CN109521021A - A kind of nuclear power plant containment shell appearance inspecting system and method - Google Patents
A kind of nuclear power plant containment shell appearance inspecting system and method Download PDFInfo
- Publication number
- CN109521021A CN109521021A CN201811493072.9A CN201811493072A CN109521021A CN 109521021 A CN109521021 A CN 109521021A CN 201811493072 A CN201811493072 A CN 201811493072A CN 109521021 A CN109521021 A CN 109521021A
- Authority
- CN
- China
- Prior art keywords
- defect
- appearance
- climbing robot
- containment
- control instruction
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8861—Determining coordinates of flaws
- G01N2021/8864—Mapping zones of defects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8867—Grading and classifying of flaws using sequentially two or more inspection runs, e.g. coarse and fine, or detecting then analysing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8874—Taking dimensions of defect into account
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/888—Marking defects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Manipulator (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The present invention provides a kind of nuclear power plant containment shell appearance inspecting system and methods, the appearance inspecting system includes units of ground station and climbing robot, units of ground station is used to send the first control instruction to climbing robot, climbing robot obtains the appearance images of containment according to the first control instruction, and the appearance images are sent to units of ground station；Units of ground station judges containment appearance with the presence or absence of defect according to appearance images, and if it exists, then issues the second control instruction to climbing robot；Climbing robot reacquires defect photo according to the second control instruction and current location information is sent to units of ground station；Units of ground station obtains flaw labeling coding according to defect photo and location information, and controls the climbing robot and defect is marked according to flaw labeling coding.Nuclear power plant containment shell appearance inspecting system provided by the invention and method realize the automatic identification processing of containment open defect, and measurement accuracy is high, and measurement result is stablized, and have good technical effect.
Description
Technical field
The present invention relates to technical field of nuclear power more particularly to a kind of nuclear power plant containment shell appearance inspecting system and methods.
Background technique
Nuclear power plant containment shell is the cylindrical prestressed reinforced concrete structure for having torispheric domes, is for hindering
Keep off last one of barrier that fission product and primary Ioops radioactive substance from fuel enter environment.Since complicated pours work
Skill and environmental factor will lead to the carbonization of containment concrete surface, the defects of crack or breakage occur, cause reinforcing bar and prestressing force
System corrosion influences its service life, it is therefore necessary to periodically open defect inspection is carried out to containment, to fully understand containment
The defect distribution of concrete outer surface verifies the build guality of containment structure.
At present, the main contents of containment visual examination are concrete surface and are hidden under surface defect in concrete
All kinds of defects in portion, and for the horizontal, perpendicular of measurement item mainly crannied width, length and the crack of rift defect design
It is positioned to inclination angle.Existing detection methods mainly in such a way that visual inspection and equipment inspection combine, use appearance
It checks and hanging basket and sets up scaffold, it is artificial to carry out visual examination close to concrete surface, and record security shell defect kind and split
The parameters such as slit width degree, length, form.However, nuclear power plant containment shell is up to 60m, it is big not only to set up scaffold preparation amount, at
This valuableness, visual examination personnel are there is also the security risk of falling from high altitude, there is also that can not carry out test under bad weather, in turn
The problems such as influencing the project plan；Meanwhile artificial eye observation resultant error is big, also requires a great deal of time after the test
Data are arranged, so that artificial range estimation has significant limitation.
Summary of the invention
There is significant limitation using artificial visual inspection existing containment visual examination in the present invention, mention
A kind of nuclear power plant containment shell appearance inspecting system and method have been supplied, defect can have been carried out to containment appearance and automaticly inspected, get rid of people
The limitation of work inspection.
The present invention is used to solve the technical solution of the above technical problem are as follows: provides a kind of nuclear power plant containment shell visual examination system
System, comprising:
Units of ground station, for sending the first control instruction to climbing robot；
The climbing robot, for obtaining the appearance images of containment according to first control instruction, and will be described
Appearance images are sent to units of ground station；
The units of ground station, for judging containment appearance with the presence or absence of defect according to the appearance images, and if it exists,
Then the second control instruction is sent to the climbing robot；
The climbing robot, for reacquiring defect photo and current position letter according to second control instruction
Breath is sent to the units of ground station；
The units of ground station for obtaining flaw labeling coding according to the defect photo and location information, and controls
The climbing robot is marked defect according to flaw labeling coding.
Wherein, the units of ground station includes:
Processing center, for sending the first control instruction to climbing robot according to setting mode, first control refers to
Order includes the automatic control instruction for manually controlling instruction with sending in automatic mode sent under manual mode.
Wherein, the climbing robot includes:
Image acquisition unit, for obtaining the appearance images and defect photo；
Control unit obtains the appearance of containment for manually controlling instruction control described image acquiring unit according to described in
Image, and the appearance images are sent to the units of ground station in real time；Or,
For obtaining the appearance images of containment according to the instruction control described image acquiring unit that automatically controls, and lead to
Crossing the appearance images judges containment appearance with the presence or absence of defect；If it exists, then the appearance images that defect will be present are sent to
The units of ground station；If it does not exist, then the climbing robot is controlled to be continuously advanced.
Wherein, the units of ground station further include:
The processing center, for receiving the appearance images of the appearance images or the existing defects；
Determining defects system, for being determined outside containment by the appearance images or the appearance images of the existing defects
It sees and whether there is defect；If it exists, then the second control instruction is issued to the climbing robot；Otherwise, Xiang Suoshu wall-climbing device
Human hair goes out advancement commands.
Wherein, the units of ground station further include:
The processing center, for receiving the defect photo；
Defect recognition system connects the processing center, for being based on image nature splicing from the defect photo
Middle extraction defect information, the defect information include defect type and flaw size.
Wherein, the units of ground station further include:
The processing center, for receiving the location information；
Positioning system connects the processing center, for the location information to be converted into coordinate information.
Wherein, the units of ground station further include:
Database Systems connect the defect recognition system and positioning system, are used for the defect information and its correspondence
Coordinate information carry out classification storage according to setting rule, obtain flaw labeling coding；
The processing center is also used to send spraying instruction to described control unit, climbs described in described control unit control
Wall robot is marked defect according to flaw labeling coding.
Wherein, the units of ground station further include:
Defect distribution drawing system connects the Database Systems, for believing the coordinate according to the coordinate information
It ceases corresponding defect information and flaw labeling coding is plotted on two-dimensional containment outer cylinder, form containment open defect cloth
Set figure.
On the other hand, a kind of nuclear power plant containment shell appearance inspection method is also provided, comprising steps of
S1, the first control instruction is sent to climbing robot；
S2, climbing robot obtain the appearance images of containment according to first control instruction；
S3, judge containment appearance with the presence or absence of defect by the appearance images；If it exists, then to the wall-climbing device
Human hair goes out the second control instruction；
S4, climbing robot reacquire defect photo and current location information according to second control instruction；
S5, flaw labeling coding is obtained according to the defect photo and location information, and controls the climbing robot and presses
Defect is marked according to flaw labeling coding.
Wherein, step S1 includes:
The first control instruction is sent to climbing robot according to setting mode；It is described when the mode that sets is manual mode
First control instruction is to manually control instruction；When the mode that sets is automatic mode, first control instruction is to automatically control
Instruction.
Wherein, step S2 includes:
The climbing robot manually controls the appearance images that instruction obtains containment according to described in, and in real time described in transmission
Appearance images；Or,
The climbing robot instructs the appearance images for obtaining containment according to described automatically control, and according to the appearance
Image judges containment appearance with the presence or absence of defect；If it exists, then the appearance images of existing defects are sent；If it does not exist, then institute
Climbing robot is stated to be continuously advanced.
Wherein, step S3 includes:
Receive the appearance images of the appearance images or the existing defects；
Determine that containment appearance whether there is defect by the appearance images or the appearance images of the existing defects；If
In the presence of then to the climbing robot the second control instruction of sending；Otherwise, Xiang Suoshu climbing robot issues advancement commands.
Wherein, step S5 includes:
The defect photo is received, and defect information is extracted from the defect photo based on image nature splicing,
The defect information includes defect type and flaw size；
The location information is received, and the location information is converted into coordinate information.
Wherein, step S5 further include:
The defect information and corresponding coordinate information are subjected to classification storage according to setting rule, obtain the defect mark
Note coding；
Spraying instruction is sent to the climbing robot, controls the climbing robot according to flaw labeling coding pair
Defect is marked；
The corresponding defect information of the coordinate information and flaw labeling coding are plotted in two dimension according to the coordinate information
On containment outer cylinder, containment open defect layout drawing is formed.
Implement technical solution provided by the invention to have the benefit that
The appearance inspecting system and method realize containment appearance by climbing robot and pattern recognition technique and lack
The automatic identification processing for falling into detection, during solving the with high costs but inefficiency and test that former testing program sets up scaffold
The risk of testing crew falling from high altitude, also solving the region that former testing program scaffold and hanging basket can not reach can not carry out directly
The problem of connecing defectoscopy；Meanwhile appearance inspecting system and method are high using the precision of automatic identification processing, measurement result is steady
It is fixed, it is big to solve the problems, such as that former testing program personnel visually observe test result error；It on the other hand, can be by after defect recognition
Defect is marked automatically according to flaw labeling coding, according to the flaw labeling that is sprayed encode can quick positioning defect areas,
Convenient for follow-up tester scene defect verification.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples, in attached drawing:
Fig. 1 is the module diagram of appearance inspecting system provided in an embodiment of the present invention；
Fig. 2 is the structural schematic diagram of appearance inspecting system provided in an embodiment of the present invention；
Fig. 3 is provided in an embodiment of the present invention to carry out diminution effect picture to original image using image gridding；
Fig. 4 (1), 4 (2), 4 (3), 4 (4) are 4 kinds of directions crack provided in an embodiment of the present invention template schematic diagram respectively；
Fig. 5 is image size reduction effect figure provided in an embodiment of the present invention；
Fig. 6 is image splicing effect figure in crack provided in an embodiment of the present invention；
Fig. 7 is the flow chart of appearance inspection method provided in an embodiment of the present invention；
Fig. 8 is the detailed step flow chart of appearance inspection method provided in an embodiment of the present invention.
Specific embodiment
In order to enable those skilled in the art that the present invention is more clearly understood, below in conjunction with attached drawing and specific implementation
Example does further detailed description to the present invention.
In order to solve to need to set up scaffold, artificial close concrete to carry out containment visual examination in the prior art
Surface carries out visual examination, leads to that preparation amount is big, risk is high, measurement result is unstable, the present invention is intended to provide
A kind of nuclear power plant containment shell appearance inspecting system and method, climbing robot obtain containment outside drawing according to the first control instruction
Then appearance images are sent to units of ground station and judge containment appearance with the presence or absence of defect by picture；If it exists, then to climbing wall machine
Device human hair send the second control instruction, and climbing robot reacquires defect photo and current location information is sent to earth station's list
Member, units of ground station is based on the defect photo by image recognition technology and location information obtains defect information and coordinate letter
Breath, and carry out classification storage according to setting rule and obtain flaw labeling coding, control climbing robot is encoded according to flaw labeling
Defect is marked, the automatic identification processing of containment open defect detection is realized, artificial visual inspection is overcome and brings
Limitation.
Embodiment one
As shown in Figure 1, the present embodiment provides a kind of nuclear power plant containment shell appearance inspecting system, including 10 He of units of ground station
Climbing robot 20, units of ground station 10 and climbing robot 20 are established wireless telecommunications by data transmission unit 30 and are connected.
Specifically, units of ground station 10, which is used to connected by the wireless telecommunications to climbing robot 20, sends the first control
Instruction, climbing robot 20 are used to obtain the appearance images of containment according to first control instruction, and by the outside drawing
As being sent to units of ground station 10；
Units of ground station 10 is used to judge containment appearance with the presence or absence of defect according to the appearance images, and if it exists, then
The second control instruction is issued to climbing robot 20；
Climbing robot 20 is used to reacquire defect photo and current location information according to second control instruction
It is sent to units of ground station 10；
Units of ground station 10 is used to obtain flaw labeling coding according to the defect photo and location information, and controls and climb wall
Robot is marked defect according to flaw labeling coding.
Specifically, units of ground station 10 includes the processing center 11 of connection data transmission unit 30,11 basis of processing center
Setting mode sends the first control instruction to climbing robot, and first control instruction is included in the mode of setting as manual mode
When send manually control sent under instruction and automatic mode automatic control instruction.
Climbing robot 20 further includes the control unit 21 for connecting data transmission unit 30, and connects control unit 21
Image acquisition unit 22.
Wherein, image acquisition unit 22 is preferably video camera, is scanned and takes pictures for the appearance to containment, to obtain
Take the appearance images and defect photo；
Control unit 21 is used to manually control the appearance that instruction control image acquisition unit 22 obtains containment according to described in
Image, and the appearance images are sent to units of ground station 10 in real time, units of ground station 10 judges according to the appearance images
Containment appearance whether there is defect；
Control unit 21 is also used to obtain the outer of containment according to the instruction control image acquisition unit 22 that automatically controls
Image is seen, and judges containment appearance with the presence or absence of defect by the appearance images according to preset image recognition technology；If
In the presence of the appearance images that defect then will be present are sent to units of ground station 10；If it does not exist, then climbing robot 20 is controlled along peace
It persistently advances full shell outer surface.
Further, as shown in connection with fig. 2, climbing robot 20 is that can complete vertical walls row in containment concrete surface
The mobile platform walked, with excellent wall adsorption power and bearing capacity.In the present embodiment, image acquisition unit 22 passes through machinery
Arm 221 is mounted on climbing robot 20, and mechanical arm 221 is electrically connected described control unit 21, for according to control unit 21
Second control instruction adjusts the position of camera, to obtain apparent, comprehensive containment appearance images.
Climbing robot 20 further includes the driving device 23 being electrically connected respectively with control unit 21, height above sea level meter 24, inclinator
25 and spraying mechanism 26.
Wherein, the operating instruction driving climbing robot 20 that driving device 23 is used to send according to control unit 21 is in safety
The walking of shell wall face, completes the direction of advance and speed control of climbing robot；In the present embodiment, driving device 23 is preferably turbine
Driving；
Height above sea level meter 24 is used to obtain the vertical height of climbing robot 20；Inclinator 25 is for obtaining climbing robot 20
Level angle；It, will be described after control unit 21 receives the vertical height and level angle since containment barrel is cylinder
Vertical height and its corresponding level angle form the location information and are sent to units of ground station 10.Preferably, height above sea level meter 24
It is set to inside climbing robot 20, influence of the extraneous wind pressure to height above sea level measurement accuracy of measurement can be reduced；
Spraying mechanism 26 includes the actuated by cams machine of the spray tank being mounted on climbing robot and connection control unit 21
Structure, control unit 21 instructs control cam drive mechanism to press the spray tank according to spraying, outside to the containment recognized
Defect is seen to be marked according to flaw labeling coding.
Climbing robot provided in this embodiment further includes falling proof device comprising hoist engine and the connection hoist engine
With the safety rope of climbing robot 20, the risk of robot falling from high altitude can be avoided under equipment fault lamp abnormal conditions, protected
Protect equipment and personnel safety.
Further, units of ground station 10 further includes the determining defects system 12 for being separately connected processing center 11, defect knowledge
Other system 13, positioning system 14, Database Systems 15 and defect distribution drawing system 16.
Specifically, processing center 11 is used to connect the reception appearance images or described in the presence of scarce by the wireless telecommunications
Sunken appearance images；
Determining defects system 12 connects processing center 11, for the appearance by the appearance images or the existing defects
Spectral discrimination containment appearance whether there is defect；If it exists, then second is issued to climbing robot 20 by processing center 11
Control instruction；Otherwise, advancement commands are issued to climbing robot 20, control climbing robot 20 moves on.
In the present embodiment, determining defects system 12 is artificial decision-making system, when the first control instruction is to manually control instruction
When, it whether there is open defect in the appearance images of 10 real-time reception of artificial judgment units of ground station；When the first control instruction is
When automatically controlling instruction, whether the appearance images for the existing defects that manual confirmation climbing robot 20 is sent are true, in this way, passing through
Climbing robot and the dual test manually reaffirmed can greatly avoid erroneous judgement problem, improve and check efficiency.
In other embodiments of the invention, determining defects system 12 can determine safety based on image recognition processing algorithm
Shell appearance whether there is defect.Certainly, determining defects system 12 can also be used image recognition processing algorithm and mutually tie with artificial judgment
Close, these all within protection scope of the present invention, this embodiment is not repeated.
Processing center 11, which is also used to connect by the wireless telecommunications, receives the defect photo；Defect recognition system 13 connects
Processing center 11 is connect, for extracting defect information, the defect letter from the defect photo based on image nature splicing
Breath includes defect type and flaw size.
It should be noted that inspection system provided in this embodiment includes determining defects system 12 and defect recognition system 12
Two identifying systems, for the presence or absence of preliminary judgement defect, defect recognition system 12 is used to lack determining determining defects system 12
After sunken presence, defect photo is reacquired by adjusting image acquisition unit 22, defect letter is obtained by analyzing defect photo
Breath.In this way, being used cooperatively by two systems, can not only guarantee to check efficiency, additionally it is possible to obtain more accurate analysis knot
Fruit reduces error.
Specifically, described image nature splicing the following steps are included:
S01, image enhancement processing is carried out to the defect photo, avoids in collection process that there are certain noises, and
The extremely strong interference that the detection of a large amount of scratch fractures of cement wall surface generates；
Described image enhancing processing includes image filtering and image gridding；Wherein, image filtering can reduce noise jamming,
Improving image quality, to improve Crack Detection precision.Image filtering processing is also known as image smoothing, if smoothly improper can make
The profile with lines etc. in crack become blurred in image.Defect recognition system 13 be respectively adopted mean filter, median filtering,
Several algorithms such as Wiener filtering handle wall surface crack image, and compare and analyze to test result, so that it is determined that most
Suitable filtering algorithm.
As shown in connection with fig. 3, described image nature splicing is preliminary in addition to being carried out using filtering algorithm in image enhancement
Denoising is outer, while also introducing image gridding method, carries out a certain proportion of diminution to image, weakens big block noise in image
Interference, eliminate the influence of isolated noise.Save primary fracture information from damage substantially in diminution figure, and is still retained in figure after reducing
Noise isolatism and locality as in can be more significant.
S02, binary conversion treatment is carried out to enhanced image；
Image binaryzation processing is that a width gray level image is converted into bianry image according to certain threshold value, generallys use threshold
It is worth cutting techniques.Wall surface crack image is handled using histogram thresholding split plot design and inter-class variance thresholding method,
During image binaryzation, according to the gray scale and directional characteristic in crack, it is based on classical edge detection method design template, from each side
To gray difference degree on judge crack there are a possibility that and directionality.
In conjunction with shown in Fig. 4 (1) -4 (4), the gamma characteristic and morphological character in crack are considered, the crack in image is more thin
Long pixel set has certain contrast with adjacent area in gray scale, therefore passes through pair between each pixel and neighborhood
Ratio can be used as whether this point belongs to the judgment basis in crack, then using 0 °, 45 °, 90 °, 135 ° totally 4 kinds of directions split
Slit die plate operator carries out adjacent slits point detection to 8 angles of each candidate pixel point to judge fractuer direction.
S03, crack is judged whether there is from the binary conversion treatment result of image；If it exists, then crack information is extracted；
Specifically, system extracts crack using length information constraint and region elongation technology, then also by image size
Original, and it is more natural by the shape in crack using morphologic dilation erosion arithmetic processing method.Since typical crack all has
There is enough length, system uses clustering algorithm to obtain a series of " regions " divided first, then by zoning
Candidate crack pixel quantity removes the noise jamming being made of a small amount of sparse pixel, mentions to extend for subsequent region
For ensureing；Heuristic search is finally used, is carried out by selected seed region and corresponding heuristic Neighborhood-region-search algorithm
Effective extension of crack area, until meeting algorithm termination condition.As shown in figure 5, will also contract before this after completing crack extract
Small image restoring is to original scale.If the pixel before restoring in image is non-crack point, institute in grid is corresponded to after reduction
There is pixel to be set as 0；Otherwise the pixel before restoring in image is crack point, then corresponds to all pixels point in grid after restoring
It is set as 255.
S04, crack carry out naturalization processing；
As shown in connection with fig. 6, i.e., fracture coupling part carries out naturalization processing, splits coupling part visually with whole
It stitches smoothless linear, has random burr, therefore use morphologic dilation erosion algorithm, obtain most similar splits
Slit state, and image mosaic is carried out to the longer crack beyond camera acquisition area to realize the whole display function in crack
Can, gained image and fracture shape, the almost the same crack of length are with uniformity, and measurement accuracy is high, improve inspection result
Reliability.
Further, processing center 11 is also used to receive the location information；Positioning system 14 connects processing center 11, uses
In the location information is converted into two-dimensional coordinate information, the positioning to climbing robot is realized.
Database Systems 15 connect defect recognition system 13 and positioning system 14, are used for the defect information and its correspondence
Coordinate information carry out classification storage according to setting rule, obtain flaw labeling coding；
Processing center 11 be also used to control unit 21 send spraying instruction, control unit 11 control spraying mechanism 26 according to
Defect is marked in the flaw labeling coding.
In the present embodiment, the flaw labeling coding that Database Systems 15 are set generally is named with ABC, and wherein A represents defect
Locating inspection area, B represent defect serial number, and C represents defect kind；Defect inspection area code number is as shown in table 1 below, safety
Shell open defect type is as shown in table 2 below.In practical applications, system is in inter-process marking of defects, flaw labeling coding
DOM001FI is the 1st crack for representing dome region, and flaw labeling coding NER005CO is the 5th for representing counterfort columnar region
Corrosion, solve the problems, such as that conventional concrete defect can not encode or title is too complicated, convenient for internal data unified management and
Distribution prevents error simultaneously, can encode quick positioning defect areas according to flaw labeling, be convenient for follow-up tester scene defect
It verifies.
The definition of 1 inspection area code name of table
2 containment open defect code definition of table
Wherein, crack refers mainly to whether concrete outer surface has crack, should check its corresponding length, width for crack
Degree, shape, orientation and its development tendency and recovery；
Corrosion be mainly the steel members such as concrete surface embedded part, penetration piece engaging portion with the presence or absence of antifouling paint it is not in place or
It is corroded caused by lacquer painting is damaged；
Seepage flow trace refers to that concrete forms the trace of water or corrosion because of seepage flow；
Stone hole nest pitted skin just refers to that concrete surface compares the concrete defect of shallow-layer, i.e., usually said stone cheats nest
And voids and pits；
Face checking refers to that concrete is non-structure stress blind crack because of surface checking caused by drying shrinkage；
Spillage when cement fragments refer to casting concrete and the cement layer thin slice for forming concrete surface；
Other defects the defects of for example concrete outer surface is mingled with silt particle, sundries.
Further, defect distribution drawing system 16 connects Database Systems 15, for according to the coordinate information by institute
It states the corresponding defect information of coordinate information and flaw labeling coding is plotted on two-dimensional containment outer cylinder, formed outside containment
See defect layout drawing.
In the present embodiment, defect distribution drawing system 16 be when completing or being partially completed containment visual examination task,
The defect information on containment outer cylinder the functional module that two dimension is redrawn be can be subjected to automatically, containment inspection and history are convenient for
Data are checked.After calling the functional module, defect distribution drawing system automatically by the defects of Database Systems information and its
Corresponding coordinate information, flaw labeling coding extract, and then draw, mark on two-dimensional containment outer cylinder
Defect information present on entire containment outer cylinder and corresponding flaw labeling coding, facilitate monitoring personnel to check.Preferably,
Output format uses CAD drafting method.
Further, units of ground station provided in this embodiment further includes motion controller 17, motion controller 17 and control
Unit 21 processed direct wireless telecommunications connection manually controls instruction to the sending of climbing robot 20 for the heart 11 in processes and climbs backward
Wall robot 20 sends instruction of advancing, and realizes the motion control to climbing robot 20, climbing robot 20 is according to the traveling
Instruction is walked.
In the present embodiment, motion controller 17 is preferably hand-held remote control unit, and controllable unit 21 directly establishes channel radio
News connection.Certainly, motion controller 17 may alternatively be integrated in processing center 11, the motion control for climbing robot.
The embodiment of the present invention realizes containment appearance inspecting system and realizes using climbing robot and earth station are simple eye
The automatic identification of open defect is handled, and capable of quickly being judged whether there is defect on containment, achieving good skill
Art effect.
Embodiment two
As shown in fig. 6, the present embodiment provides a kind of nuclear power plant containment shell appearance inspection methods, comprising steps of
S1, the first control instruction is sent to climbing robot；
S2, climbing robot obtain the appearance images of containment according to first control instruction；
S3, judge containment appearance with the presence or absence of defect by the appearance images；If it exists, then to the wall-climbing device
Human hair goes out the second control instruction；
S4, climbing robot reacquire defect photo and current location information according to second control instruction；
S5, flaw labeling coding is obtained according to the defect photo and location information, and controls the climbing robot and presses
Defect is marked according to flaw labeling coding.
Specifically, step S1 includes:
The first control instruction is sent to climbing robot according to setting mode；It is described when the mode that sets is manual mode
First control instruction is to manually control instruction；When the mode that sets is automatic mode, first control instruction is to automatically control
Instruction.
Further, step S2 is specifically included:
Climbing robot manually controls instruction according to described in and obtains the appearance images of containment, and sends the appearance in real time
Image；
Climbing robot instructs the appearance images for obtaining containment according to described automatically control, and according to the appearance images
Judge containment appearance with the presence or absence of defect；If it exists, then the appearance images of existing defects are sent；If it does not exist, then wall machine is climbed
Device people is continuously advanced according to setting path along containment outer surface；
S3, judge containment appearance with the presence or absence of defect by the appearance images；If it exists, then to the wall-climbing device
Human hair goes out the second control instruction；
Further, step S3 is specifically included:
The appearance images of the appearance images or the existing defects are received, and pass through the appearance images or the presence
The appearance images of defect determine that containment appearance whether there is defect；If it exists, then the second control is issued to the climbing robot
System instruction；If it does not exist, then advancement commands are issued to the climbing robot, controls the climbing robot and moves on.
Further, step S4 is specifically included:
Position of the climbing robot according to second control instruction adjustment camera, reacquires defect photo；
S41, the climbing robot obtain the vertical height and level angle of current location according to second control instruction, and by institute
It states vertical height and its corresponding level angle forms the location information.
Further, step S5 is specifically included:
S51, the defect photo is received, and defect letter is extracted from the defect photo based on image nature splicing
Breath, the defect information includes defect type and flaw size；
S52, the location information is received, and the location information is converted into coordinate information；
S53, the defect information and corresponding coordinate information are subjected to classification storage according to setting rule, obtain described lack
Fall into label coding；
S54, Xiang Suoshu climbing robot send spraying instruction, control the climbing robot and compile according to the flaw labeling
Defect is marked in code；S55, the corresponding defect information of the coordinate information and flaw labeling coding are drawn according to the coordinate
System forms containment open defect layout drawing on two-dimensional Security shell outer cylinder.
It should be noted that appearance inspection method provided in this embodiment may refer to the appearance introduced in previous embodiment
Inspection system, details are not described herein again.
In conclusion a kind of nuclear power plant containment shell appearance inspecting system provided in this embodiment and method utilize wall-climbing device
People and pattern recognition technique realize the automatic identification processing of containment open defect detection, solve former testing program and set up foot
The risk of testing crew falling from high altitude during the with high costs but inefficiency and test of hand cradle, while also solving former test side
The region that case scaffold and hanging basket can not reach can not carry out the problem of direct defectoscopy；Meanwhile appearance inspecting system and side
Method is high using the precision of automatic identification processing, and measurement result is stablized, and solves former testing program personnel and visually observes test result
The big problem of error；On the other hand, it can be encoded according to flaw labeling after defect recognition and automatic marking is carried out to defect, and can be certainly
It is dynamic to draw containment open defect layout drawing, solve asking for the huge workload that containment general defect is drawn after the test
Topic, while quick positioning defect areas can be encoded according to flaw labeling, it is verified convenient for follow-up tester scene defect information.
It should be understood that for those of ordinary skills, it can be modified or changed according to the above description,
And all these modifications and variations should all belong to the protection domain of appended claims of the present invention.
Claims (14)
1. a kind of nuclear power plant containment shell appearance inspecting system characterized by comprising
Units of ground station, for sending the first control instruction to climbing robot；
The climbing robot, for obtaining the appearance images of containment according to first control instruction, and by the appearance
Image is sent to units of ground station；
The units of ground station, for according to the appearance images judge containment appearance whether there is defect, and if it exists, then to
The climbing robot sends the second control instruction；
The climbing robot, for reacquiring defect photo and current location information hair according to second control instruction
It send to the units of ground station；
The units of ground station, for obtaining flaw labeling coding according to the defect photo and location information, and described in control
Climbing robot is marked defect according to flaw labeling coding.
2. appearance inspecting system according to claim 1, which is characterized in that the units of ground station includes:
Processing center, for sending the first control instruction, the first control instruction packet to climbing robot according to setting mode
Include the automatic control instruction for manually controlling instruction with sending in automatic mode sent in a manual mode.
3. appearance inspecting system according to claim 2, which is characterized in that the climbing robot includes:
Image acquisition unit, for obtaining the appearance images and defect photo；
Control unit obtains the outside drawing of containment for manually controlling instruction control described image acquiring unit according to described in
Picture, and the appearance images are sent to the units of ground station in real time；Or,
For obtaining the appearance images of containment according to the instruction control described image acquiring unit that automatically controls, and pass through institute
Stating appearance images judges containment appearance with the presence or absence of defect；If it exists, then will be present defect appearance images be sent to it is described
Units of ground station；If it does not exist, then the climbing robot is controlled to be continuously advanced.
4. appearance inspecting system according to claim 2, which is characterized in that the units of ground station further include:
The processing center, for receiving the appearance images of the appearance images or the existing defects；
Determining defects system, for determining that containment appearance is by the appearance images or the appearance images of the existing defects
No existing defects；If it exists, then the second control instruction is issued to the climbing robot；Otherwise, Xiang Suoshu wall-climbing device human hair
Advancement commands out.
5. appearance inspecting system according to claim 4, which is characterized in that the units of ground station further include:
The processing center, for receiving the defect photo；
Defect recognition system connects the processing center, for being mentioned from the defect photo based on image nature splicing
Defect information is taken, the defect information includes defect type and flaw size.
6. appearance inspecting system according to claim 5, which is characterized in that the units of ground station further include:
The processing center, for receiving the location information；
Positioning system connects the processing center, for the location information to be converted into coordinate information.
7. appearance inspecting system according to claim 6, which is characterized in that the units of ground station further include:
Database Systems connect the defect recognition system and positioning system, are used for the defect information and its corresponding seat
It marks information and carries out classification storage according to setting rule, obtain the flaw labeling coding；
The processing center is also used to send spraying instruction to described control unit, climbs wall machine described in described control unit control
Device people is marked defect according to flaw labeling coding.
8. appearance inspecting system according to claim 7, which is characterized in that the units of ground station further include:
Defect distribution drawing system connects the Database Systems, for according to the coordinate information by the coordinate information pair
Defect information and the flaw labeling coding answered are plotted on two-dimensional containment outer cylinder, form containment open defect arrangement
Figure.
9. a kind of nuclear power plant containment shell appearance inspection method, which is characterized in that comprising steps of
S1, the first control instruction is sent to climbing robot；
S2, climbing robot obtain the appearance images of containment according to first control instruction；
S3, judge containment appearance with the presence or absence of defect by the appearance images；If it exists, then to the wall-climbing device human hair
Second control instruction out；
S4, climbing robot reacquire defect photo and current location information according to second control instruction；
S5, flaw labeling coding is obtained according to the defect photo and location information, and controls the climbing robot according to institute
Flaw labeling coding is stated defect is marked.
10. appearance inspection method according to claim 9, which is characterized in that step S1 includes:
The first control instruction is sent to climbing robot according to setting mode；When the mode that sets is manual mode, described first
Control instruction is to manually control instruction；When the mode that sets is automatic mode, first control instruction is to automatically control instruction.
11. appearance inspection method according to claim 10, which is characterized in that step S2 includes:
The climbing robot manually controls instruction according to described in and obtains the appearance images of containment, and sends the appearance in real time
Image；Or,
The climbing robot instructs the appearance images for obtaining containment according to described automatically control, and according to the appearance images
Judge containment appearance with the presence or absence of defect；If it exists, then the appearance images of existing defects are sent；If it does not exist, then described to climb
Wall robot is continuously advanced.
12. appearance inspection method according to claim 11, which is characterized in that step S3 includes:
Receive the appearance images of the appearance images or the existing defects；
Determine that containment appearance whether there is defect by the appearance images or the appearance images of the existing defects；If depositing
Then the second control instruction is being issued to the climbing robot；Otherwise, Xiang Suoshu climbing robot issues advancement commands.
13. appearance inspection method according to claim 9, which is characterized in that step S5 includes:
The defect photo is received, and defect information is extracted from the defect photo based on image nature splicing, it is described
Defect information includes defect type and flaw size；
The location information is received, and the location information is converted into coordinate information.
14. appearance inspection method according to claim 13, which is characterized in that step S5 further include:
The defect information and corresponding coordinate information are subjected to classification storage according to setting rule, the flaw labeling is obtained and compiles
Code；
Spraying instruction is sent to the climbing robot, the climbing robot is controlled and encodes according to the flaw labeling to defect
It is marked；
The corresponding defect information of the coordinate information and flaw labeling coding are plotted in two-dimensional Security according to the coordinate information
On shell outer cylinder, containment open defect layout drawing is formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811493072.9A CN109521021A (en) | 2018-12-07 | 2018-12-07 | A kind of nuclear power plant containment shell appearance inspecting system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811493072.9A CN109521021A (en) | 2018-12-07 | 2018-12-07 | A kind of nuclear power plant containment shell appearance inspecting system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109521021A true CN109521021A (en) | 2019-03-26 |
Family
ID=65794890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811493072.9A Pending CN109521021A (en) | 2018-12-07 | 2018-12-07 | A kind of nuclear power plant containment shell appearance inspecting system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109521021A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112750127A (en) * | 2021-02-04 | 2021-05-04 | 深圳市泽峰光电科技有限公司 | Image processing method for log end face measurement |
| CN113223739A (en) * | 2021-04-16 | 2021-08-06 | 中广核工程有限公司 | Nuclear power station containment test system under depth defense |
| CN113588666A (en) * | 2021-08-12 | 2021-11-02 | 同济大学 | Curtain wall defect detection, positioning and statistical method |
| CN116701894A (en) * | 2023-06-07 | 2023-09-05 | 深圳市胜美德电子有限公司 | Electronic product test control system based on multi-environment simulation |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103364408A (en) * | 2013-07-10 | 2013-10-23 | 三峡大学 | Method for detecting underwater surface crack of hydraulic concrete structure by using underwater robot system |
| CN104048969A (en) * | 2014-06-19 | 2014-09-17 | 樊晓东 | Tunnel defect recognition method |
| CN104076040A (en) * | 2014-07-02 | 2014-10-01 | 中广核工程有限公司 | Nuclear power plant concrete building defect detection method, equipment and system |
| CN205302962U (en) * | 2015-10-21 | 2016-06-08 | 国核华清（北京）核电技术研发中心有限公司 | A wall climbing robot that is used for nuclear power station steel containment wall to detect and system thereof |
| CN105811297A (en) * | 2016-03-03 | 2016-07-27 | 国网山东省电力公司茌平县供电公司 | Automatic insulator maintenance method for use in power transmission and distribution system |
| CN106918600A (en) * | 2017-04-07 | 2017-07-04 | 江苏博虏智能科技有限公司 | A kind of web surface defects detection and labeling method based on machine vision |
| CN207104930U (en) * | 2017-08-30 | 2018-03-16 | 海安科大机器人科技有限公司 | Wall spray robot is climbed based on mechanical arm |
| CN108918539A (en) * | 2018-07-27 | 2018-11-30 | 同济大学 | A kind of apparent Defect inspection device and method of tunnel structure |
-
2018
- 2018-12-07 CN CN201811493072.9A patent/CN109521021A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103364408A (en) * | 2013-07-10 | 2013-10-23 | 三峡大学 | Method for detecting underwater surface crack of hydraulic concrete structure by using underwater robot system |
| CN104048969A (en) * | 2014-06-19 | 2014-09-17 | 樊晓东 | Tunnel defect recognition method |
| CN104076040A (en) * | 2014-07-02 | 2014-10-01 | 中广核工程有限公司 | Nuclear power plant concrete building defect detection method, equipment and system |
| CN205302962U (en) * | 2015-10-21 | 2016-06-08 | 国核华清（北京）核电技术研发中心有限公司 | A wall climbing robot that is used for nuclear power station steel containment wall to detect and system thereof |
| CN105811297A (en) * | 2016-03-03 | 2016-07-27 | 国网山东省电力公司茌平县供电公司 | Automatic insulator maintenance method for use in power transmission and distribution system |
| CN106918600A (en) * | 2017-04-07 | 2017-07-04 | 江苏博虏智能科技有限公司 | A kind of web surface defects detection and labeling method based on machine vision |
| CN207104930U (en) * | 2017-08-30 | 2018-03-16 | 海安科大机器人科技有限公司 | Wall spray robot is climbed based on mechanical arm |
| CN108918539A (en) * | 2018-07-27 | 2018-11-30 | 同济大学 | A kind of apparent Defect inspection device and method of tunnel structure |
Non-Patent Citations (1)
| Title |
|---|
| 普尔菲勒迪斯: "《铁路管理与工程》", 31 May 2012, 中国铁道出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112750127A (en) * | 2021-02-04 | 2021-05-04 | 深圳市泽峰光电科技有限公司 | Image processing method for log end face measurement |
| CN113223739A (en) * | 2021-04-16 | 2021-08-06 | 中广核工程有限公司 | Nuclear power station containment test system under depth defense |
| CN113223739B (en) * | 2021-04-16 | 2023-08-22 | 中广核工程有限公司 | Nuclear power station containment test system under deep defense |
| CN113588666A (en) * | 2021-08-12 | 2021-11-02 | 同济大学 | Curtain wall defect detection, positioning and statistical method |
| CN116701894A (en) * | 2023-06-07 | 2023-09-05 | 深圳市胜美德电子有限公司 | Electronic product test control system based on multi-environment simulation |
| CN116701894B (en) * | 2023-06-07 | 2024-04-30 | 深圳市胜美德电子有限公司 | Electronic product test control system based on multi-environment simulation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109521021A (en) | A kind of nuclear power plant containment shell appearance inspecting system and method | |
| CN106548182B (en) | Pavement crack detection method and device based on deep learning and main cause analysis | |
| CN108022235B (en) | Method for identifying defects of key components of high-voltage transmission iron tower | |
| CN104535356B (en) | A kind of roller steel wire rope rope fault monitoring method and system based on machine vision | |
| CN112396658B (en) | Indoor personnel positioning method and system based on video | |
| CN109858367B (en) | Visual automatic detection method and system for worker through supporting unsafe behaviors | |
| CN112837290B (en) | Crack image automatic identification method based on seed filling algorithm | |
| CN113903081A (en) | Visual identification artificial intelligence alarm method and device for images of hydraulic power plant | |
| CN111275679A (en) | Solar cell defect detection system and method based on image | |
| CN109711348A (en) | Intelligent monitoring method and system based on the long-term real-time architecture against regulations in hollow panel | |
| CN112149512A (en) | Helmet wearing identification method based on two-stage deep learning | |
| CN109712127A (en) | A kind of electric transmission line fault detection method for patrolling video flowing for machine | |
| CN111127399A (en) | Underwater pier disease identification method based on deep learning and sonar imaging | |
| CN112528979B (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN112465706A (en) | Automatic gate container residual inspection method | |
| CN109949209A (en) | A kind of rope detection and minimizing technology based on deep learning | |
| CN112287823A (en) | Facial mask identification method based on video monitoring | |
| CN106600580A (en) | Hough transform-based abnormal recognition method and system of power line | |
| CN116152748A (en) | River and lake supervision method and system based on blue algae identification | |
| CN105023018A (en) | Jet code detection method and system | |
| CN111667473A (en) | Insulator hydrophobicity grade judging method based on improved Canny algorithm | |
| CN114943841A (en) | Method and device for assisting operation safety control based on image recognition | |
| CN114973320A (en) | Underground coal mine personnel detection method based on depth information | |
| CN113179389A (en) | System and method for identifying crane jib of power transmission line dangerous vehicle | |
| CN116309447B (en) | Dam slope crack detection method based on deep learning |
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 |