CN117173183B - Railway wagon fault detection method based on artificial intelligence - Google Patents
Railway wagon fault detection method based on artificial intelligence Download PDFInfo
- Publication number
- CN117173183B CN117173183B CN202311452020.8A CN202311452020A CN117173183B CN 117173183 B CN117173183 B CN 117173183B CN 202311452020 A CN202311452020 A CN 202311452020A CN 117173183 B CN117173183 B CN 117173183B
- Authority
- CN
- China
- Prior art keywords
- damage
- railway
- area
- calculating
- similarity
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 19
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 14
- 230000006378 damage Effects 0.000 claims abstract description 146
- 239000002184 metal Substances 0.000 claims abstract description 54
- 229910052751 metal Inorganic materials 0.000 claims abstract description 54
- 239000011159 matrix material Substances 0.000 claims abstract description 43
- 238000012423 maintenance Methods 0.000 claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000003708 edge detection Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000003902 lesion Effects 0.000 claims description 6
- 230000004044 response Effects 0.000 claims description 6
- 208000031074 Reinjury Diseases 0.000 claims description 5
- 208000027418 Wounds and injury Diseases 0.000 claims description 4
- 208000014674 injury Diseases 0.000 claims description 4
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims 2
- SAZUGELZHZOXHB-UHFFFAOYSA-N acecarbromal Chemical compound CCC(Br)(CC)C(=O)NC(=O)NC(C)=O SAZUGELZHZOXHB-UHFFFAOYSA-N 0.000 claims 2
- 229910052742 iron Inorganic materials 0.000 claims 1
- 238000004901 spalling Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 description 6
- 238000012512 characterization method Methods 0.000 description 3
- 239000000306 component Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000004227 thermal cracking Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention relates to the field of railway wagon fault detection, in particular to an artificial intelligence-based railway wagon fault detection method. The method comprises the following steps: the method comprises the steps of obtaining a metal falling block peeling damage area and a crack area by carrying out edge detection on a wheel tread gray level image of a railway wagon; respectively constructing a gray level co-occurrence matrix for the metal falling block peeling damage area and the crack area to obtain characteristic information; acquiring position information of a metal falling block peeling damage area and position information of a crack area; constructing a comprehensive damage characteristic matrix, and calculating the similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons; and calculating a first damage interval and a second damage interval, and generating and sending all railway wagon information and maintenance instructions which need maintenance. By the technical scheme, the railway wagon needing to be overhauled can be found more accurately and efficiently.
Description
Technical Field
The present invention relates generally to the field of railway wagon fault detection. More particularly, the invention relates to an artificial intelligence-based railway wagon fault detection method.
Background
The rail wagon is a core component of a rail wagon running part and is the most difficult component with the most faults and maintenance, and the faults mainly comprise two major types of wheel tread stripping and polygonal abrasion of the rail wagon. Tread stripping refers to the phenomena of metal chipping damage and fish scale or crazing-like thermal cracking areas exhibited on the tread circumference or part of the circumference during operation of the wheels of a railway wagon due to braking action or wheel rail rolling contact fatigue action.
The prior Chinese patent train wheel set monitoring system and train wheel set monitoring method with the issued publication number of CN111071291B discloses an independent train monitoring system and train monitoring method, wherein the train wheel set is abnormally monitored in a vibration monitoring mode.
The monitoring of the fault book of the single train in the prior art lacks certainty, and similar vibration caused by non-train faults can be detected, so that the detection accuracy and the detection efficiency are low.
Disclosure of Invention
In order to solve one or more of the above technical problems, the present invention proposes an edge detection method for a gray image of a tread of a railway wagon, comprising the steps of: the method comprises the steps of obtaining a metal falling block peeling damage area and a crack area by carrying out edge detection on a wheel tread gray level image of a railway wagon; respectively constructing a gray level co-occurrence matrix for the metal falling block peeling damage area and the crack area to obtain characteristic information; acquiring position information of a metal falling block peeling damage area and position information of a crack area; constructing a comprehensive damage characteristic matrix, and calculating the similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons; and calculating a first damage interval and a second damage interval, and generating and sending all railway wagon information and maintenance instructions which need maintenance. To this end, the present invention provides solutions in various aspects as follows.
In one embodiment, the metal falling block peeling damage area and the crack area are obtained by performing edge detection on the wheel tread gray level image; constructing a first gray level co-occurrence matrix according to the metal falling and peeling damage area to obtain damage characteristic information of the metal falling and peeling damage area, and constructing a second gray level co-occurrence matrix according to the crack area to obtain texture characteristic information of the crack area; calculating first position information and second position information, and constructing a comprehensive damage characteristic matrix related to the first position information, the second position information, damage characteristic information and texture characteristic information, wherein the first position information is the position information of a metal falling block peeling damage area, and the second position information is the position information of a crack area; randomly extracting two rows of railway wagons, calculating the similarity of the wheel tread comprehensive damage areas of the two rows of railway wagons, generating judging information of the same damage of the two rows of railway wagons in response to the similarity being larger than a preset threshold, and calculating a first damage interval, wherein the first damage interval is an intersection of the two rows of railway wagons with the same damage; calculating the similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons passing through the first damage area in a preset time period; and reserving the railway wagons with the similarity larger than a preset threshold value, obtaining a running route intersection, generating a second damage interval, and generating and sending all railway wagons information and maintenance instructions which need maintenance.
In one embodiment, the similarity of the wheel tread comprehensive damage areas of the two trains of railway trucks satisfies the relationship:
wherein,for the similarity of the comprehensive lesion areas, +.>Similarity of peeling damaged area for metal falling block, < >>For the similarity of crack areas, +.>Feature matrix of metal falling block peeling damage area of first-row railway wagon>Feature matrix of metal falling block peeling damage area of second-row railway wagon>Characteristic matrix for the crack zone of the first train rail wagon,>and the feature matrix is the feature matrix of the crack area of the second train of railway trucks.
In one embodiment, calculating the first location information, the second location information comprises the steps of: calculating a damage vertical distance, and taking the center of any one of the metal falling block peeling damage areas or the center of any one of the crack areas as a starting point and taking the inner side boundary of the railway wagon as an end point to obtain the damage vertical distance vertical to the inner side boundary of the railway wagon; calculating the width of the railway wagon according to the number of gray image pixels of the tread of the wheel; calculating first position information, wherein the first position information is the ratio of the damage vertical distance of a metal falling block peeling damage area to the width of the railway wagon; and calculating second position information, wherein the second position information is the ratio of the damage vertical distance of the crack area to the width of the railway wagon.
In one embodiment, the obtaining damage characteristic information of the metal chipping damage region and obtaining texture characteristic information of the crack region includes the steps of: constructing a first gray level co-occurrence matrix according to the metal falling block peeling damage area; calculating damage characteristic information of a metal falling block peeling damage area, wherein the damage characteristic information is the confusion degree of the first gray level co-occurrence matrix; constructing a second gray level co-occurrence matrix according to the crack area; and calculating texture feature information of the crack area, wherein the texture feature information is the energy feature of the second gray level co-occurrence matrix.
In one embodiment, generating the second injury interval comprises the steps of: generating judging information that the two railway wagons are worn with the similarity being greater than a preset threshold value when passing through the same track in response to the same damage to the wheel treads of the two railway wagons; calculating a first damage interval, wherein the calculation formula is as follows:
wherein,for the first injury zone, ++>For the driving route of the railway wagon of column A, < > for the railway wagon of column A>Is the driving route of the railway wagon of the B column.
Based on the first damaged section, obtaining wheel tread gray level images of all railway wagons of a driving route passing through the first damaged section according to a camera, calculating the similarity of wheel tread comprehensive damaged areas of any two railway wagons in the wheel tread gray level images of all railway wagons, and reserving a driving route set of all railway wagons with the similarity larger than a preset threshold valueAnd calculating a second damage interval, wherein the calculation formula is as follows:
wherein,for the driving route of the first train of rail wagons with similarity greater than the preset threshold value, +.>Is of similarity greater thanThe driving route of the second train of railway wagons with preset threshold value +.>The>Train route of rail wagon->Is the second injury interval.
In one embodiment, the method further comprises the steps of: traversing to obtain a second damaged section, obtaining a running route section with damage, generating a maintenance instruction, and counting the number of passing railway wagons of each running route, wherein the higher the number of passing railway wagons is, the higher the maintenance priority of the running route is.
The invention has the following beneficial effects:
1. compared with the prior art, the method can mark the track section with damage, thereby overhauling all railway wagons passing through the section, being accurate and efficient, and not needing to detect faults one by one railway wagons.
2. The number of the railway freight cars is very small, the gray level images of the wheel treads of a small number of railway freight cars are collected, and finally the intersection of the running paths of the small number of railway freight cars is obtained, so that the damaged road sections of the railway freight cars can be determined, all railway freight cars passing through the road sections can be reversely pushed out, the probability of faults of the railway freight cars is high, the step is traversed, whether a large number of railway freight cars have faults or not can be rapidly checked, and the labor is saved.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. In the drawings, embodiments of the invention are illustrated by way of example and not by way of limitation, and like reference numerals refer to similar or corresponding parts and in which:
FIG. 1 is a flow chart of a rail wagon fault detection method based on artificial intelligence according to an embodiment of the invention;
FIG. 2 is a schematic illustration of a railway wagon wheel tread damage based on an artificial intelligence based railway wagon fault detection method in accordance with an embodiment of the present invention;
fig. 3 is a schematic diagram of damage vertical distance based on an artificial intelligence based railway wagon fault detection method according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that when the terms "first," "second," and the like are used in the claims, the specification and the drawings of the present invention, they are used merely for distinguishing between different objects and not for describing a particular sequential order. The terms "comprises" and "comprising" when used in the specification and claims of the present invention are taken to specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The invention provides a railway wagon fault detection method based on artificial intelligence. As shown in FIG. 1, an artificial intelligence based railway wagon fault detection method comprises steps S1-S5, which are described in detail below.
S1, obtaining a metal falling block peeling damage area and a crack area by carrying out edge detection on a wheel tread gray image of a railway wagon.
In one embodiment, all wheel treads of the railway wagon are shot through fixed cameras arranged at parallel positions on two sides beside the railway wagon track, RGB (RED GREEN BLUE) images of the wheel treads of the railway wagon are obtained, the RGB images of the wheel treads of the railway wagon are converted into gray images, and the gray images of the wheel treads of the railway wagon are obtained.
Performing target detection on the wheel tread gray level images, marking a metal falling block peeling damage area and a crack area in the acquired multiple wheel tread gray level images, as shown in fig. 2, which is a wheel tread damage schematic diagram of a railway wagon, putting the multiple wheel tread gray level images carrying the marks into a preset detection model for training, and responding to the loss function being smaller than a preset value to obtain a target detection model.
The wheel tread gray level image obtained in real time is put into a preset target detection model to obtain the wheel tread gray level image, and the candidate frames obtained in the target detection process are the circumscribed rectangle of the metal falling block peeling damage area and the circumscribed rectangle of the crack area, and are not the actual metal falling block peeling damage area and the crack area, so that the wheel tread damage gray level image is required to be subjected to edge detection, and the metal falling block peeling damage area and the crack area are obtained.
S2, respectively constructing gray level co-occurrence matrixes for the metal falling block peeling damage area and the crack area to obtain characteristic information.
Specifically, a first gray level co-occurrence matrix is constructed according to a metal falling block peeling damage area, damage characteristic information of the metal falling block peeling damage area is obtained, a second gray level co-occurrence matrix is constructed according to a crack area, and texture characteristic information of the crack area is obtained.
In one embodiment, a first gray level co-occurrence matrix is constructed according to a metal falling block peeling damage area, the confusion of the first gray level co-occurrence matrix is calculated as the complexity of the metal falling block peeling damage area, and damage characteristic information of the metal falling block peeling damage area is recorded and usedAnd (3) representing. The damage characteristic information of the metal falling block peeling damage area can use the first gray level symbiotic momentEntropy value of the matrix. The more chaotic the damage condition in the damaged area of the metal falling block and peeling of the wheel tread is, the damage characteristic information value of the damaged area of the metal falling block and peeling is +.>The larger. The first gray level co-occurrence matrix is a number matrix of gray level gradient changes and contains texture information in the image.
And constructing a second gray level co-occurrence matrix according to the crack region, calculating the energy characteristics of the second gray level co-occurrence matrix, and representing the grain thickness of the crack region. The thicker the tread crack texture, the greater the texture feature value a of the crack region.
S3, acquiring position information of a metal chipping and peeling damage area and position information of a crack area.
As shown in fig. 3, a schematic diagram of a damage vertical distance is shown, specifically, a damage vertical distance perpendicular to an inner boundary of a railway wagon is obtained by taking a center of any one metal falling block peeling damage area or a center of any one crack area as a starting point and taking the inner boundary of the railway wagon as an ending point, namely the damage vertical distance; and calculating the width of the railway wagon according to the number of the gray image pixels of the tread of the wheel.
Calculating first position information, wherein the first position information is the ratio of the damage vertical distance of a metal falling block peeling damage area to the width of the railway wagon; and calculating second position information, wherein the second position information is the ratio of the damage vertical distance of the crack area to the width of the railway wagon.
In one embodiment, the position of the camera is close to one side of the track when the railway wagon is in driving in the shooting process, when the railway wagon passes from far to near, the gray level images of the wheel treads of each frame are changed from small to large, so that the gray level images of the wheel treads of each frame in the shooting process are inconsistent, but the ratio of the vertical distance to the gray level images of the inner side boundary of the railway wagon is consistent with the gray level images of the wheel treads of each frame, the vertical distance between the damaged area and the inner side boundary of the railway wagon is required to be constructed, the center of the minimum circumscribed rectangle of the metal falling and peeling damaged area is marked, the center of the minimum circumscribed rectangle of the crack area is marked, the center of any metal falling and peeling damaged area or the center of the crack area is taken as the starting point, the inner side boundary of the railway wagon is taken as the end point, the damaged vertical distance perpendicular to the inner side boundary of the railway wagon is obtained, and the width of the railway wagon can be directly obtained through the number of pixels of the gray level images of the wheel treads.
First position informationSecond position information for the ratio of the damage vertical distance of the damaged area of the metal falling block to the width of the railway wagon>The ratio of the vertical distance of damage to the cracked area to the width of the railway wagon.
S4, constructing a comprehensive damage characteristic matrix, and calculating the similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons.
Specifically, a comprehensive damage characteristic matrix is constructed according to the first position information, the second position information, the damage characteristic information and the texture characteristic information. And calculating the similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons, and generating judging information of the same damage of the two trains of railway wagons in response to the similarity being greater than a preset threshold value.
In one embodiment, a comprehensive lesion characterization matrix is constructed from the first location information, the second location information, the lesion characterization information, and the texture characterization information.
As when the rail is damaged, all the wheel treads on the same side of the rail passing through the same position can be similarly damaged. For example, a camera shoots nodes, and all shot images are wheel tread gray images on the same side of the railway freight car, so that all wheel tread gray images shot by the camera are wheel tread gray images on the same side, wheel tread damage feature matrixes on the same side of all railway freight cars recorded by the camera are calculated, similarity of wheel tread damage feature matrixes on the same side of each railway freight car is calculated, and the same side with similar damage can be found. The similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons meets the relation:
wherein,for the similarity of the comprehensive lesion areas, +.>Similarity of peeling damaged area for metal falling block, < >>For the similarity of crack areas, +.>Feature matrix of metal falling block peeling damage area of railway freight car in line>Feature matrix of metal falling block peeling damage area of another train of railway freight car>Characteristic matrix for a train of railway wagon crack areas, < >>For the characteristic matrix of another train of railway wagon crack areas, < > for>For vector distance calculation, an example may be euclidean distance.
Comprehensive injurySimilarity of regionsThe larger the damage area of the rail wagons, the more similar the two trains.
Setting a similarity threshold of the comprehensive damage areas, and when the similarity is larger than a preset threshold, indicating that the two railway wagons have similar damage areas, and reserving the current railway wagons, otherwise, not considering the current railway wagons. Illustratively, the similarity threshold for the composite lesion area is set to 0.9.
S5, calculating a first damage interval and a second damage interval, and generating and sending all railway wagon information and maintenance instructions which need maintenance.
Specifically, calculating the similarity of the wheel tread comprehensive damage areas of any two trains of all the railway wagons passing through a first damage interval in a preset time period, wherein the first damage interval is the intersection of two train running routes with the same damage; and reserving railway wagons with the similarity larger than a preset threshold value, obtaining a running route intersection, marking the running route intersection as a second damage interval, generating information of all railway wagons needing maintenance, and sending maintenance instructions.
In one embodiment, taking any damaged area in the wheel tread of one train of railway freight car as an example, if the damaged area of the wheel tread of another train of railway freight car exists and the damaged area, the similarity of the comprehensive damaged area is obtainedIf the speed is more than 0.9, recording the driving route of the railway wagon as +.>The driving route of another train of railway freight car is +.>。
In response to the same damage to the wheel treads of the two trains, generating judging information that the two trains are worn with similarity greater than a preset threshold value when passing through the same track;
calculating a first injury interval:
wherein,for the first injury zone, ++>For the driving route of the railway wagon of column A, < > for the railway wagon of column A>Is the driving route of the railway wagon of the B column.
Based on the first damage section, obtaining wheel tread gray level images of all railway wagons of a driving route passing through the first damage section according to a camera, calculating the similarity of wheel tread comprehensive damage areas of any two railway wagons in the wheel tread gray level images of all railway wagons, and reserving a driving route set of all railway wagons with the similarity larger than a preset threshold valueCalculating a second injury interval:
wherein,for the driving route of the first train of rail wagons with similarity greater than the preset threshold value, +.>For the driving route of the second train of rail wagons with similarity greater than the preset threshold value, +.>The>Train route of rail wagon->Is the second injury interval.
Traversing to obtain a second damaged section, obtaining a running route section with damage, generating a maintenance instruction, counting the number of passing railway wagons of each running route, and if the number of passing railway wagons is larger, the maintenance priority of the running route is higher.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (5)
1. The railway wagon fault detection method based on artificial intelligence is characterized by comprising the following steps:
obtaining a metal falling block peeling damage area and a crack area by carrying out edge detection on the gray level image of the wheel tread;
constructing a first gray level co-occurrence matrix according to the metal falling and peeling damage area to obtain damage characteristic information of the metal falling and peeling damage area, and constructing a second gray level co-occurrence matrix according to the crack area to obtain texture characteristic information of the crack area;
calculating first position information and second position information, and constructing a comprehensive damage characteristic matrix related to the first position information, the second position information, damage characteristic information and texture characteristic information, wherein the first position information is the position information of a metal falling block peeling damage area, and the second position information is the position information of a crack area;
randomly extracting two rows of railway wagons, calculating the similarity of the wheel tread comprehensive damage areas of the two rows of railway wagons, generating judging information of the same damage of the two rows of railway wagons in response to the similarity being larger than a preset threshold, and calculating a first damage interval, wherein the first damage interval is an intersection of the two rows of railway wagons with the same damage;
calculating the similarity of the wheel tread comprehensive damage areas of any two trains of railway wagons passing through the first damage area in a preset time period;
the rail wagons with the similarity larger than a preset threshold value are reserved, a running route intersection is obtained, a second damage interval is generated, and all rail wagons information and maintenance instructions needing maintenance are generated and sent;
the step of calculating the first position information and the second position information comprises the following steps:
calculating a damage vertical distance, and taking the center of any one of the metal falling block peeling damage areas or the center of any one of the crack areas as a starting point and taking the inner side boundary of the railway wagon as an end point to obtain the damage vertical distance vertical to the inner side boundary of the railway wagon;
calculating the width of the railway wagon according to the number of gray image pixels of the tread of the wheel;
calculating first position information, wherein the first position information is the ratio of the damage vertical distance of a metal falling block peeling damage area to the width of the railway wagon;
and calculating second position information, wherein the second position information is the ratio of the damage vertical distance of the crack area to the width of the railway wagon.
2. The method for detecting the faults of the railway freight car based on artificial intelligence according to claim 1, wherein the similarity of the wheel tread comprehensive damage areas of the two railway freight cars meets the following relation:
wherein,for the similarity of the comprehensive lesion areas, +.>Similarity of peeling damaged area for metal falling block, < >>For the similarity of crack areas, +.>Feature matrix of metal falling block peeling damage area of first-row railway wagon>Feature matrix of metal falling block peeling damage area of second-row railway wagon>Characteristic matrix for the crack zone of the first train rail wagon,>feature matrix for crack area of second-row railway wagon, < > for railway wagon>For vector distance calculation.
3. The method for detecting the faults of the railway freight car based on artificial intelligence according to claim 1, wherein the steps of obtaining damage characteristic information of a damaged area of metal falling block spalling and obtaining texture characteristic information of a cracked area comprise the following steps:
constructing a first gray level co-occurrence matrix according to the metal falling block peeling damage area;
calculating damage characteristic information of a metal falling block peeling damage area, wherein the damage characteristic information is the confusion degree of the first gray level co-occurrence matrix;
constructing a second gray level co-occurrence matrix according to the crack area;
and calculating texture feature information of the crack area, wherein the texture feature information is the energy feature of the second gray level co-occurrence matrix.
4. The method for detecting a rail wagon fault based on artificial intelligence according to claim 1, wherein the step of generating the second damage interval comprises the steps of:
generating judging information that the two railway wagons are worn with the similarity being greater than a preset threshold value when passing through the same track in response to the same damage to the wheel treads of the two railway wagons;
calculating a first damage interval, wherein the calculation formula is as follows:
wherein,for the first injury zone, ++>For the driving route of the railway wagon of column A, < > for the railway wagon of column A>Iron of column BThe driving route of the road truck;
based on the first damaged section, obtaining wheel tread gray level images of all railway wagons of a driving route passing through the first damaged section according to a camera, calculating the similarity of wheel tread comprehensive damaged areas of any two railway wagons in the wheel tread gray level images of all railway wagons, and reserving a driving route set of all railway wagons with the similarity larger than a preset threshold valueAnd calculating a second damage interval, wherein the calculation formula is as follows:
wherein,for the driving route of the first train of rail wagons with similarity greater than the preset threshold value, +.>For the driving route of the second train of rail wagons with similarity greater than the preset threshold value, +.>For the driving route of the first rail wagon with similarity greater than the preset threshold value, +.>Is the second injury interval.
5. The method for detecting a rail wagon fault based on artificial intelligence according to claim 1, further comprising the steps of:
traversing to obtain a second damaged section, obtaining a running route section with damage, generating a maintenance instruction, and counting the number of passing railway wagons of each running route, wherein the higher the number of passing railway wagons is, the higher the maintenance priority of the running route is.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311452020.8A CN117173183B (en) | 2023-11-03 | 2023-11-03 | Railway wagon fault detection method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311452020.8A CN117173183B (en) | 2023-11-03 | 2023-11-03 | Railway wagon fault detection method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117173183A CN117173183A (en) | 2023-12-05 |
| CN117173183B true CN117173183B (en) | 2024-01-26 |
Family
ID=88930281
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311452020.8A Active CN117173183B (en) | 2023-11-03 | 2023-11-03 | Railway wagon fault detection method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117173183B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102310174A (en) * | 2011-09-07 | 2012-01-11 | 中国科学院金属研究所 | Method and device for improving metal solidification defects and refining solidification textures |
| CN104608799A (en) * | 2014-12-12 | 2015-05-13 | 郑州轻工业学院 | Information fusion technology based train wheel set tread damage online detection and recognition method |
| CN106683076A (en) * | 2016-11-24 | 2017-05-17 | 南京航空航天大学 | Texture feature clustering-based locomotive wheelset tread damage detection method |
| CN109145492A (en) * | 2018-09-10 | 2019-01-04 | 西南交通大学 | Tire tread defects analogy method and device |
| CN110276297A (en) * | 2019-06-21 | 2019-09-24 | 中国神华能源股份有限公司 | Train tyre tread wear detecting system and method |
| CN112907524A (en) * | 2021-02-04 | 2021-06-04 | 哈尔滨市科佳通用机电股份有限公司 | Method for detecting fault of fire-proof plate of rail wagon based on image processing |
| CN217623587U (en) * | 2022-06-14 | 2022-10-21 | 南京航空航天大学 | A piston shoe structure for assisting rail damage detects |
| CN115631174A (en) * | 2022-11-01 | 2023-01-20 | 南通皋亚钢结构有限公司 | Mechanical part surface defect identification method based on visual image data |
| CN116168028A (en) * | 2023-04-25 | 2023-05-26 | 中铁电气化局集团有限公司 | High-speed rail original image processing method and system based on edge filtering under low visibility |
-
2023
- 2023-11-03 CN CN202311452020.8A patent/CN117173183B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102310174A (en) * | 2011-09-07 | 2012-01-11 | 中国科学院金属研究所 | Method and device for improving metal solidification defects and refining solidification textures |
| CN104608799A (en) * | 2014-12-12 | 2015-05-13 | 郑州轻工业学院 | Information fusion technology based train wheel set tread damage online detection and recognition method |
| CN106683076A (en) * | 2016-11-24 | 2017-05-17 | 南京航空航天大学 | Texture feature clustering-based locomotive wheelset tread damage detection method |
| CN109145492A (en) * | 2018-09-10 | 2019-01-04 | 西南交通大学 | Tire tread defects analogy method and device |
| CN110276297A (en) * | 2019-06-21 | 2019-09-24 | 中国神华能源股份有限公司 | Train tyre tread wear detecting system and method |
| CN112907524A (en) * | 2021-02-04 | 2021-06-04 | 哈尔滨市科佳通用机电股份有限公司 | Method for detecting fault of fire-proof plate of rail wagon based on image processing |
| CN217623587U (en) * | 2022-06-14 | 2022-10-21 | 南京航空航天大学 | A piston shoe structure for assisting rail damage detects |
| CN115631174A (en) * | 2022-11-01 | 2023-01-20 | 南通皋亚钢结构有限公司 | Mechanical part surface defect identification method based on visual image data |
| CN116168028A (en) * | 2023-04-25 | 2023-05-26 | 中铁电气化局集团有限公司 | High-speed rail original image processing method and system based on edge filtering under low visibility |
Non-Patent Citations (2)
| Title |
|---|
| " 城轨车辆齿轮箱滚动轴承在线状态监测与故障诊断";赵永鑫;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;第C033-399页 * |
| "基于Lamb 的金属薄板损伤主动监测技术研究";解维华等;《压电与声光》;第30卷(第3期);第349-352页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117173183A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111080598B (en) | Bolt and nut missing detection method for coupler yoke key safety crane | |
| CN110047070B (en) | Method and system for identifying rail wear degree | |
| CN111080603B (en) | Method for detecting breakage fault of shaft end bolt of railway wagon | |
| CN109766746B (en) | Track foreign matter detection method for aerial video of unmanned aerial vehicle | |
| CN111080607B (en) | Rolling bearing oil slinging fault detection method based on image recognition | |
| CN110276297B (en) | Train tread wear detection system and method | |
| CN110222436B (en) | Method and device for evaluating health state of train parts and storage medium | |
| CN113221839B (en) | Automatic truck image identification method and system | |
| CN111079734B (en) | Method for detecting foreign matters in triangular holes of railway wagon | |
| CN111080614A (en) | Method for identifying damage to rim and tread of railway wagon wheel | |
| CN113516629A (en) | Intelligent detection system for TFDS passing operation | |
| CN106683040B (en) | Infrared panoramic image splicing method based on NCC algorithm | |
| CN108010044B (en) | Video boundary detection method | |
| CN117173183B (en) | Railway wagon fault detection method based on artificial intelligence | |
| CN105957352B (en) | Parking offense detection method and device | |
| CN113506269A (en) | Turnout and non-turnout rail fastener positioning method based on deep learning | |
| CN113469974A (en) | Method and system for monitoring grate plate state of pellet chain grate | |
| CN111091557B (en) | Method and system for detecting breakage fault of flange of bearing saddle of railway wagon | |
| Lai et al. | Machine vision analysis of the energy efficiency of intermodal freight trains | |
| CN116631187A (en) | Intelligent acquisition and analysis system for case on-site investigation information | |
| CN115541591A (en) | Method and system for detecting abrasion edge of carbon pantograph slider of train | |
| CN115631146A (en) | Image-based pantograph carbon slide strip defect detection method | |
| CN115424128A (en) | Fault image detection method and system for lower link of freight car bogie | |
| Amu et al. | Detection of wheel discoloration using R-CNN | |
| Sunkara et al. | Analysis and classification of railway track surfaces based on image processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |