical formula:
in the formula, area (P) represents a target candidate region generated through an image recognition algorithm, and area (G) represents a real target candidate region.
According to another aspect of the present application, the present application further includes an on-line monitoring system for a power transmission line of a power system, wherein: comprising
The camera is used for acquiring the monitoring field video;
the video image preprocessing module is used for preprocessing the acquired video images by the camera;
the wind speed sensor is used for detecting the wind speed grade;
the processor is used for executing the step of the power transmission line on-line monitoring method of the power system;
and the server is used for executing the R-CNN image recognition model to judge the foreign matter invasion condition.
Based on the technical scheme, the method and the system for on-line monitoring of the power transmission line of the power system have the following technical effects:
the method is used for monitoring foreign matter invasion of the power transmission line, the environment is divided into a relatively static environment and a relatively dynamic environment according to the condition of the wind speed of the environment, in the relatively static environment, an image similarity calculation program is operated by a processor of a camera for judging the foreign matter invasion condition, in the relatively dynamic environment, a monitoring video is uploaded to a server, and the foreign matter invasion condition is judged by an image recognition model; in a relatively static environment, the uploading operation of the image and the operation of the image recognition model are not needed, so that the energy consumption is reduced, and the solar energy is better applied to each module of the power transmission line online monitoring system.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a judging scheme of foreign matter invasion in the prior art;
fig. 2 is a flowchart of an online monitoring method for a power transmission line of a power system according to an embodiment of the present application;
fig. 3 is a schematic diagram of a high-pixel monitoring and snapshot integrated camera according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings of the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The concepts related to the present application will be described with reference to the accompanying drawings. It should be noted that the following descriptions of the concepts are only for making the content of the present application easier to understand, and do not represent a limitation on the protection scope of the present application.
Aiming at the defects of the technical scheme, the invention provides an online monitoring method and an online monitoring system for a power transmission line of a power system, which are used for monitoring the invasion danger of foreign matters of the power transmission line; as shown in fig. 2, an on-line monitoring method for a power transmission line of a power system is characterized by comprising the following steps:
step 1: collecting a monitoring field video through a camera of an online monitoring system;
as shown in fig. 3, the camera is a high-pixel monitoring snapshot integrated camera, and monitoring data is collected through the camera; meanwhile, in order to enable devices such as a camera to normally work in an outdoor independent environment of a power transmission line, the embodiment selects a solar cell panel to generate power, and a lithium battery with proper capacity is required to be provided as a standby battery for the solar cell panel to be used for processing monitoring data in consideration of the situations that the solar cell panel cannot generate power at night and generates less power in overcast and rainy days.
Furthermore, in order to enable the camera to have a better monitoring effect, specific consideration should be given to the installation position of the camera, on one hand, the voltage level of the power transmission line is different, and the requirements on the safe distance of the power transmission line are different, so that the safe distance between the camera and the power transmission line needs to be determined, and the installation of the camera cannot influence the normal working operation of the power transmission line; meanwhile, the definition of the image shot by the camera is considered, the installation position of the camera cannot be too high, and the camera is too high in installation, so that the monitored object distance is too far, and the difficulty is increased for image identification.
Step 2: the camera performs preprocessing operation on the collected video images;
due to the influence of field environment, light, weather and the like, some noise is mixed in the collected field video, so that the video image needs to be preprocessed; the preprocessing operation comprises color conversion, image enhancement and image filtering; the color conversion mainly completes the conversion from a color image to a gray image, so as to be beneficial to further image recognition; the image enhancement is to carry out image enhancement on the converted image, adjust the image contrast, highlight the target detail characteristics, and the image filtering is to eliminate noise interference through filtering, so as to improve the accuracy of feature extraction; in a word, the purpose of preprocessing an image is to remove noise in the image, highlight useful information in the image, solve exposure and focusing problems during image acquisition, and improve image contrast.
Step 3: the wind speed sensor detects the wind speed level;
the model of the wind speed sensor is Siemens wind speed sensor QVM 62.1.1, has the characteristics of small volume, simple and direct installation, high measurement accuracy, wide range, good stability, strong anti-interference capability and the like, and has good application in a national power grid;
step 4: judging the wind speed grade, and if the wind speed grade is smaller than 6 grades, performing step 5; if the wind speed level is greater than 6 levels, executing the step 6;
when the wind speed is less than 6 levels, at the moment, the main factor of the risk of foreign matter invasion to the transmission line channel is the phenomenon of nonstandard construction of the line channel; the foreign matter invasion is in a relatively static environment, so that the examination of an image recognition system is small, at the moment, whether the foreign matter invasion condition exists can be judged through the image similarity comparison, and the running similarity comparison program has low requirements on hardware, so that the processor of the camera can be used for running the similarity comparison program, the step of uploading video data to a data transmission of a server is avoided, and the energy is saved; when the wind speed is greater than 6 levels, the transmission line and the surrounding environment, particularly surrounding trees, are in a dynamic process, the foreign matter invasion is in the dynamic environment, and the result error obtained by comparing the pattern similarity is larger, so that whether the foreign matter invasion exists is intelligently judged through an image recognition model, and the image recognition model needs to run in a large processor, and therefore, video data need to be uploaded to a server in the dynamic environment to judge the foreign matter invasion;
step 5: converting the video image subjected to preprocessing operation into a video frame picture by using a processor of a video camera, and performing similarity calculation on the video frame picture at the current moment and the video frame picture at the previous 10s moment; judging the invasion condition of foreign matters;
specifically, the similarity calculation is performed through a perceptual hash algorithm, the perceptual hash algorithm can generate a fingerprint character string for each picture, and then fingerprints are compared to judge the similarity of the two pictures;
further, the step of similarity determination includes:
(1) And reducing the size of the video frame picture. The pictures to be compared are reduced to a size of 10 x 10, and the scaling is because the resolution of original pictures is generally very high, the number of pixels is too large, and the energy consumption during model operation is increased, so that the video frame pictures need to be scaled to be very small;
(2) And carrying out gray scale processing on the video frame picture. The details of the zoomed video frame picture are hidden, but the zoomed video frame picture is not enough, because the zoomed video frame picture is colorful, if RGB values are directly used for contrasting color intensity differences, calculation is still complex, so that an original image is converted into a gray image, and the comparison dimension of a three-dimensional space is reduced to one-dimensional comparison;
(3) Performing discrete cosine transform, wherein the discrete cosine transform can decompose the video frame picture to obtain a transform coefficient matrix of 32x 32, and the principle of the discrete cosine transform is as follows:
wherein F (i, j) is an original video frame picture, F (u, v) is a result after discrete cosine transform, N is a pixel point of the video frame picture, and c (u) and c (v) are compensation coefficients;
(4) Calculating to obtain the average value of the transformation coefficient matrix;
(5) A hash value is calculated. Carrying out hash operation on elements in the transformation coefficient matrix to form a 64-bit binary number, wherein the element value in the matrix is greater than or equal to the average value calculated in the step 4 and is 1, and the element value is 0 when the element value is smaller than the average value calculated in the step 4, and the calculated element value is the fingerprint of the video frame picture;
(6) And (5) comparing the similarity of the video frame pictures through the fingerprints in the step 5.
When converting the preprocessed image into video frame pictures, converting the preprocessed image into a plurality of video frame pictures in a one-second one-frame mode for facilitating similarity calculation of subsequent video frame pictures; of course, if more detailed conversion is performed to improve the monitoring accuracy, for example, the preprocessed image is converted into a plurality of video frame pictures in the form of n frames of one second, n is a natural number greater than or equal to 2;
according to the foregoing description, in this embodiment, the processor of the video machine is used to run the similarity comparison program to perform foreign matter intrusion determination, so as to reduce energy consumption, and in this embodiment, similarity calculation is performed between the video frame picture at the current time and the video frame picture at the previous 10s time; therefore, for the purpose of improving the monitoring precision, the similarity calculation can be performed between the video frame picture at the current moment and the video frame picture at the previous n seconds (n is a positive integer smaller than 10), but the energy consumption is increased, so that the energy consumption condition of the camera is considered when the n value is determined;
for example, the video frame picture at 9 am for 20 minutes and 20s is p1, at this time, the processor of the camera will find the video frame picture p2 at 9 am for 10 seconds, and calculate the similarity between p1 and p2, so as to make the subsequent judgment;
further, the process of judging the foreign matter invasion condition specifically comprises the following steps: if the similarity of the two video frame pictures is less than 80%, judging that foreign matter invasion situation possibly exists, and triggering an alarm device by an online monitoring system to remind an operator on duty;
step 6: uploading the video image subjected to the preprocessing operation to a server, running an image recognition model, and judging the invasion condition of foreign matters;
the image recognition model is specifically an R-CNN image recognition model, wherein the R-CNN is an area image recognition intelligent algorithm architecture based on deep learning, and the architecture comprises three components, namely a feature extraction part, a region to be selected generation part and a target classification part; the feature extraction part and the candidate region generation part function to form a candidate region generation network of R-CNN, and the feature extraction part and the target classification part together form an R-CNN detector; that is, the two modules of the candidate region generation network and the R-CNN detector together form an R-CNN architecture, and the two modules share a feature extraction convolution layer; the feature extraction of the image data is realized through a convolutional neural network, and the convolutional neural network comprises a convolutional layer, a pooling layer and a normalization layer; the convolution layer processes the input image data by utilizing a convolution kernel, so that the expansion of the data dimension can be realized, and the characteristic with higher robustness can be learned; in the deep learning process, after the input original data is convolved, a characteristic response diagram is further obtained by performing nonlinear processing on an activation function; when a convolution kernel is considered to have a certain characteristic, then its convolution result is to input a response corresponding to that characteristic; although the local connection and parameter sharing features in the convolution process can greatly reduce the number of connections between the input and the convolution kernels, the dimension of the feature map is still large and needs to be further reduced in the pooling layer. The space dimension reduction of the pooling layer is realized through downsampling, statistical information is extracted in the downsampling process, the space dimension between layers is reduced, and the calculated amount is simplified;
it is worth emphasizing that in the target classification part, an effective object frame is formed by adopting a non-maximum suppression method, then the characteristics of the area where the effective object frame is positioned are extracted through pooling Chi Chihua treatment, and a prediction function is adopted to predict a target class, namely a boundary frame;
specifically, there are a lot of repeated areas inevitably existing between the various candidate areas generated by the candidate area generating network, and in order to remove these repeated areas, the embodiment removes the repeated areas according to the score of the cross ratio; the intersection ratio refers to the degree of overlap between the target candidate region and the real target candidate region generated by the image recognition algorithm, namely the ratio of the intersection area of the two regions to the union area of the two regions; expressed by a mathematical formula:
in the formula, area (P) represents a target candidate region generated through an image recognition algorithm, and area (G) represents a real target candidate region.
According to another embodiment of the present application, the present application further includes an online monitoring system for a power transmission line of a power system, and is characterized in that: comprising
The camera is used for acquiring the monitoring field video;
the video image preprocessing module is used for preprocessing the acquired video images by the camera;
the wind speed sensor is used for detecting the wind speed grade;
the processor is used for executing the step of the power transmission line on-line monitoring method of the power system;
and the server is used for executing the R-CNN image recognition model to judge the foreign matter invasion condition.
The above examples and/or embodiments are merely for illustrating the preferred embodiments and/or implementations of the present technology, and are not intended to limit the embodiments and implementations of the present technology in any way, and any person skilled in the art should be able to make some changes or modifications to the embodiments and/or implementations without departing from the scope of the technical means disclosed in the present disclosure, and it should be considered that the embodiments and implementations are substantially the same as the present technology.
Claims (7)
1. An on-line monitoring method for a power transmission line of a power system comprises the following steps:
step 1: collecting a monitoring field video through a camera of an online monitoring system;
step 2: the camera performs preprocessing operation on the collected video images;
step 3: the wind speed sensor detects the wind speed level;
step 4: judging the wind speed grade, and if the wind speed grade is smaller than 6 grades, performing step 5; if the wind speed level is greater than 6 levels, executing the step 6;
step 5: converting the video image subjected to preprocessing operation into a video frame picture by using a processor of a video camera, and performing similarity calculation on the video frame picture at the current moment and the video frame picture at the previous 10s moment; judging the invasion condition of foreign matters;
step 6: uploading the video image subjected to the preprocessing operation to a server, running an image recognition model, and judging the invasion condition of foreign matters;
in the step 4: defining the environment at the moment as a relatively static environment when the wind speed is less than 6 levels, and defining the environment at the moment as a relatively dynamic environment when the wind speed is greater than 6 levels; according to different wind speeds, different methods are selected to judge foreign matter invasion;
the step 5 specifically comprises the following steps: when the video camera is in a relatively static environment, judging whether foreign matter invasion exists or not by utilizing a processor of the video camera through image similarity comparison, and when the video camera is in a dynamic environment, uploading the preprocessed video data to a server for foreign matter invasion judgment;
the similarity comparison and judgment steps are as follows:
(1) Reducing the size of the video frame picture, and reducing the picture to be compared to a size of 10 x 10;
(2) Gray scale processing is carried out on the video frame picture;
(3) Performing discrete cosine transform, wherein the discrete cosine transform can decompose the video frame picture to obtain a transform coefficient matrix of 32x 32, and the principle of the discrete cosine transform is as follows:
wherein F (i, j) is an original video frame picture, i and j represent pixel coordinates of the original video frame picture, F (u, v) is an outcome after discrete cosine transform, u and v represent pixel coordinates of the outcome after discrete transform, N is a pixel point of the video frame picture, and c (u) and c (v) are compensation coefficients;
(4) Calculating to obtain the average value of the transformation coefficient matrix;
(5) Calculating a hash value, carrying out hash operation on elements in the transformation coefficient matrix to form a 64-bit binary number, wherein the element value in the matrix is greater than or equal to the average value calculated in the step 4 and is 1, and the element value is less than 0, and the calculated number is the fingerprint of the video frame picture;
(6) And (5) comparing the similarity of the video frame pictures through the fingerprints in the step 5.
2. The on-line monitoring method for the power transmission line of the power system according to claim 1, wherein the camera is a high-pixel monitoring and snapshot integrated camera, a solar cell panel is selected for power generation, and a lithium battery is further provided as a standby battery.
3. The method for on-line monitoring of a power system transmission line according to claim 1, wherein the preprocessing operation includes color conversion, image enhancement, image filtering; the color conversion mainly completes the conversion from a color image to a gray image; the image enhancement is to carry out image enhancement on the converted image, adjust the image contrast and highlight the target detail characteristics, and the image filtering is to eliminate noise interference through filtering, so that the accuracy of characteristic extraction is improved.
4. The method for on-line monitoring of a power transmission line of a power system according to claim 1, wherein the wind speed sensor is a siemens wind speed sensor QVM 62.1.1.
5. The method according to claim 1, wherein in step 5, if the similarity of the two video frame pictures is less than 80%, it is determined that a foreign matter invasion condition may exist, and the online monitoring system triggers an alarm device to remind an operator on duty.
6. The on-line monitoring method for the power transmission line of the power system according to claim 1, wherein the image recognition model is an R-CNN image recognition model and comprises three components, namely a feature extraction part, a region to be selected generation part and a target classification part;
forming an effective object frame in the target classification part by adopting a non-maximum value inhibition method, then carrying out pooling Chi Chihua treatment, extracting the characteristics of the area where the effective object frame is positioned, and predicting the target class, namely the boundary frame by adopting a prediction function;
many repeated areas exist among the various candidate areas formed by the candidate area generating network, so that the repeated areas are removed according to the score of the cross-over ratio (IOU); expressed by a mathematical formula:
in the formula, area (P) represents a target candidate region generated through an image recognition algorithm, and area (G) represents a real target candidate region.
7. An on-line monitoring system for a power transmission line of a power system is characterized in that: comprising
The camera is used for acquiring the monitoring field video;
the video image preprocessing module is used for preprocessing the acquired video images by the camera;
the wind speed sensor is used for detecting the wind speed grade;
a processor for performing the steps of the power transmission line on-line monitoring method of a power system of any one of claims 1-6;
and the server is used for executing the R-CNN image recognition model to judge the foreign matter invasion condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211373997.6A CN115601707B (en) | 2022-11-03 | 2022-11-03 | On-line monitoring method and system for power transmission line of power system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211373997.6A CN115601707B (en) | 2022-11-03 | 2022-11-03 | On-line monitoring method and system for power transmission line of power system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115601707A CN115601707A (en) | 2023-01-13 |
| CN115601707B true CN115601707B (en) | 2024-01-23 |
Family
ID=84853183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211373997.6A Active CN115601707B (en) | 2022-11-03 | 2022-11-03 | On-line monitoring method and system for power transmission line of power system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115601707B (en) |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080025706A (en) * | 2008-01-24 | 2008-03-21 | 이여한 | Forest fire prevention auto system and structure |
| CN103366516A (en) * | 2013-07-30 | 2013-10-23 | 陈勃闻 | Intelligent power transmission line monitoring system and method |
| CN103647942A (en) * | 2013-11-30 | 2014-03-19 | 山东信通电器有限公司 | Comprehensive transmission line monitoring device with intelligent video damage-by-external-force prevention function |
