CN113297940A - High-speed rail pantograph working state detection method - Google Patents
High-speed rail pantograph working state detection method Download PDFInfo
- Publication number
- CN113297940A CN113297940A CN202110534957.4A CN202110534957A CN113297940A CN 113297940 A CN113297940 A CN 113297940A CN 202110534957 A CN202110534957 A CN 202110534957A CN 113297940 A CN113297940 A CN 113297940A
- Authority
- CN
- China
- Prior art keywords
- pantograph
- image
- alarm
- speed rail
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/182—Network patterns, e.g. roads or rivers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Current-Collector Devices For Electrically Propelled Vehicles (AREA)
Abstract
The invention discloses a method for detecting the working state of a high-speed rail pantograph, which comprises the following steps: s1, collecting the pantograph position image by using a high-speed camera, creating a real-time image thread, and reading each frame of image data frame by frame; s2, converting the read image into an RGB format, and screening out a clear image; s3, sending the RGB format image into an image analysis module for recognition, recognizing the pantograph, judging the size defect of the pantograph, calculating whether the contact gap distance between the pantograph and the high-voltage transmission line exceeds a threshold value, and if the contact gap distance exceeds the threshold value, judging that a fault exists; and S4, obtaining an analysis result, and alarming according to the abnormal state and the fault of the pantograph. The invention solves the problems that the existing pantograph state of the high-speed rail cannot be accurately detected in real time and the fault cannot be reported in time.
Description
Technical Field
The invention relates to the technical field of railway transportation safety, in particular to a method for detecting the working state of a high-speed rail pantograph.
Background
The high-speed rail has long driving mileage and high speed, in the high-speed driving process, the high-voltage wire above the roof needs to transmit electric energy to the vehicle body through the pantograph, the pantograph is in continuous contact friction with the power transmission line, meanwhile, when the high-speed rail is driven in a track-changing mode, the pantograph needs to be switched to the corresponding power transmission line, the friction inevitably generates abrasion and deformation, and in order to guarantee the safe driving of the high-speed rail, the state of the pantograph needs to be monitored in real time
In order to ensure that the pantograph of the high-speed rail is not worn off, the pantograph is inspected and replaced at regular time at present, but the inspection and replacement are carried out when the high-speed rail stops, the pantograph cannot be replaced when the high-speed rail runs, only the pantograph material, the pantograph structure and the special graphite material which is improved in a real-time monitoring layer are improved, the two sides of the high-speed rail are provided with the towers and are crossed at intervals, so that the high-voltage wire is connected to be actually snakelike, namely Z-shaped, the high-voltage wire has the advantages that the same position of the pantograph cannot contact with the electric wire for a long time, the electric wire moves left and right on the pantograph, the temperature of the same contact point cannot be overhigh, excessive abrasion and naked fire are caused, the abrasion area of the abrasion-resistant material is increased, the abrasion-able time is prolonged, meanwhile, the monitoring camera is arranged at the top of the vehicle, and a special high-speed rail safety worker is arranged to check the real-time state of the pantograph through human eyes, thereby playing the real-time control of the pantograph state. The graphite of the pantograph is continuously worn and thinned, every time the lower support rises by one millimeter, the pantograph runs by thousands of meters again for a high-speed rail, the pantograph cannot be worn off as long as the graphite is periodically replaced, and the contact between the pantograph and a wire is unstable unless the height of a rail fluctuates, so that the pantograph is damaged by discharge. The prior art scheme mainly depends on experience and manual work to realize the operation, because the pantograph state is related to the operation safety of a high-speed rail, the potential hazards are discovered in time at the first time, and the prior art scheme cannot realize quick problem discovery and report.
Disclosure of Invention
Therefore, the invention provides a method for detecting the working state of a high-speed rail pantograph, which aims to solve the problems that the state of the existing high-speed rail pantograph cannot be accurately detected in real time and faults cannot be reported in time.
In order to achieve the above purpose, the invention provides the following technical scheme:
the invention discloses a method for detecting the working state of a high-speed rail pantograph, which comprises the following steps:
s1, collecting the pantograph position image by using a high-speed camera, creating a real-time image thread, and reading each frame of image data frame by frame;
s2, converting the read image into an RGB format, and screening out a clear image;
s3, sending the RGB format image into an image analysis module for recognition, recognizing the pantograph, judging the size defect of the pantograph, calculating whether the contact gap distance between the pantograph and the high-voltage transmission line exceeds a threshold value, and if the contact gap distance exceeds the threshold value, judging that a fault exists;
and S4, obtaining an analysis result, and alarming according to the abnormal state and the fault of the pantograph.
Further, the high-speed camera is installed on the top of the high-speed rail, inclines upwards at an angle of 45 degrees, and shoots images of the pantograph and the power transmission line.
Further, the image shot by the high-speed camera is converted into an RGB format through image processing equipment, the image processing equipment judges the image quality, filters out images with poor quality, and performs image correction and image enhancement on the images with good image quality.
Further, the image analysis module performs region segmentation on the corrected and enhanced RGB format image, obtains a feature vector for each region by using a convolutional neural network, extracts features corresponding to each RoI from the full-image features by using a RoI posing Layer, predicts the position of the pantograph from the target image through a full-connection Layer of 68 dimensions and 180 dimensions, respectively, and calculates the center point position thereof.
Furthermore, the image analysis module firstly obtains an image area of the pantograph as a reference for an overlapping area of the power transmission line and the pantograph, performs prepositioning on the position of the power transmission line, extracts the positioned interested area, performs local high-resolution fine detection, and performs segmentation detection on the area.
Further, the image analysis module identifies the position of the pantograph from the target image, judges the form of the pantograph through the identification model, and judges whether a fault exists according to the closeness of the contact point of the pantograph and the power transmission line.
Furthermore, a large amount of standard data are input into the recognition model in advance for training, data of the normal form of the pantograph and data of the closeness of the contact point of the pantograph and the power transmission line are imported into the recognition model, and the recognition model compares the received real-time image data with the standard data to judge whether the abnormality occurs.
Further, the image analysis module analyzes the obtained result to judge that the pantograph and the power transmission line are abnormal, and then sends out an alarm signal through the alarm module, and stores alarm pictures and videos for identifying faults into a specified directory.
Further, the alarm data generated by the alarm module is persisted to a database, if the storage fails, the stored alarm picture and video information are deleted, and if the storage of the alarm information is successful, the corresponding alarm data packet in the alarm data analysis queue is deleted.
Furthermore, the alarm module sends alarm information through sound and light, and meanwhile, the corresponding alarm information is pushed by the high-speed rail dispatching platform and the mobile terminal corresponding to the manager and is processed in time.
The invention has the following advantages:
the invention discloses a method for detecting the working state of a high-speed rail pantograph, which comprises the steps of installing a high-speed camera behind the top pantograph of a high-speed rail, collecting an image of the position of the pantograph in real time, decoding the image, sending the decoded image to an analysis algorithm module running on a calculation unit for recognition, analyzing whether the pantograph is deformed or not, judging whether the pantograph and a high-voltage wire are in good contact or not, and immediately notifying an equipment state detection and alarm module once the deformation of the pantograph exceeds a threshold value or the contact gap between the pantograph and the high-voltage wire exceeds the threshold value. The real-time automatic detection of the running states of the pantograph and the transmission line is realized, the fault is reported in time, the labor intensity of workers is reduced, and the fault is processed in time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
The structures, ratios, sizes, and the like shown in the present specification are only used for matching with the contents disclosed in the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions that the present invention can be implemented, so that the present invention has no technical significance, and any structural modifications, changes in the ratio relationship, or adjustments of the sizes, without affecting the effects and the achievable by the present invention, should still fall within the range that the technical contents disclosed in the present invention can cover.
Fig. 1 is a flowchart of a method for detecting a working state of a high-speed rail pantograph according to an embodiment of the present invention;
Detailed Description
The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
The embodiment discloses a method for detecting the working state of a high-speed rail pantograph, which comprises the following steps:
s1, collecting the pantograph position image by using a high-speed camera, creating a real-time image thread, and reading each frame of image data frame by frame;
s2, converting the read image into an RGB format, and screening out a clear image;
s3, sending the RGB format image into an image analysis module for recognition, recognizing the pantograph, judging the size defect of the pantograph, calculating whether the contact gap distance between the pantograph and the high-voltage transmission line exceeds a threshold value, and if the contact gap distance exceeds the threshold value, judging that a fault exists;
and S4, obtaining an analysis result, and alarming according to the abnormal state and the fault of the pantograph.
The high-speed camera is arranged at the top of the high-speed rail, inclines upwards at an angle of 45 degrees and shoots images of the pantograph and the power transmission line; the image shot by the high-speed camera is converted into an RGB format through image processing equipment, the image processing equipment judges the image quality, filters out images with poor quality, and performs image correction and image enhancement on the images with good image quality. In the running process of a high-speed rail, practical environment interference sources such as backlight, insufficient light, complex and variable background and the like exist, so that the situation that collected images are not suitable for algorithm judgment exists, in order to improve the identification accuracy and reduce the misjudgment rate, images with poor quality need to be filtered, the quality of the collected images is judged frame by introducing an image quality evaluation algorithm, and the images with good quality are screened out. Because the real-time data of the pantograph is collected in a moving state, the data of the high-speed camera is continuously transmitted to the computing unit through the network and is transmitted to the analysis algorithm module, if the data is blocked, the pressure of the computing unit is high, and even the computing unit goes down, so that a high requirement is provided for the processing speed of the analysis algorithm module, and the requirement of real-time processing must be met.
And detecting the pantograph frame by frame in the video image screened out from the image quality module and outputting the position of the pantograph. In this embodiment, an acquired image is subjected to region segmentation, a convolutional neural network is used for each region to obtain a feature vector, a RoI Pooling Layer is used to extract features corresponding to each RoI from the features of the whole image, the features respectively pass through a 68-dimensional and 180-dimensional full connection Layer, the 68-dimensional and 180-dimensional full connection layers are parallel, the former is classified output, the latter is regression output, the position of the pantograph is predicted from the target image, and the center point position of the pantograph is calculated.
The high-speed rail running environment can have the condition of many high voltage transmission lines, but only have one or two high voltage transmission lines and pantograph contact simultaneously, whether high voltage transmission line intersects with the pantograph, need consider the environment that the actual scene is complicated to and the high voltage transmission line adopts the method of degree of depth study for the space visual angle of pantograph, make the algorithm have sufficient generalization.
In order to solve the problems that the high-voltage power transmission line is small in target size and long and thin, the high-voltage power transmission line and a pantograph are easy to overlap, if special processing is not conducted, the characteristics of the high-voltage power transmission line are difficult to extract through a neural network, the high-voltage power transmission line needs to be well perceived in the global state to achieve good positioning, high-voltage power transmission line sub-line regions must be located in the range of the pantograph region, the high-voltage power transmission line sub-line regions must depend on global information of the pantograph position region to achieve good positioning, the image region of the pantograph is obtained first to serve as a reference, pre-positioning is conducted on the position of the power transmission line, the located interested regions are extracted, local high-resolution fine detection is conducted, then segmentation detection is conducted on the regions, and therefore the detection speed is improved. The image analysis module identifies the position of the pantograph from the target image, judges the form of the pantograph through the identification model and judges whether a fault exists according to the closeness of a contact point of the pantograph and the power transmission line.
The identification model is pre-input with a large amount of standard data for training, data of the normal form of the pantograph and data of the contact point tightness of the pantograph and the power transmission line are imported into the identification model, and the identification model compares the received real-time image data with the standard data to judge whether the abnormality occurs. And the image analysis module analyzes the obtained result to judge that the pantograph and the power transmission line are abnormal, sends out an alarm signal through the alarm module and stores alarm pictures and videos for identifying faults into a specified directory.
And the alarm data generated by the alarm module is persisted to a database, if the storage fails, the stored alarm picture and video information are deleted, and if the storage of the alarm information succeeds, the corresponding alarm data packet in the alarm data analysis queue is deleted. Alarm module sends alarm information through sound, light, and corresponding alarm information is pushed with the mobile terminal that corresponds the managers to the high-speed railway dispatch platform simultaneously, in time handles.
After the mobile terminals of the railway dispatching platform and the managers receive the alarm information, specific fault types are judged by combining the pictures and the video information in the alarm data, and repair preparation work is carried out, so that the maintenance efficiency can be improved.
According to the detection method for the working state of the high-speed rail pantograph, the high-speed camera is installed at the rear part of the high-speed rail top pantograph, images of the position of the pantograph are collected in real time, the images are decoded and then are sent to the analysis algorithm module running on the computing unit for recognition, whether the pantograph is deformed or not is analyzed, whether the contact between the pantograph and a high-voltage wire is good or not is judged, and once the fact that the deformation of the pantograph exceeds a threshold value or the contact gap between the pantograph and the high-voltage wire exceeds the threshold value is found, the device state detection and alarm module is immediately notified. The real-time automatic detection of the running states of the pantograph and the transmission line is realized, the fault is reported in time, the labor intensity of workers is reduced, and the fault is processed in time.
Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (10)
1. A method for detecting the working state of a high-speed rail pantograph is characterized by comprising the following steps:
s1, collecting the pantograph position image by using a high-speed camera, creating a real-time image thread, and reading each frame of image data frame by frame;
s2, converting the read image into an RGB format, and screening out a clear image;
s3, sending the RGB format image into an image analysis module for recognition, recognizing the pantograph, judging the size defect of the pantograph, calculating whether the contact gap distance between the pantograph and the high-voltage transmission line exceeds a threshold value, and if the contact gap distance exceeds the threshold value, judging that a fault exists;
and S4, obtaining an analysis result, and alarming according to the abnormal state and the fault of the pantograph.
2. The method as claimed in claim 1, wherein the high speed camera is installed on the top of the high speed railway, is tilted at an angle of 45 ° upward, and takes an image of the pantograph and the power line.
3. The method as claimed in claim 2, wherein the images captured by the high speed camera are converted into RGB format by an image processing device, the image processing device determines the image quality, filters out poor quality images, and performs image correction and image enhancement on good quality images.
4. The method as claimed in claim 1, wherein the image analysis module performs region segmentation on the corrected and enhanced RGB format image, obtains a feature vector for each region by using a convolutional neural network, extracts features corresponding to each RoI from the full-image features by using a rotolining Layer, respectively passes through full connection layers of 68 dimensions and 180 dimensions, predicts the position of the pantograph from the target image, and calculates the position of the center point of the pantograph.
5. The method as claimed in claim 1, wherein the image analysis module first obtains an image area of the pantograph as a reference for an overlapping area of the power line and the pantograph, pre-positions the position of the power line, extracts a region of interest, performs local high-resolution fine detection, and performs segmentation detection on the region.
6. The method as claimed in claim 4, wherein the image analysis module identifies the position of the pantograph from the target image, determines the shape of the pantograph through the identification model, and determines whether there is a fault according to the closeness of the contact point between the pantograph and the power transmission line.
7. The method as claimed in claim 6, wherein the recognition model is pre-entered with a large amount of standard data for training, and the data of the normal form of the pantograph and the closeness data of the contact point between the pantograph and the power transmission line are imported into the recognition model, and the recognition model compares the received real-time image data with the standard data to determine whether an abnormality occurs.
8. The method as claimed in claim 1, wherein the image analysis module analyzes the result to determine whether the pantograph and the power line are abnormal, and sends out an alarm signal through the alarm module, and stores an alarm picture and a video for identifying the fault in a designated directory.
9. The method for detecting the operating state of the high-speed railway pantograph according to claim 8, wherein the alarm data generated by the alarm module is persisted to a database, and if the warehousing fails, the stored alarm pictures and video information are deleted, and if the warehousing succeeds, the corresponding alarm data packets in the alarm data analysis queue are deleted.
10. The method for detecting the operating condition of the pantograph of the high-speed rail according to claim 8, wherein the alarm module sends alarm information through sound and light, and meanwhile, the corresponding alarm information is pushed by the high-speed rail dispatching platform and a mobile terminal corresponding to a manager and is processed in time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110534957.4A CN113297940A (en) | 2021-05-17 | 2021-05-17 | High-speed rail pantograph working state detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110534957.4A CN113297940A (en) | 2021-05-17 | 2021-05-17 | High-speed rail pantograph working state detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113297940A true CN113297940A (en) | 2021-08-24 |
Family
ID=77322370
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110534957.4A Pending CN113297940A (en) | 2021-05-17 | 2021-05-17 | High-speed rail pantograph working state detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113297940A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113859312A (en) * | 2021-09-30 | 2021-12-31 | 中车青岛四方机车车辆股份有限公司 | Pantograph fault alarm method and device based on vehicle-mounted PHM and rail vehicle |
-
2021
- 2021-05-17 CN CN202110534957.4A patent/CN113297940A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113859312A (en) * | 2021-09-30 | 2021-12-31 | 中车青岛四方机车车辆股份有限公司 | Pantograph fault alarm method and device based on vehicle-mounted PHM and rail vehicle |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111507308B (en) | Transformer substation safety monitoring system and method based on video identification technology | |
CN109842787B (en) | Method and system for monitoring high-altitude parabolic object | |
CN111432179A (en) | Intelligent coal conveying belt inspection system and method based on computer vision | |
CN102842211B (en) | Monitoring and early warning system and monitoring and early warning method for prevention of external force of transmission line based on image recognition | |
CN110127479B (en) | Elevator door switch abnormity detection method based on video analysis | |
CN104966304A (en) | Kalman filtering and nonparametric background model-based multi-target detection tracking method | |
US20230005176A1 (en) | Throwing position acquisition method and apparatus, computer device and storage medium | |
CN111222478A (en) | Construction site safety protection detection method and system | |
KR102207393B1 (en) | Expressway traffic monitoring system of transport information center CCTV using artificial intelligence | |
CN113139482A (en) | Method and device for detecting traffic abnormity | |
CN106454330B (en) | The fuzzy method for detecting abnormality of vision signal | |
Aydın et al. | Particle swarm based arc detection on time series in pantograph-catenary system | |
CN112560816A (en) | Equipment indicator lamp identification method and system based on YOLOv4 | |
Ketcham et al. | The intruder detection system for rapid transit using CCTV surveillance based on histogram shapes | |
CN113297940A (en) | High-speed rail pantograph working state detection method | |
CN114022846A (en) | Anti-collision monitoring method, device, equipment and medium for working vehicle | |
CN117197713A (en) | Extraction method based on digital video monitoring system | |
CN111160321A (en) | Storage battery car goes up terraced detection and early warning system | |
CN115631625A (en) | Smart mine management and control system based on big data platform | |
CN111523386A (en) | Machine vision-based high-speed railway platform door monitoring and protecting method and system | |
CN109488383B (en) | Intelligent video analysis system and application method thereof | |
CN114179866A (en) | Track turnout opening direction safety detection system and detection method | |
CN112232235B (en) | Intelligent factory remote monitoring method and system based on 5G sum algorithm | |
CN113222548A (en) | Community property eagle eye monitoring system and use method | |
CN117035669A (en) | Enterprise safety production management method and system based on artificial intelligence |
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 |