CN113236653A - Bolt structure with loosening detection function and bolt loosening detection method - Google Patents
Bolt structure with loosening detection function and bolt loosening detection method Download PDFInfo
- Publication number
- CN113236653A CN113236653A CN202110427605.9A CN202110427605A CN113236653A CN 113236653 A CN113236653 A CN 113236653A CN 202110427605 A CN202110427605 A CN 202110427605A CN 113236653 A CN113236653 A CN 113236653A
- Authority
- CN
- China
- Prior art keywords
- bolt
- warning area
- nut
- state
- bolt structure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 claims description 24
- 238000012856 packing Methods 0.000 claims description 23
- 238000012423 maintenance Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 21
- 238000013135 deep learning Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 11
- 230000005284 excitation Effects 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000011176 pooling Methods 0.000 description 6
- 241001669679 Eleotris Species 0.000 description 5
- 206010040007 Sense of oppression Diseases 0.000 description 5
- 229910000831 Steel Inorganic materials 0.000 description 5
- 238000009434 installation Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 5
- 239000010959 steel Substances 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 239000010426 asphalt Substances 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000001125 extrusion Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000003466 welding Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011083 cement mortar Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000004567 concrete Substances 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 239000011513 prestressed concrete Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000011150 reinforced concrete Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16B—DEVICES FOR FASTENING OR SECURING CONSTRUCTIONAL ELEMENTS OR MACHINE PARTS TOGETHER, e.g. NAILS, BOLTS, CIRCLIPS, CLAMPS, CLIPS OR WEDGES; JOINTS OR JOINTING
- F16B39/00—Locking of screws, bolts or nuts
- F16B39/22—Locking of screws, bolts or nuts in which the locking takes place during screwing down or tightening
- F16B39/24—Locking of screws, bolts or nuts in which the locking takes place during screwing down or tightening by means of washers, spring washers, or resilient plates that lock against the object
- F16B39/26—Locking of screws, bolts or nuts in which the locking takes place during screwing down or tightening by means of washers, spring washers, or resilient plates that lock against the object with spring washers fastened to the nut or bolt-head
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16B—DEVICES FOR FASTENING OR SECURING CONSTRUCTIONAL ELEMENTS OR MACHINE PARTS TOGETHER, e.g. NAILS, BOLTS, CIRCLIPS, CLAMPS, CLIPS OR WEDGES; JOINTS OR JOINTING
- F16B43/00—Washers or equivalent devices; Other devices for supporting bolt-heads or nuts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Mechanical Engineering (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The invention relates to the technical field of rail safety detection, in particular to a bolt structure with a loosening detection function and a bolt loosening detection method. According to the invention, when the bolt structure is loosened, the shielding piece and the warning area are staggered, the warning area is partially or completely in an exposed state according to different loosening degrees of the nuts, and whether the warning area on the bolt structure is exposed or not is directly observed, so that whether the bolt structure is loosened or not can be directly judged, and the portability of track maintenance and detection is effectively improved.
Description
Technical Field
The invention relates to the technical field of rail safety detection, in particular to a bolt structure with a loosening detection function and a bolt loosening detection method.
Background
The track refers to a route laid by strip-shaped steel for running of trains, trams and the like. With the increase of the railway transportation volume and the increase of the axle weight and the running speed of the locomotive, a plurality of novel tracks such as a seamless track, a wide sleeper track, an integral track bed track, a plate track and the like appear in succession; the seamless track is also called as a welding long steel rail track, and is a track for welding common steel rails without leaving rail gaps, each welded steel rail is not less than 200 meters in length, and the actual application is generally 800-1000 meters or more. The wide sleeper line is also called sleeper slab line, and is constructed by using prestressed concrete sleeper slabs which are densely laid on a compacted road bed, and filling asphalt or other materials between slab joints. The concrete (generally provided with steel bars) for the whole track bed line is directly poured on a stable and solid roadbed without using a common sleeper and a novel line of a gravel track bed. A plate-type track is a track made of reinforced concrete and emulsified asphalt cement mortar as regulation layer (or a layer of elastic high-molecular material as cushion layer) under the plate.
In the process of track installation and operation, due to the influence of train running vibration, bolts for connection and fastening are easy to loosen.
Therefore, the bolt structure with the loosening detection function and the bolt loosening detection method are simple in structure, convenient to use and capable of effectively improving bolt loosening condition detection efficiency.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a bolt structure with a loosening detection function and a bolt loosening detection method.
The purpose of the invention is realized by the following technical scheme:
the utility model provides a bolt structure with not hard up detection function, includes nut and packing ring, the nut bottom is provided with shielding piece, be provided with boss and warning area on the packing ring, boss and warning area set up on same circumference, shielding piece includes first state and second state, and under the first state, the nut is in the stable connection state, and shielding piece shields the warning area, and under the second state, the nut is in not hard up state, and shielding piece and warning area dislocation, warning area are in part or whole exposure state.
Further, an annular connecting piece is arranged at the bottom of the nut, and the shielding piece is arranged on the side wall of the annular connecting piece. In the actual use process, when the bolt structure is installed, the nut and the washer are sleeved on the bolt, the nut is rotated, the distance between the washer and the head of the bolt is adjusted, the track mechanical unit is fixedly connected by utilizing the matching of the nut, the washer and the bolt, when the nut is screwed down and is stably connected with the bolt, the shielding piece completely shields the warning area on the washer and is abutted against the side wall of the boss, and at the moment, the bolt structure is shown to be in a stable connection state; and when the not hard up condition appeared in bolt structure, the nut after becoming flexible will drive the shielding piece rotation, lets dislocation between shielding piece and the warning district, and is different according to the not hard up degree of nut, and the warning district will be partly or whole in the state of exposing, and at this moment, whether warning district in the direct observation bolt structure exposes, just can directly judge whether not hard up of bolt structure, effectively improves the portability to the track maintenance detection.
Further, the outer diameter of the annular connecting piece is the same as the diameter of the circumcircle of the nut.
Further, the shield is of an arch-shaped structure.
Furthermore, boss and warning area are domes respectively.
Furthermore, the shielding piece is of a 180-degree arch structure, and the boss and the warning area are of 90-degree arch structures respectively.
Further, the boss and the warning area are arranged adjacently. The shielding piece is arranged to be in a 180-degree arch structure, the boss and the warning area are respectively arranged to be in a 90-degree arch structure, the boss and the warning area are arranged adjacently, the nut is in a stable connection state when the shielding piece is in a first state, the shielding piece shields the warning area, meanwhile, the shielding piece is abutted against one side wall of the boss, at the moment, the boss is located on the circumference of the gasket, 1/4 is the boss, 1/2 is shielded by the shielding piece, and 1/4 is in an exposed state; and when the shielding piece was in and the second state, shielding piece and warning area dislocation, at this moment, utilized the opposite side lateral wall of boss, can carry on spacingly to the shielding piece to a certain extent, restraint nut's not hard up degree realizes exposing warning area, shows not hard up warning information's basis, prevents effectively that bolt structure from further becoming flexible, improves bolt structure's safety in utilization.
Furthermore, the outer diameter of the shielding piece is not smaller than the outer diameter of the warning area, and the inner diameter of the shielding piece is not larger than the inner diameter of the warning area.
Further, the friction coefficient of the upper surface of the gasket where the warning area is located is smaller than that of the lower surface of the gasket. Through being less than the lower surface friction system of packing ring with the coefficient of friction of warning district place gasket upper surface, let the packing ring be difficult for taking place to slide with mechanical unit, let the difficult nut that is become flexible after the packing ring drive the rotation simultaneously, guarantee that the warning district on the packing ring is changeed after the nut is not hard up and is revealed out.
Further, the surface in warning district is provided with the dope layer. In the installation use of bolt structure, according to the difference of installation requirement, can select the control boss to be in the low point, let bolt structure can not cause the unexpected condition that shows in warning district to take place because of receiving the boss dead weight influence, also can select the control boss to be in the high point, and possess and receive the dead weight and have the trend to low point pivoted, after bolt structure is not hard up, the packing ring is because the extrusion force that receives reduces when unable to overcome the influence of boss dead weight, the packing ring just can receive the gravity influence of boss to take place the rotation, through the rotation of packing ring, and the rotation of the nut that becomes flexible, expose warning district maximum range, show the not hard up warning information of this bolt structure.
Further, the shielding piece is abutted against the side wall of the boss in the first state.
Further, boss and warning area set up on same circumference, warning area and boss adjacent setting, and the one end that the boss was kept away from in the warning area is provided with elasticity locating part. Under the first state, elasticity locating part receives the shielding piece oppression to be in the shrink state, when bolt assembly is not hard up, nut after not hard up will drive the shielding piece rotatory, let the dislocation between shielding piece and the warning area, treat that the not hard up degree of shielding piece exceeds the predetermined value, shielding piece can rotate 90 promptly, let warning area and elasticity locating part expose completely, at this moment, elasticity locating part loses the oppression of shielding piece, extend at once, the lateral wall in warning area is kept away from to the cooperation boss, spacing to the shielding piece, let shielding piece and packing ring joint, make the warning area keep exposing the state always, let the information in warning area can not receive bolt assembly further not hard up and shielded once more, improve equipment's use convenience.
Further, the diameter of the inner arc of the boss is not smaller than the outer diameter of the annular connecting piece.
A rail bolt looseness detection method is based on the bolt structure and is used for detecting through a method comprising the following steps: the method comprises the steps of obtaining picture information of a bolt on a track, judging whether a warning area on a bolt structure is in an exposed state or not according to the picture information, if part or all of the warning area is in the exposed state, indicating that the bolt is loosened, and if the warning area is not exposed, indicating that the bolt is not loosened.
Furthermore, in the process of judging the picture information, intelligent analysis and judgment are carried out on the picture information by adopting a deep learning method, the deep learning method is realized on the basis of a Tensorflow deep learning framework, and a data source uses an MNIST data set and carries out model training by respectively adopting a softmax regression algorithm and CNN deep learning.
Further, the deep network structure for CNN deep learning comprises a convolutional layer, an excitation layer, a pooling layer and a full-link layer
The convolutional layer is: the operation of performing inner product on the image and the filter matrix is convolution, the convolution layer can be used for acquiring the characteristics of the image through the convolution operation, the modules are stacked, different convolution kernels are used for acquiring higher-order characteristics, and in the stacked depth structure, the weight parameters of the convolution kernels are shared, so that the depth structure can acquire higher-dimensional characteristics, but the parameter quantity is not greatly increased;
the excitation layer is as follows: the excitation layer is not explicitly indicated in the figure, and is usually used for selecting an appropriate excitation function after the convolution operation to process the result of the convolution operation; the function of the activation function is to reserve and map the characteristics of the activated neurons through the function, so as to solve the nonlinear problem of the network structure;
the pooling layer is as follows: and the data is downsampled, so that overfitting can be avoided, and the operations such as data characteristics and data calculation amount are reduced.
The full connection layer: the connection of the layer is the connection of the neurons in the neural network structure between each layer; the fully connected layer plays a role of a classifier in the whole convolutional neural network; mapping original data to a hidden layer feature space through operations of a convolutional layer, a pooling layer, an activation function layer and the like, wherein a full connection layer is used for mapping learned distributed feature representation to a sample mark space; a layer of structure which can be customized after passing through the layer is used for classification or regression; the full connection layer is realized by acquiring more characteristics of high-dimensional images through a depth network structure to process the images, and controlling parameters through convolution operation.
The invention has the beneficial effects that: according to the bolt structure with the loosening detection function and the bolt loosening detection method, when the bolt structure is loosened, the loosened nut drives the shielding piece to rotate, the shielding piece and the warning area are staggered, part or all of the warning area is in an exposed state according to different loosening degrees of the nut, and at the moment, whether the warning area on the bolt structure is exposed or not is directly observed, so that whether the bolt structure is loosened or not can be directly judged, and the portability of rail maintenance detection is effectively improved.
Drawings
FIG. 1 is a schematic structural view of a bolt structure according to the present invention;
FIG. 2 is a schematic structural view of the nut of the present invention;
FIG. 3 is a schematic view of the construction of the gasket of the present invention;
FIG. 4 is a schematic view of the construction of the shield of the present invention;
FIG. 5 is a graph showing the results of the detection in the test example of the present invention;
FIG. 6 is a graph showing another detection result of the detection method in the test example of the present invention;
in the figure, 1, a nut; 2. a gasket; 3. a shield; 4. a boss; 5. a warning area; 6. an elastic limiting part; 7. a limiting plate; 8. a groove; 9. the elastic piece is clamped.
Detailed Description
The technical solutions of the present invention are further described in detail below with reference to the accompanying drawings, but the scope of the present invention is not limited to the following.
As shown in fig. 1-3, a bolt structure with become flexible and detect function, includes nut and 1 packing ring 2, 1 bottom of nut is provided with shielding piece 3, be provided with boss 4 and warning area 5 on the packing ring 2, boss 4 and warning area 5 set up on same circumference, shielding piece 3 includes first state and second state, and under the first state, nut 1 is in stable connection state, and shielding piece 3 shields warning area 5, and under the second state, nut 1 is in not hard up state, and shielding piece 3 misplaces with warning area 5, and warning area 5 is in partial or whole exposure state.
Specifically, the bottom of the nut 1 is provided with an annular connecting piece, and the shielding piece 3 is arranged on the side wall of the annular connecting piece. In the actual use process, when the bolt structure is installed, the nut 1 and the washer 2 are sleeved on the bolt, the nut 1 is rotated, the distance between the washer 2 and the head of the bolt is adjusted, the track mechanical unit is fixedly connected by utilizing the matching of the nut 1, the washer 2 and the bolt, when the nut 1 is screwed down and is stably connected with the bolt, the shielding piece 3 completely shields the warning area 5 on the washer 2, and the shielding piece 3 is abutted against the side wall of the boss 1, at the moment, the bolt structure is shown to be in a stable connection state; and when the not hard up condition appeared in the bolt structure, nut 1 after not hard up will drive shielding piece 3 rotatory, lets dislocation between shielding piece 3 and the warning district 5, and according to the not hard up degree difference of nut 1, warning district 5 will be partly or all in the state of exposing, and at this moment, whether warning district 5 in the direct observation bolt structure exposes, just can directly judge whether not hard up of bolt structure, effectively improves the portability to the track maintenance detection.
Specifically, the outer diameter of the annular connector is the same as the diameter of the circumcircle of the nut 1.
In particular, the shutter 3 is of arched configuration.
Specifically, the boss 4 and the warning area 5 are respectively of an arch structure.
Specifically, the shielding member 3 is of an arch structure of 180 degrees, and the boss 4 and the warning area 5 are of arch structures of 90 degrees respectively.
Specifically, the boss 4 and the warning area 5 are adjacently arranged. By arranging the shielding piece 3 into a 180-degree arch structure, arranging the boss 4 and the warning area 5 into 90-degree arch structures respectively, and arranging the boss 4 and the warning area 5 adjacently, the nut 1 is in a stable connection state when the shielding piece 3 is in a first state, the shielding piece 3 shields the warning area 5, meanwhile, the shielding piece 3 is abutted against one side wall of the boss 4, at the moment, the boss 4 is arranged on the circumference of the gasket 2, 1/4 is the boss 4, 1/2 is shielded by the shielding piece 3, and 1/4 is in an exposed state; and when shielding piece 3 was in and the second state, shielding piece 3 misplaced with warning area 5, at this moment, utilized boss 4's opposite side lateral wall, can carry on spacingly to shielding piece 3 to a certain extent, and restraint nut 1's not hard up degree realizes exposing warning area 5, shows not hard up warning information's basis, prevents effectively that bolt structure is further not hard up, improves bolt structure's safety in utilization.
Specifically, the outer diameter of the shielding piece 3 is not smaller than the outer diameter of the warning area 5, and the inner diameter of the shielding piece 3 is not larger than the inner diameter of the warning area 5.
In particular, the friction coefficient of the upper surface of the washer 2 on which said warning zone 5 is located is lower than the friction system of the lower surface of the washer 2. Through being less than the lower surface friction system of packing ring 2 with the coefficient of friction of 5 surperficial washers in warning area 2, let packing ring 2 be difficult for taking place to slide with mechanical unit, let packing ring 2 be difficult for being driven by the nut after becoming flexible simultaneously and rotate, guarantee that warning area 5 on packing ring 2 is more easily revealed out after nut 1 is become flexible.
Specifically, the surface of warning district 5 is provided with the dope layer. In the installation use process of bolt structure, according to the difference of installation requirement, can select control boss 4 to be in the low point, let bolt structure can not cause the unexpected condition that shows in warning district to take place because of receiving the influence of boss dead weight, also can select control boss 4 to be in the high point, and possess and receive the dead weight and have the trend that rotates to the low point, after bolt structure is not hard up, when the packing ring reduces to unable the influence of overcoming boss 4 dead weight because the extrusion force that receives, packing ring 2 just can be influenced by the gravity of boss 4 and take place the rotation, through the rotation of packing ring 2, and the rotation of the nut 1 that becomes flexible, will expose warning district 5 maximum ranges, show the not hard up warning information of this bolt structure.
Specifically, in the first state of the shield 3, the shield 3 abuts against the side wall of the boss 1.
Specifically, boss 3 and warning area 5 set up on same circumference, warning area 5 sets up with boss 4 is adjacent, and warning area 5 keeps away from the one end of boss 4 and is provided with elasticity locating part. Under the first state, elasticity locating part receives shielding piece 3 oppression to be in the shrink state, when bolt assembly is not hard up, nut 1 after not hard up will drive shielding piece 3 rotatory, let dislocation between shielding piece 3 and the warning area 5, treat that shielding piece 3 not hard up degree exceeds the predetermined value, shielding piece 3 rotates 90 promptly, let warning area 5 and elasticity locating part expose completely, at this moment, elasticity locating part loses shielding piece 3's oppression, the extension at once, cooperation boss 4 keeps away from the lateral wall of warning area 5, it is spacing to carry out shielding piece 3, let shielding piece 3 and packing ring 2 joint, make warning area 5 keep exposing the state always, let the information of warning area 5 can not receive bolt assembly further not hard up and be shielded once more, improve equipment's use convenience.
Specifically, as shown in fig. 4, a limiting plate 7 is arranged at the bottom of the shielding piece 3, the limiting plate 7 is of an arch structure, and the limiting plate 7 is connected with the shielding piece 3 through a clamping elastic piece 9.
Specifically, the limiting plate 7 is in an arch structure of 150-175 degrees. Preferably, the limiting plate 7 is in an arch structure of 170 degrees.
Specifically, the bottom of shielding member 3 is provided with recess 8, recess 8 and limiting plate 7 phase-match, joint elastic component 9 evenly sets up in recess 8. In the first state, the nut 1 is stably connected, the elastic limiting piece is pressed by the shielding piece 3 to be in a contraction state, and the shielding piece 3 simultaneously compresses and clamps the elastic piece 9 and is positioned in the groove 8; at the not hard up degree of shielding member 3 and surpassing the predetermined value, shielding member 3 rotates to let warning district 5 and elasticity locating part expose completely promptly, at this moment, elasticity locating part loses shielding member 3's oppression, extend at once, the lateral wall of warning district 5 is kept away from to cooperation boss 4, constitute a spacing region, simultaneously, joint elastic component 9 releases recess 8 with limiting plate 7, push limiting plate 7 into in the spacing region, spacing to limiting plate 7, let shielding member 3 and packing ring 2 joint, make warning district 5 keep exposing the state always, let the information in warning district 5 can not receive bolt assembly further not hard up and shielded once more, improve equipment's use convenience.
Preferably, the clamping elastic piece 9 is partially embedded on the inner wall of the groove 8.
Specifically, the diameter of the inner arc of the boss 4 is not less than the outer diameter of the annular connecting piece.
A rail bolt looseness detection method is based on the bolt structure and is used for detecting through a method comprising the following steps: the method comprises the steps of obtaining picture information of a bolt on a track, judging whether a warning area on a bolt structure is in an exposed state or not according to the picture information, if part or all of the warning area is in the exposed state, indicating that the bolt is loosened, and if the warning area is not exposed, indicating that the bolt is not loosened.
Specifically, in the process of judging the picture information, intelligent analysis and judgment are carried out on the picture information by adopting a deep learning method, the deep learning method is realized on the basis of a Tensorflow deep learning framework, and a data source uses an MNIST data set and carries out model training by respectively adopting a softmax regression algorithm and CNN deep learning.
Specifically, the deep network structure for CNN deep learning comprises a convolutional layer, an excitation layer, a pooling layer and a full-connection layer
The convolutional layer is: the operation of performing inner product on the image and the filter matrix is convolution, the convolution layer can be used for acquiring the characteristics of the image through the convolution operation, the modules are stacked, different convolution kernels are used for acquiring higher-order characteristics, and in the stacked depth structure, the weight parameters of the convolution kernels are shared, so that the depth structure can acquire higher-dimensional characteristics, but the parameter quantity is not greatly increased;
the excitation layer is as follows: the excitation layer is not explicitly indicated in the figure, and is usually used for selecting an appropriate excitation function after the convolution operation to process the result of the convolution operation; the function of the activation function is to reserve and map the characteristics of the activated neurons through the function, so as to solve the nonlinear problem of the network structure;
the pooling layer is as follows: and the data is downsampled, so that overfitting can be avoided, and the operations such as data characteristics and data calculation amount are reduced.
The full connection layer: the connection of the layer is the connection of the neurons in the neural network structure between each layer; the fully connected layer plays a role of a classifier in the whole convolutional neural network; mapping original data to a hidden layer feature space through operations of a convolutional layer, a pooling layer, an activation function layer and the like, wherein a full connection layer is used for mapping learned distributed feature representation to a sample mark space; a layer of structure which can be customized after passing through the layer is used for classification or regression; the full connection layer is realized by acquiring more characteristics of high-dimensional images through a depth network structure to process the images, and controlling parameters through convolution operation.
Test examples
When the detection method is adopted to detect the loosening condition of the bolt structure on the track, the unmanned aerial vehicle is used for fixed-point back-and-forth aerial photography monitoring to acquire real-time images; the camera drivers of labview, NI Vision Development, Module Vision Development kit and NI Vision acquisition Software are selected, and the specific modules and processes comprise,
1. camera acquisition module
A VI is newly built, a front panel diagram is organized, and a program block diagram is opened; then, selecting an IMAQ Create data cache region and creating a character string constant for the IMAQ Create data cache region;
IMAQdx Open Camera: opening a camera, loading a camera configuration file, and creating a session handle of IMAQdx for the camera;
3. call IMAQdx Configure Grab: finishing the initialization of the Grab, executing the capture images continuously circulating on the buffer zone ring by the Grab, and acquiring the high-speed images by using the Grab VI;
4. calling IMAQdx Grab to acquire the latest image output of the current frame and creating an image display for the latest image output; this VI is only invoked after the IMAQdx configuration Grab VI is invoked; if the image type does not match the video format of the camera, this VI will change the image type to the appropriate format;
5. displaying pictures
Before displaying the picture, converting the collected picture into a gray picture by using an IMAQ ExtractSingleColorplane, thereby facilitating the subsequent image processing operation;
6. convolving acquired images
The convolution is carried out in a conditional structure, so that the function of convolution on the image can be realized by pressing a convolution button; if this condition is false, no action is done within the structure;
(1) firstly, a character string is taken to store a convolution kernel; then, calling IMAQ building Kernel, and constructing a convolution matrix through converting character strings;
(2) secondly, calling IMAQ volume, and filtering the image by using a linear filter; the output of the function is a convoluted picture; one of the two input pictures is the picture obtained in the acquisition step, the other picture is the picture obtained in the acquisition step, the picture is copied and stored in the other memory after being acquired, namely, a picture buffer zone is also established for storing the copy, and the other input is the convolution matrix obtained in the step (1); finally, a normalization factor is added, elements in the matrix are summed, and then the sum is divided by the normalization factor;
(3) copying and displaying pictures
Establishing a picture buffer area again, storing the convolved picture, copying the obtained convolved picture, otherwise replacing the convolved picture in a later while loop, and storing the picture which is not the convolved picture; the picture display calls a picture display window.
7. Saving the convolved pictures
Saving the convolved picture is in another conditional structure because the saved button function is to be implemented; calling a file dialog box for path selection during storage; then, calling the IMAQ Write BMP File, and storing the picture in a BMP format; saving to create a picture buffer a picture is copied to a buffer in a convolutional conditional structure and then saved by pulling a line in the buffer to an IMAQ Write BMP File.
8. Realize the real-time collection of pictures and the continuous operation of programs
Using a while loop to display the IMAQdx Grab and the subsequent pictures in the two condition structures and the picture acquisition, and then creating an input control under the loop condition to realize the exit button function of the program;
the back end carries out intelligent analysis and judgment on the acquired image information by adopting a deep learning method, the deep learning method is realized based on a Tensorflow deep learning framework, and a data source uses an MNIST data set and carries out model training by respectively adopting a softmax regression algorithm and CNN deep learning; whether the warning area on the bolt structure is in an exposed state or not is judged through a deep learning method, if part or all of the warning area is in the exposed state, the bolt is loosened, and if the warning area is not exposed, the bolt is not loosened, and the result is shown in fig. 5 and 6.
The foregoing is illustrative of the preferred embodiments of this invention, and it is to be understood that the invention is not limited to the precise form disclosed herein and that various other combinations, modifications, and environments may be resorted to, falling within the scope of the concept as disclosed herein, either as described above or as apparent to those skilled in the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. The utility model provides a bolt structure with become flexible and detect function, its characterized in that, includes nut and packing ring, the nut bottom is provided with shielding piece, be provided with boss and warning area on the packing ring, boss and warning area set up on same circumference, shielding piece includes first state and second state, and under the first state, the nut is in stable connection state, and shielding piece shields the warning area, and under the second state, the nut is in not hard up state, and shielding piece and warning area dislocation, warning area are in part or all exposure states.
2. The bolt structure with the loosening detection function as claimed in claim 1, wherein the nut is provided at a bottom thereof with an annular connecting member, and the shielding member is provided on a side wall of the annular connecting member.
3. The bolt structure with the loosening detection function according to claim 2, wherein the outer diameter of the annular connecting member is the same as the diameter of the circumscribed circle of the nut.
4. A bolt structure with a looseness detection function according to claim 2, wherein said shielding member is an arch structure.
5. The bolt structure with the function of detecting looseness according to claim 4, wherein the boss and the warning area are respectively of an arch structure.
6. The bolt structure with the loosening detection function as claimed in claim 4, wherein the shielding member has a 180 ° arch structure, and the boss and the warning region have 90 ° arch structures, respectively.
7. The bolt structure with the loosening detection function as claimed in claim 1, wherein the friction coefficient of the upper surface of the washer where the warning area is located is smaller than that of the lower surface friction system of the washer.
8. A rail bolt looseness detection method is characterized in that based on the bolt structure of any one of claims 1 to 7, detection is performed through a method comprising the following steps: the method comprises the steps of obtaining picture information of a bolt on a track, judging whether a warning area on a bolt structure is in an exposed state or not according to the picture information, if part or all of the warning area is in the exposed state, indicating that the bolt is loosened, and if the warning area is not exposed, indicating that the bolt is not loosened.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110427605.9A CN113236653B (en) | 2021-04-21 | 2021-04-21 | Bolt structure with loosening detection function and bolt loosening detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110427605.9A CN113236653B (en) | 2021-04-21 | 2021-04-21 | Bolt structure with loosening detection function and bolt loosening detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113236653A true CN113236653A (en) | 2021-08-10 |
| CN113236653B CN113236653B (en) | 2022-05-27 |
Family
ID=77128561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110427605.9A Active CN113236653B (en) | 2021-04-21 | 2021-04-21 | Bolt structure with loosening detection function and bolt loosening detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113236653B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114582104A (en) * | 2022-03-03 | 2022-06-03 | 国网山东省电力公司电力科学研究院 | Power transmission tower bolt looseness early warning device |
| CN116838693A (en) * | 2023-09-01 | 2023-10-03 | 广东电网有限责任公司广州供电局 | Early warning device for preventing fastener from loosening and power system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6789990B1 (en) * | 1998-10-16 | 2004-09-14 | Paul Anthony Reginald Harris | Fastener device |
| CN108071655A (en) * | 2016-11-16 | 2018-05-25 | 孙愉后 | Indicate the fastener of the loose amount of moving back |
| CN110246132A (en) * | 2019-06-23 | 2019-09-17 | 中车青岛四方车辆研究所有限公司 | Rail vehicle bolt looseness detection method and system |
| CN110242659A (en) * | 2018-03-07 | 2019-09-17 | 福建省晋江市励精汽配有限公司 | Fastener |
| CN110273909A (en) * | 2019-06-05 | 2019-09-24 | 上海应谱科技有限公司 | A kind of intelligent fastener and on-line monitoring system that can perceive fastener nut loosening |
| CN110349104A (en) * | 2019-07-08 | 2019-10-18 | 创新奇智(南京)科技有限公司 | A kind of method of the COMPUTER DETECTION with groove bolt looseness, computer-readable medium and system |
| CN211474627U (en) * | 2019-12-31 | 2020-09-11 | 重庆国翰能源发展有限公司 | Visual connecting piece of elasticity |
-
2021
- 2021-04-21 CN CN202110427605.9A patent/CN113236653B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6789990B1 (en) * | 1998-10-16 | 2004-09-14 | Paul Anthony Reginald Harris | Fastener device |
| CN108071655A (en) * | 2016-11-16 | 2018-05-25 | 孙愉后 | Indicate the fastener of the loose amount of moving back |
| CN110242659A (en) * | 2018-03-07 | 2019-09-17 | 福建省晋江市励精汽配有限公司 | Fastener |
| CN110273909A (en) * | 2019-06-05 | 2019-09-24 | 上海应谱科技有限公司 | A kind of intelligent fastener and on-line monitoring system that can perceive fastener nut loosening |
| CN110246132A (en) * | 2019-06-23 | 2019-09-17 | 中车青岛四方车辆研究所有限公司 | Rail vehicle bolt looseness detection method and system |
| CN110349104A (en) * | 2019-07-08 | 2019-10-18 | 创新奇智(南京)科技有限公司 | A kind of method of the COMPUTER DETECTION with groove bolt looseness, computer-readable medium and system |
| CN211474627U (en) * | 2019-12-31 | 2020-09-11 | 重庆国翰能源发展有限公司 | Visual connecting piece of elasticity |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114582104A (en) * | 2022-03-03 | 2022-06-03 | 国网山东省电力公司电力科学研究院 | Power transmission tower bolt looseness early warning device |
| CN116838693A (en) * | 2023-09-01 | 2023-10-03 | 广东电网有限责任公司广州供电局 | Early warning device for preventing fastener from loosening and power system |
| CN116838693B (en) * | 2023-09-01 | 2023-11-21 | 广东电网有限责任公司广州供电局 | Early warning device for preventing fastener from loosening and power system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113236653B (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113236653B (en) | Bolt structure with loosening detection function and bolt loosening detection method | |
| CN108109390A (en) | A kind of road traffic changeable driveway regulator control system and regulation and control method | |
| CN107909575B (en) | Binocular vision on-line detection device and detection method for running state of vibrating screen | |
| DE112019000630T5 (en) | System and method for determining the work of a work vehicle and method for generating a taught-in model | |
| CN111126460B (en) | Automatic pavement disease inspection method, medium, equipment and device based on artificial intelligence | |
| DE102005023672A1 (en) | Method for controlling a machine with electric drive | |
| CN106228812A (en) | Illegal vehicle image-pickup method and system | |
| CN111626382A (en) | Rapid intelligent identification method and system for cleanliness of vehicle on construction site | |
| CN109410583A (en) | Traffic information collection and analysis system and operation method based on automobile data recorder | |
| CN111338256A (en) | Intelligent sanitation vehicle monitoring system of intelligent sanitation cloud platform | |
| CN108665699A (en) | There is the method and device in place in a kind of prediction vehicle | |
| CN115759471A (en) | AI photovoltaic cleaning prediction system based on multi-mode fusion | |
| JPH05332915A (en) | Hydrosphere monitoring apparatus and purifying apparatus | |
| CN109131239B (en) | Vehicle cleaning parameter setting method and vehicle washing machine | |
| CN2896399Y (en) | Electronic police device for monitoring running red-light | |
| CN117902417A (en) | Safety control method, device, equipment and medium for elevator | |
| CN112967259B (en) | Plasma arc welding perforation state prediction method and system based on molten pool image | |
| CN1874500A (en) | Monitor and control alarm device in infrared video, and method | |
| CN209671933U (en) | A kind of highway digitlization construction video panorama monitoring device | |
| CN112181331A (en) | Consistency detection system for published contents of variable information publishing terminal | |
| Hicks et al. | A strategy for integrated monitoring | |
| CN113089607B (en) | Intelligent river surge cleaning system and control method | |
| Hu et al. | Design of pipe-inspection robot based on yolov3 | |
| CN207232721U (en) | A kind of prefabricated pumping plant of intelligent integral | |
| CN211577693U (en) | Monitoring device system for improving government to control raise dust by 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230926 Address after: 056000 West of Longcao Village, Liuying Town, Yongnian District, Handan City, Hebei Province Patentee after: Hebei Shouhong Metal Products Co.,Ltd. Address before: 230000 floor 1, building 2, phase I, e-commerce Park, Jinggang Road, Shushan Economic Development Zone, Hefei City, Anhui Province Patentee before: Dragon totem Technology (Hefei) Co.,Ltd. Effective date of registration: 20230926 Address after: 230000 floor 1, building 2, phase I, e-commerce Park, Jinggang Road, Shushan Economic Development Zone, Hefei City, Anhui Province Patentee after: Dragon totem Technology (Hefei) Co.,Ltd. Address before: 610031 No. two, section 111, ring road, Chengdu, Sichuan, China Patentee before: SOUTHWEST JIAOTONG University |
|
| TR01 | Transfer of patent right |