| CN203761618U (en) * | 2013-11-13 | 2014-08-06 | 湖南赛道科技有限公司 | Intelligent street lamp control system based on traffic flow |
| CN105915846A (en) * | 2016-04-26 | 2016-08-31 | 成都通甲优博科技有限责任公司 | Monocular and binocular multiplexed invading object monitoring method and system |
| CN106056821A (en) * | 2016-08-13 | 2016-10-26 | 哈尔滨理工大学 | Power-transmission-line foreign-matter-invasion intelligent-video on-line monitoring evaluation system |
| CN205862497U (en) * | 2016-08-13 | 2017-01-04 | 哈尔滨理工大学 | Electric line foreign matter intrusion intelligent video on-line monitoring assessment system |
| CN107749142A (en) * | 2017-11-21 | 2018-03-02 | 海南电网有限责任公司电力科学研究院 | A kind of anti-mountain fire early warning system of transmission line of electricity and its method for early warning |
| WO2018130016A1 (en) * | 2017-01-10 | 2018-07-19 | 哈尔滨工业大学深圳研究生院 | Parking detection method and device based on monitoring video |
| WO2019109524A1 (en) * | 2017-12-07 | 2019-06-13 | 平安科技（深圳）有限公司 | Foreign object detection method, application server, and computer readable storage medium |
| CN209823793U (en) * | 2019-05-25 | 2019-12-20 | 许昌初心智能电气科技有限公司 | High-voltage line on-line monitoring device |
| CN110769195A (en) * | 2019-10-14 | 2020-02-07 | 国网河北省电力有限公司衡水供电分公司 | Intelligent monitoring and recognizing system for violation of regulations on power transmission line construction site |
| CN211509200U (en) * | 2020-04-09 | 2020-09-15 | 段宏 | Transmission line prevents outer broken monitored control system based on artificial intelligence technique |
| CN111754714A (en) * | 2020-07-08 | 2020-10-09 | 南阳师范学院 | Security monitoring system and monitoring method thereof |
| CN111830070A (en) * | 2020-08-10 | 2020-10-27 | 中海石油气电集团有限责任公司 | Automatic defect identification and judgment system and method based on edge calculation |
| KR102309077B1 (en) * | 2021-05-27 | 2021-10-06 | 주식회사 부력에너지 | Solar power generation system and method applying sensor-based safety diagnosis technology |
| CN214748120U (en) * | 2021-05-06 | 2021-11-16 | 筠连县福强农业科技有限公司 | Fruit tree growth cycle monitoring and counting system |
| WO2022052475A1 (en) * | 2020-09-14 | 2022-03-17 | 上海商汤智能科技有限公司 | Image capture processing method, apparatus and device, storage medium, and program product |
| CN114627388A (en) * | 2022-03-23 | 2022-06-14 | 南方电网数字电网研究院有限公司 | Power transmission line foreign matter detection equipment and foreign matter detection method thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11455877B2 (en) * | 2021-02-23 | 2022-09-27 | Arash Aharpour | System and method of reducing energy consumption of datalogger devices while maintaining high sampling rate and real time alarm function |
-
2022
- 2022-11-03 CN CN202211373997.6A patent/CN115601707B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080025706A (en) * | 2008-01-24 | 2008-03-21 | 이여한 | Forest fire prevention auto system and structure |
| CN103366516A (en) * | 2013-07-30 | 2013-10-23 | 陈勃闻 | Intelligent power transmission line monitoring system and method |
| CN203761618U (en) * | 2013-11-13 | 2014-08-06 | 湖南赛道科技有限公司 | Intelligent street lamp control system based on traffic flow |
| CN103647942A (en) * | 2013-11-30 | 2014-03-19 | 山东信通电器有限公司 | Comprehensive transmission line monitoring device with intelligent video damage-by-external-force prevention function |
| CN105915846A (en) * | 2016-04-26 | 2016-08-31 | 成都通甲优博科技有限责任公司 | Monocular and binocular multiplexed invading object monitoring method and system |
| CN106056821A (en) * | 2016-08-13 | 2016-10-26 | 哈尔滨理工大学 | Power-transmission-line foreign-matter-invasion intelligent-video on-line monitoring evaluation system |
| CN205862497U (en) * | 2016-08-13 | 2017-01-04 | 哈尔滨理工大学 | Electric line foreign matter intrusion intelligent video on-line monitoring assessment system |
| WO2018130016A1 (en) * | 2017-01-10 | 2018-07-19 | 哈尔滨工业大学深圳研究生院 | Parking detection method and device based on monitoring video |
| CN107749142A (en) * | 2017-11-21 | 2018-03-02 | 海南电网有限责任公司电力科学研究院 | A kind of anti-mountain fire early warning system of transmission line of electricity and its method for early warning |
| WO2019109524A1 (en) * | 2017-12-07 | 2019-06-13 | 平安科技（深圳）有限公司 | Foreign object detection method, application server, and computer readable storage medium |
| CN209823793U (en) * | 2019-05-25 | 2019-12-20 | 许昌初心智能电气科技有限公司 | High-voltage line on-line monitoring device |
| CN110769195A (en) * | 2019-10-14 | 2020-02-07 | 国网河北省电力有限公司衡水供电分公司 | Intelligent monitoring and recognizing system for violation of regulations on power transmission line construction site |
| CN211509200U (en) * | 2020-04-09 | 2020-09-15 | 段宏 | Transmission line prevents outer broken monitored control system based on artificial intelligence technique |
| CN111754714A (en) * | 2020-07-08 | 2020-10-09 | 南阳师范学院 | Security monitoring system and monitoring method thereof |
| CN111830070A (en) * | 2020-08-10 | 2020-10-27 | 中海石油气电集团有限责任公司 | Automatic defect identification and judgment system and method based on edge calculation |
| WO2022052475A1 (en) * | 2020-09-14 | 2022-03-17 | 上海商汤智能科技有限公司 | Image capture processing method, apparatus and device, storage medium, and program product |
| CN214748120U (en) * | 2021-05-06 | 2021-11-16 | 筠连县福强农业科技有限公司 | Fruit tree growth cycle monitoring and counting system |
| KR102309077B1 (en) * | 2021-05-27 | 2021-10-06 | 주식회사 부력에너지 | Solar power generation system and method applying sensor-based safety diagnosis technology |
| CN114627388A (en) * | 2022-03-23 | 2022-06-14 | 南方电网数字电网研究院有限公司 | Power transmission line foreign matter detection equipment and foreign matter detection method thereof |
Non-Patent Citations (2)
| Title |
|---|
| 基于大数据的隧道通风智能控制***;王坚;刘晓娜;孟引鹏;;科技风(25);36-37 * |
| 基于深度强化学习与图像智能识别的输电线路在线监测***;叶俊健;邓伟锋;徐常志;赵丽娜;;工业技术创新(03);76-79 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115601707A (en) | 2023-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110232380B (en) | Fire night scene restoration method based on Mask R-CNN neural network | |
| CN111583198A (en) | Insulator picture defect detection method combining FasterR-CNN + ResNet101+ FPN | |
| CN112950576B (en) | Power transmission line defect intelligent identification method and system based on deep learning | |
| CN115620239B (en) | Point cloud and video combined power transmission line online monitoring method and system | |
| CN111047598B (en) | Deep learning-based ultraviolet discharge light spot segmentation method and device for power transmission and transformation equipment | |
| CN110728212B (en) | Road well lid monitoring device and monitoring method based on computer vision | |
| CN118038153A (en) | Method, device, equipment and medium for identifying external damage prevention of distribution overhead line | |
| CN114325856A (en) | Power transmission line foreign matter monitoring method based on edge calculation | |
| CN115601707B (en) | On-line monitoring method and system for power transmission line of power system | |
| CN117036825A (en) | Solar cell panel detection method, medium and system | |
| CN115240394B (en) | Method and system for monitoring and early warning water level of accident oil pool of transformer substation | |
| CN116612065A (en) | YOLO v 5-based intelligent identification method for transmission line inspection image defects | |
| CN113689399B (en) | Remote sensing image processing method and system for power grid identification | |
| CN115760616A (en) | Human body point cloud repairing method and device, electronic equipment and storage medium | |
| CN114926774A (en) | Method and device for identifying photovoltaic module fire based on video image | |
| CN114926424A (en) | Detection method and electronic equipment | |
| Zhou et al. | Water leakage detection and its application of turbine floor equipment based on MaskRCNN | |
| CN111524158A (en) | Method for detecting foreground target in complex scene of hydraulic engineering | |
| CN116664484A (en) | Scene-adaptive parameterized photovoltaic module loosening detection method and system | |
| CN116071656B (en) | Intelligent alarm method and system for infrared image ponding detection of underground transformer substation | |
| Gang et al. | Research on key technology of infrared detection of power equipment | |
| Ma | Face Recognition of Intelligent Building based on Super-Resolution Reconstruction of Visual Image | |
| CN117710297A (en) | High-precision detection method and system for power distribution network insulator | |
| CN118038021A (en) | Transformer substation operation site foreign matter intrusion detection method based on improvement yolov4 | |
| CN118071821A (en) | Method and device for determining position information, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |