CN107680195A - A kind of transformer station intelligent robot inspection Computer Aided Analysis System and method - Google Patents
A kind of transformer station intelligent robot inspection Computer Aided Analysis System and method Download PDFInfo
- Publication number
- CN107680195A CN107680195A CN201711115053.8A CN201711115053A CN107680195A CN 107680195 A CN107680195 A CN 107680195A CN 201711115053 A CN201711115053 A CN 201711115053A CN 107680195 A CN107680195 A CN 107680195A
- Authority
- CN
- China
- Prior art keywords
- identification
- deep learning
- module
- intelligent robot
- transformer station
- 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
- 238000007689 inspection Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004458 analytical method Methods 0.000 title claims abstract description 25
- 238000013135 deep learning Methods 0.000 claims abstract description 73
- 230000005856 abnormality Effects 0.000 claims abstract description 24
- 230000008569 process Effects 0.000 claims abstract description 11
- 238000003909 pattern recognition Methods 0.000 claims abstract description 5
- 230000003936 working memory Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 8
- 239000000203 mixture Substances 0.000 claims description 6
- 230000007547 defect Effects 0.000 claims description 4
- 230000002708 enhancing effect Effects 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000007711 solidification Methods 0.000 claims description 3
- 230000008023 solidification Effects 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013434 data augmentation Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 229910052573 porcelain Inorganic materials 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02B—BOARDS, SUBSTATIONS OR SWITCHING ARRANGEMENTS FOR THE SUPPLY OR DISTRIBUTION OF ELECTRIC POWER
- H02B3/00—Apparatus specially adapted for the manufacture, assembly, or maintenance of boards or switchgear
Abstract
The invention discloses a kind of transformer station intelligent robot inspection Computer Aided Analysis System and method, including image taking servo module, deep learning identification module, abnormality reasoning module, wherein, image taking servo module is used for the sample image for assisting intelligent robot to shoot and inputs above-mentioned sample image for deep learning identification module；Deep learning identification module structure three-level schema identification system, equipment, part, state are identified respectively for three-level schema identification system, and every grade of pattern-recognition system carries out process identification to sample image and exports recognition result to image taking servo module and abnormality reasoning module；Abnormality reasoning module makes inferences engine and inference conclusion is showed into user according to the status information of equipment of deep learning identification module identification.The present invention has higher accuracy, stronger generalization ability compared to the image-recognizing method of traditional feature based.
Description
Technical field
The present invention relates to intelligent substation patrol technical field, more specifically to a kind of transformer station intelligent robot
Inspection Computer Aided Analysis System and method.
Background technology
Daily Round Check is to ensure the most important Working Means of substation safety stable operation, with the quick hair of social economy
Exhibition, power grid construction dynamics of investment continue to increase, supplement and the nonsynchronous contradiction of power grid construction development speed of technical skills personnel
Show increasingly.
Traditional substation inspection work is planned to daily by the timing node of formulation by substation field operations staff
Ask to the transformer station administered and do regularly periodic inspection, record relevant device service data.In electric power great development
Under background, the transformer station that same operation maintenance personnel is administered is more, corresponding primary equipment enormous amount, will be repeated daily
Property patrol task, not only easily produce be sick of psychology, can not ensure make an inspection tour quality.Meanwhile Personal Skills' water of operation maintenance personnel
There is gap between flat, operation maintenance personnel is easy to only rely on by means of personal experience, to judge, not ensure during equipment routing inspection
Unified standard.These bring very big challenge to substation field inspection work, and also the safe and stable operation of power network is caused
Great potential safety hazard.
With the development that intelligent robot technology applies, Intelligent Mobile Robot enters the practical stage, and robot can
To complete the tour work of outdoor primary equipment some projects in transformer station according to preplanned mission.It presets rail in transformer station's interior edge
Road and default inspection point provided carry out the tour work of specification with following the prescribed order, and carry out oil level reading, meter is made a copy of, temperature detection and outer
The work such as detection are seen, during tour, the view data for constantly being collected itself uploads to monitoring backstage for robot,
If it find that irregular operating state, monitoring backstage can realize alarm, remind operations staff to pay attention to, and can generate abnormal report
Accuse.The equipment of transformer station is maked an inspection tour using intelligent patrol robot, avoiding manual patrol subjective aspect may cause partially
Difference, all tour data of synchronous recording, reliable basis are provided for accident analysis, while it is possible during tour to avoid personnel
The danger run into.
But existing crusing robot is using fixed point definite angle shot image, the side learnt from the background using conventional machines
The method of formula identification of defective.The identification of conventional machines study failure needs priori design feature parameter, characteristic parameter selection
Quality will directly affect the effect of fault identification, and the image photographed due to the environment such as camera site, angle, illumination because
Element influences, and causes the accuracy fluctuation of fault identification very big, it is impossible to meet the requirement of practical application.
The content of the invention
The technical problem to be solved in the present invention is, there is provided a kind of transformer station intelligent robot inspection Computer Aided Analysis System and
Method.
The technical solution adopted for the present invention to solve the technical problems is：It is auxiliary to construct a kind of transformer station intelligent robot inspection
Analysis system, including image taking servo module, deep learning identification module, abnormality reasoning module are helped, wherein, the figure
As shooting servo module is used to assist the sample image of intelligent robot shooting and inputs above-mentioned sample for deep learning identification module
This image；Deep learning identification module structure three-level schema identification system, the three-level schema identification system pair are set respectively
Standby, part, state are identified, every grade of pattern-recognition system to sample image carry out process identification and by recognition result export to
Described image shoots servo module and abnormality reasoning module；The abnormality reasoning module is according to deep learning identification mould
The status information of equipment of block identification makes inferences engine and inference conclusion is showed into user.
In such scheme, the three-level schema identification system includes first order equipment identification layer, second level part identifies
Layer and third level state recognition layer, the first order equipment identification layer are used to identify transformer station's main equipment, the second level part
The part composition that identification layer is used to identify on main equipment, the third level state recognition layer are used for identification component with the presence or absence of different
Often.
In such scheme, described image shooting servo module determines intelligence machine using deep learning algorithm of target detection
The position of reference object in image captured by people, and focal length and shooting angle are adjusted by camera.
In such scheme, intelligent robot shooting sample image corresponding to reference object include transformer station's main equipment,
On main equipment part composition, component detail region, successively with the first order equipment identification layer, second level part identification layer and
Third level state recognition layer is corresponding.
In such scheme, the deep learning identification module uses two kinds of sides of offline deep learning and online deep learning
Formula carries out process identification to the power equipment feature of sample image, and offline deep learning mode builds deep learning platform, uses
Depth convolutional neural networks are trained to the target signature of representative power equipment in sample image, and online deep learning mode is adopted
Sample image is incorporated into deep learning framework with incremental learning and enhancing learning art, the model of offline deep learning mode is entered
Row optimization.
In such scheme, the abnormality reasoning module includes working memory, rule base and inference engine three parts,
Engine is made inferences using Drools rules, the working memory is used for the equipment state that storage depth study identification module obtains
Feature, the rule base are used for the expertise for storing solidification, and the inference engine draws including pattern matcher, agenda and execution
Three parts are held up, the inference engine is used for the equipment state feature that user is input in working memory and the rule in rule base
Matched and drawn a conclusion.
The present invention also provides a kind of transformer station intelligent robot method for inspecting, and the method for inspecting comprises the following steps：
Step S010, intelligent robot, which reaches, specifies inspection point, shoots the initial pictures of main equipment in transformer station, and pass through
Communication is sent to above-mentioned transformer station intelligent robot inspection Computer Aided Analysis System, image taking servo module profit in standing
The position of reference object in image captured by intelligent robot is determined with deep learning algorithm of target detection, and is adjusted by camera
Whole focal length and shooting angle are to photograph the centre position that the panorama of main equipment and main equipment in transformer station are located at image, intelligence
After robot adjusts shooting strategy, first sample image is shot, and send to intelligent substation machine by wireless transmission method
Device people's inspection Computer Aided Analysis System.
Step S020, the first sample image photographed are sent into deep learning identification module, deep learning identification module pair
Sample image carries out process identification, including all parts of transformer and its position, for each part, deep learning identification mould
Block sends corresponding shooting strategy to image taking servo module, specifies intelligent robot to carry out secondary shooting to part.
Step S030, intelligent robot receive shooting strategy, adjust camera, and secondary bat is carried out for specified parts
Take the photograph, and obtained secondary image is wirelessly deposited to transformer station intelligent robot inspection Computer Aided Analysis System；
Step S040, deep learning identification module carries out defect to secondary image, fault type recognizes, and identification is obtained
Equipment state export to abnormality reasoning module.
Step S050, abnormality reasoning module are pushed away according to the status information of equipment of deep learning identification module identification
Inference conclusion is simultaneously showed user by reason engine.
Implement a kind of transformer station intelligent robot inspection Computer Aided Analysis System of the present invention and method, have below beneficial to effect
Fruit：
1st, the present invention is realized to the semantic excavation of acquired image progress, and profit using artificial intelligence, depth learning technology
It is auxiliary with incremental learning, depth enhancing learning art, transformer station intelligent robot inspection of the structure with object automatic tracking technology
Analysis system is helped, realizes substation equipment failure, the INTELLIGENT IDENTIFICATION of defect, auxiliary operation maintenance personnel routine work.
2nd, the present invention is compared to the image-recognizing method of traditional feature based, has higher accuracy, stronger general
Change ability.
3rd, groundwork of the invention realizes that hardware components can coordinate with existing intelligent inspection robot in software,
It is significantly cost-effective.
Brief description of the drawings
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing：
Fig. 1 is a kind of applied topology structure chart of transformer station intelligent robot inspection Computer Aided Analysis System；
Fig. 2 is a kind of Whole Work Flow figure of transformer station intelligent robot inspection Computer Aided Analysis System；
Fig. 3 is the workflow diagram of deep learning identification module in the present invention；
Fig. 4 is the workflow diagram of abnormality reasoning module in the present invention.
Embodiment
In order to which technical characteristic, purpose and the effect of the present invention is more clearly understood, now compares accompanying drawing and describe in detail
The embodiment of the present invention.
As shown in Fig. 2 the present invention provides a kind of transformer station intelligent robot inspection Computer Aided Analysis System, including image taking
Servo module, deep learning identification module, abnormality reasoning module.Wherein, image taking servo module passes through deep learning
First, second grade of conclusion fed back of network, strategy is shot by the automatic tracking cell adjust automatically of object, instructs object to clap
Take the photograph unit and targetedly gather picture material, picture material is divided into any angle image, equipment according to the different levels of application
Three kinds of panoramic picture and fine features image；Deep learning identification module mainly applies the knowledge of deep learning real-time performance object
Not, positioning and state identification, deep learning network is divided into three-level schema identification, respectively for visual angle image, equipment panorama
Image and fine features image, for preceding two-stage mainly for the identification and positioning of target, its conclusion will instruct object from motion tracking list
Member, afterbody are state identification, identify the state of different parts；Abnormality reasoning module obtains deep learning identification module
The all parts state fusion arrived arranges, and forms status information of equipment set, by rule-based reasoning unit, derives whether equipment is located
User is presented to by web modes in abnormality, and by the reasoning results.
Image taking servo module, deep learning identification module, abnormality reasoning module are described in detail below：
Image taking servo module utilizes reference object automatic tracking technology, adjust automatically shooting strategy, assists intelligent machine
The sample image of device people shooting, and input above-mentioned sample image for deep learning identification module.Image taking servo module utilizes
The algorithm of target detection of deep learning end to end based on homing method determines reference object in image captured by intelligent robot
Position, by camera control API, adjust image taking strategy, including adjustment focal length, control shooting angle up and down,
To ensure the integrality of reference object and definition.
Deep learning identification module structure three-level schema identification system, three-level schema identify that system is known including first order equipment
Other layer, second level part identification layer and third level state recognition layer, first order equipment identification layer are used to identify transformer station's main equipment,
Such as transformer, reactor, breaker, disconnecting switch, the part composition that second level part identification layer is used to identify on main equipment,
Body, sleeve pipe, conservator, the radiator of such as transformer, third level state recognition layer are used for identification component and whether there is exception, such as
The filth of bushing shell for transformer porcelain screen, crack, breakage etc..Reference object corresponding to the sample image of intelligent robot shooting includes power transformation
Part composition in station owner's equipment, main equipment, component detail region, know successively with first order equipment identification layer, second level part
Other layer and third level state recognition layer are corresponding.Equipment, part, state are identified respectively for three-level schema identification system, every grade
Pattern-recognition system carries out process identification to sample image and exports recognition result to image taking servo module and abnormal shape
State reasoning module.In the present embodiment, deep learning identification module uses two kinds of sides of offline deep learning and online deep learning
Formula carries out process identification to the power equipment feature of sample image, and offline deep learning mode builds deep learning platform, uses
Depth convolutional neural networks are trained to the target signature of representative power equipment in sample image, and online deep learning mode is adopted
Sample image is incorporated into deep learning framework with incremental learning and enhancing learning art, the model of offline deep learning mode is entered
Row optimization.
As shown in figure 3, the mode of operation of the deep learning identification module of the present embodiment is divided into deep learning network struction ring
Save (offline deep learning) and the application link (online deep learning) of deep learning network, the use of deep learning identification module
It is divided into following steps, specifically includes：
Step 1), in the module before use, first initialization training is carried out to deep neural network, using offline depth
Spend the method for study, by the training set of images of preprepared tape label, using be not limited to data augmentation, image preprocessing,
The skills such as netinit, the selecting of activation primitive, different regularization methods, complete the initialization training of deep neural network.
Step 2), the deep neural network model trained is applied in ONLINE RECOGNITION, carries out identification, the positioning of target
And state identification, image of the input from actual photographed of model.
Step 3), after system applies a period of time, deep learning model needs to obtain newly from interactive process
Knowledge, constantly improve recognition accuracy, it is therefore desirable to form incremental learning sample using ONLINE RECOGNITION picture.Simultaneously as should
Needs, it may be necessary to identify the new kind equipment that "current" model can not identify, thus need preparation one to have been subjected to
The newly-increased image set of demarcation, collectively as incremental learning sample.
Step 4), using incremental learning sample, by Incremental Learning Algorithm, complete excellent to the parameter of deep learning network
Change.
Abnormality reasoning module is regular using Drools according to the status information of equipment of deep learning identification module identification
Make inferences engine and inference conclusion is showed into user.Abnormality reasoning module includes working memory, rule base and reasoning
Engine three parts, working memory are used for the equipment state feature that storage depth study identification module obtains, and rule base is used to store
The expertise of solidification, inference engine include pattern matcher, agenda and enforcement engine three parts, and inference engine is used for user
The equipment state feature being input in working memory is matched and drawn a conclusion with the rule in rule base.
As shown in figure 4, the inference step of the Drools rule-based reasoning engines in the present embodiment is as follows：
Step 1), status information of equipment set is stored in working memory, Land use models adaptation and the production in rule base
Raw formula rule is matched, and produces matched rule.Wherein, in the known integrated circuit it is a fact that one of relation between description object and between object properties
Group data；Production rule is the reasoning sentence being made up of condition and conclusion, is the binary that knowledge presentation is carried out with first order logic
Structure, shaped like when<conditions>then<actions>.
Step 2), matching process may cause multiple rules to be simultaneously activated, and the rule that inference engine conflicts these is united
One is put into conflict set.
Step 3), agenda utilize conflict decision strategy, the regular order of reasonable arrangement activation, while by the rule of activation
It is put into agenda.
Step 4), the rule in agenda, repeat step 2 are sequentially carried out by enforcement engine) step 4) is arrived, until strictly all rules
All it is performed and finishes, exports the conclusion after similar expert reasoning thereafter.
The present invention also provides a kind of transformer station intelligent robot method for inspecting, and the method for inspecting comprises the following steps：
Step S010, as shown in figure 1, intelligent robot in inspection operation, runs along the inspection channel delimited, arrived
Stop up to pre-set inspection point, camera is adjusted to predetermined angle, shoots the initial graph of main equipment in transformer station
Picture, and it is sent to by communication in station background system (the i.e. above-mentioned transformer station intelligent robot inspection in master control building
Computer Aided Analysis System), image taking servo module determines figure captured by intelligent robot using deep learning algorithm of target detection
The position of reference object as in, and focal length and shooting angle adjusted by camera to photograph the complete of main equipment in transformer station
Scape and main equipment are located at the centre position of image, after intelligent robot adjusts shooting strategy, shoot first sample image, and lead to
Wireless transmission method is crossed to deposit to background system.
Step S020, the main equipment image photographed are sent into deep learning identification module, and deep learning identification module is to sample
This image carries out object (including all parts of transformer and its position, part include body, sleeve pipe, conservator and radiator etc.)
Identification.For each part, deep learning identification module specifies corresponding shooting strategy to send to image taking servo module, refers to
Determine intelligent robot and secondary shooting is carried out to part.
Step S030, intelligent robot receive shooting strategy, adjust camera, and secondary bat is carried out for specified parts
Take the photograph, and obtained secondary image is wirelessly deposited to background system；
Step S040, deep learning identification module carries out defect to secondary image, fault type recognizes, and identification is obtained
Equipment state export to abnormality reasoning module.
Step S050, abnormality reasoning module are pushed away according to the status information of equipment of deep learning identification module identification
Inference conclusion is simultaneously showed user by reason engine.
Embodiments of the invention are described above in conjunction with accompanying drawing, but the invention is not limited in above-mentioned specific
Embodiment, above-mentioned embodiment is only schematical, rather than restricted, one of ordinary skill in the art
Under the enlightenment of the present invention, in the case of present inventive concept and scope of the claimed protection is not departed from, it can also make a lot
Form, these are belonged within the protection of the present invention.
Claims (7)
1. a kind of transformer station intelligent robot inspection Computer Aided Analysis System, it is characterised in that including image taking servo module, depth
Degree study identification module, abnormality reasoning module, wherein, described image shooting servo module is used to assist intelligent robot to clap
The sample image taken the photograph simultaneously inputs above-mentioned sample image for deep learning identification module；The deep learning identification module builds three-level
Equipment, part, state are identified respectively for pattern-recognition system, the three-level schema identification system, every grade of pattern-recognition body
System carries out process identification to sample image and exports recognition result to described image to shoot servo module and abnormality reasoning
Module；The abnormality reasoning module makes inferences engine simultaneously according to the status information of equipment of deep learning identification module identification
Inference conclusion is showed into user.
2. a kind of transformer station intelligent robot inspection Computer Aided Analysis System according to claim 1, it is characterised in that described
Three-level schema identification system includes first order equipment identification layer, second level part identification layer and third level state recognition layer, described
First order equipment identification layer is used to identify transformer station's main equipment, and the second level part identification layer is used to identify the portion on main equipment
Part forms, and the third level state recognition layer is used for identification component with the presence or absence of abnormal.
3. a kind of transformer station intelligent robot inspection Computer Aided Analysis System according to claim 2, it is characterised in that described
Image taking servo module determines reference object in image captured by intelligent robot using deep learning algorithm of target detection
Position, and focal length and shooting angle are adjusted by camera.
A kind of 4. transformer station intelligent robot inspection Computer Aided Analysis System according to claim 3, it is characterised in that intelligence
It is thin that reference object corresponding to the sample image of robot shooting includes transformer station's main equipment, the part composition on main equipment, part
Region is saved, it is corresponding with the first order equipment identification layer, second level part identification layer and third level state recognition layer successively.
5. a kind of transformer station intelligent robot inspection Computer Aided Analysis System according to claim 2, it is characterised in that described
Deep learning identification module is special to the power equipment of sample image using offline deep learning and online deep learning two ways
Sign carries out process identification, and offline deep learning mode builds deep learning platform, using depth convolutional neural networks to sample graph
The target signature of representative power equipment is trained as in, and online deep learning mode is using incremental learning and enhancing learning art
Sample image is incorporated into deep learning framework, the model of offline deep learning mode is optimized.
6. a kind of transformer station intelligent robot inspection Computer Aided Analysis System according to claim 1, it is characterised in that described
Abnormality reasoning module includes working memory, rule base and inference engine three parts, is made inferences and drawn using Drools rules
Hold up, the working memory is used for the equipment state feature that storage depth study identification module obtains, and the rule base is used to store
The expertise of solidification, the inference engine include pattern matcher, agenda and enforcement engine three parts, and the inference engine is used
Matched and drawn a conclusion with the rule in rule base in the equipment state feature that user is input in working memory.
7. a kind of transformer station intelligent robot method for inspecting, it is characterised in that the method for inspecting comprises the following steps：
Step S010, intelligent robot, which reaches, specifies inspection point, shoots the initial pictures of main equipment in transformer station, and by station
Communication is sent to any one of the claim 1-6 transformer station intelligent robot inspection Computer Aided Analysis Systems, described
Image taking servo module determines reference object in image captured by intelligent robot using deep learning algorithm of target detection
Position, and focal length and shooting angle adjusted by camera to photograph the panorama of main equipment in transformer station and main equipment is located at
The centre position of image, after intelligent robot adjusts shooting strategy, first sample image is shot, and pass through wireless transmission method
Send to transformer station intelligent robot inspection Computer Aided Analysis System；
Step S020, the first sample image photographed are sent into the deep learning identification module, and the deep learning identifies mould
Block carries out process identification, including all parts of transformer and its position, for each part, the depth to sample image
Practise identification module to send corresponding shooting strategy to image taking servo module, specify intelligent robot to carry out part secondary
Shooting；
Step S030, intelligent robot receive shooting strategy, adjust camera, and secondary shooting is carried out for specified parts, and
Obtained secondary image is wirelessly deposited to the transformer station intelligent robot inspection Computer Aided Analysis System；
Step S040, the deep learning identification module carries out defect to secondary image, fault type recognizes, and identification is obtained
Equipment state export to the abnormality reasoning module.
Step S050, the abnormality reasoning module are pushed away according to the status information of equipment of deep learning identification module identification
Inference conclusion is simultaneously showed user by reason engine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711115053.8A CN107680195A (en) | 2017-11-13 | 2017-11-13 | A kind of transformer station intelligent robot inspection Computer Aided Analysis System and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711115053.8A CN107680195A (en) | 2017-11-13 | 2017-11-13 | A kind of transformer station intelligent robot inspection Computer Aided Analysis System and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107680195A true CN107680195A (en) | 2018-02-09 |
Family
ID=61146989
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711115053.8A Pending CN107680195A (en) | 2017-11-13 | 2017-11-13 | A kind of transformer station intelligent robot inspection Computer Aided Analysis System and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107680195A (en) |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537792A (en) * | 2018-04-17 | 2018-09-14 | 成都思晗科技股份有限公司 | A kind of electric power defect image recognition methods based on convolutional neural networks |
| CN108597052A (en) * | 2018-04-24 | 2018-09-28 | 合肥市艾特米信息技术有限公司 | Instrument automatic inspection device |
| CN108847968A (en) * | 2018-06-04 | 2018-11-20 | 国网浙江省电力有限公司 | Monitoring accident, anomalous event identification and multidimensional analysis method |
| CN109005391A (en) * | 2018-09-14 | 2018-12-14 | 四川能投云电科技有限公司 | A kind of camera moving method of detection failure |
| CN109102669A (en) * | 2018-09-06 | 2018-12-28 | 广东电网有限责任公司 | A kind of transformer substation auxiliary facility detection control method and its device |
| CN109117789A (en) * | 2018-08-10 | 2019-01-01 | 国网上海市电力公司 | A kind of switchgear inspection image segmentation configuration processing method |
| CN109615086A (en) * | 2018-10-11 | 2019-04-12 | 国网浙江省电力有限公司电力科学研究院 | A kind of generation method and system of O&M assisted tag |
| CN109725233A (en) * | 2018-12-12 | 2019-05-07 | 杭州申昊科技股份有限公司 | A kind of intelligent substation inspection system and its method for inspecting |
| CN110142785A (en) * | 2019-06-25 | 2019-08-20 | 山东沐点智能科技有限公司 | A kind of crusing robot visual servo method based on target detection |
| CN110246175A (en) * | 2019-05-24 | 2019-09-17 | 国网安徽省电力有限公司检修分公司 | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera |
| CN110266938A (en) * | 2018-12-27 | 2019-09-20 | 全球能源互联网研究院有限公司 | Substation equipment intelligence image pickup method and device based on deep learning |
| CN110351527A (en) * | 2019-07-25 | 2019-10-18 | 友谊国际工程咨询有限公司 | The other robot system of standing of engineering construction intelligence |
| CN110497331A (en) * | 2018-05-16 | 2019-11-26 | 李迪 | Hidden danger is assisted to identify crusing robot |
| CN110610136A (en) * | 2019-08-21 | 2019-12-24 | 广西电网有限责任公司电力科学研究院 | Transformer substation equipment identification module and identification method based on deep learning |
| CN111046943A (en) * | 2019-12-09 | 2020-04-21 | 国网智能科技股份有限公司 | Method and system for automatically identifying state of isolation switch of transformer substation |
| CN111123916A (en) * | 2019-12-05 | 2020-05-08 | 国网浙江省电力有限公司温州供电公司 | Method and device for power inspection robot |
| CN111157531A (en) * | 2019-12-25 | 2020-05-15 | 南京悠阔电气科技有限公司 | Transformer substation equipment defect identification method based on deep learning |
| CN111239768A (en) * | 2020-01-13 | 2020-06-05 | 南京七宝机器人技术有限公司 | Method for automatically constructing map and searching inspection target by electric power inspection robot |
| EP3706268A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706267A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706266A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706270A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706269A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| CN111682440A (en) * | 2020-06-17 | 2020-09-18 | 贵州电网有限责任公司 | Digital substation inspection method |
| CN111917178A (en) * | 2020-06-23 | 2020-11-10 | 国网浙江乐清市供电有限公司 | Patrol method of operation and maintenance integrated transformer substation |
| CN112004012A (en) * | 2020-08-26 | 2020-11-27 | 西安咏圣达电子科技有限公司 | Switch cabinet running state management and detection system and method and computer equipment |
| CN112383698A (en) * | 2020-10-09 | 2021-02-19 | 西安星闪数创智能科技有限公司 | Transformer platform district terminal intelligent monitoring system |
| CN112526254A (en) * | 2020-11-10 | 2021-03-19 | 国网内蒙古东部电力有限公司 | High-precision positioning inspection monitoring method capable of resisting strong electromagnetic interference |
| CN113131378A (en) * | 2021-04-13 | 2021-07-16 | 国网黑龙江省电力有限公司鹤岗供电公司 | Intelligent inspection system and method for secondary system of transformer substation based on image recognition |
| CN113205177A (en) * | 2021-04-25 | 2021-08-03 | 广西大学 | Electric power terminal identification method based on incremental collaborative attention mobile convolution |
| CN114665608A (en) * | 2022-05-19 | 2022-06-24 | 国网智能科技股份有限公司 | Intelligent sensing inspection system and method for transformer substation |
| CN115150559A (en) * | 2022-09-06 | 2022-10-04 | 国网天津市电力公司高压分公司 | Remote vision system with acquisition self-adjustment calculation compensation and calculation compensation method |
| CN115223384A (en) * | 2021-03-29 | 2022-10-21 | 东风汽车集团股份有限公司 | Vehicle data display method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722166A (en) * | 2012-06-29 | 2012-10-10 | 山东电力集团公司电力科学研究院 | Intelligent vision detection system and state detection method of transformer substation device |
| CN106228246A (en) * | 2016-07-15 | 2016-12-14 | 江苏科技大学 | Based on semantic unattended duty transformer substation monitoring system and method |
| CN106597231A (en) * | 2016-11-11 | 2017-04-26 | 上海交通大学 | GIS fault detection system and method based on multi-source information fusion and deep learning network |
| CN106649915A (en) * | 2016-09-08 | 2017-05-10 | 国家电网公司 | Early warning system for defects of transmission and distribution power grid equipment |
-
2017
- 2017-11-13 CN CN201711115053.8A patent/CN107680195A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722166A (en) * | 2012-06-29 | 2012-10-10 | 山东电力集团公司电力科学研究院 | Intelligent vision detection system and state detection method of transformer substation device |
| CN106228246A (en) * | 2016-07-15 | 2016-12-14 | 江苏科技大学 | Based on semantic unattended duty transformer substation monitoring system and method |
| CN106649915A (en) * | 2016-09-08 | 2017-05-10 | 国家电网公司 | Early warning system for defects of transmission and distribution power grid equipment |
| CN106597231A (en) * | 2016-11-11 | 2017-04-26 | 上海交通大学 | GIS fault detection system and method based on multi-source information fusion and deep learning network |
Cited By (48)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537792A (en) * | 2018-04-17 | 2018-09-14 | 成都思晗科技股份有限公司 | A kind of electric power defect image recognition methods based on convolutional neural networks |
| CN108597052A (en) * | 2018-04-24 | 2018-09-28 | 合肥市艾特米信息技术有限公司 | Instrument automatic inspection device |
| CN110497331A (en) * | 2018-05-16 | 2019-11-26 | 李迪 | Hidden danger is assisted to identify crusing robot |
| CN108847968B (en) * | 2018-06-04 | 2021-06-01 | 国网浙江省电力有限公司 | Monitoring accident and abnormal event identification and multidimensional analysis method |
| CN108847968A (en) * | 2018-06-04 | 2018-11-20 | 国网浙江省电力有限公司 | Monitoring accident, anomalous event identification and multidimensional analysis method |
| CN109117789A (en) * | 2018-08-10 | 2019-01-01 | 国网上海市电力公司 | A kind of switchgear inspection image segmentation configuration processing method |
| CN109102669A (en) * | 2018-09-06 | 2018-12-28 | 广东电网有限责任公司 | A kind of transformer substation auxiliary facility detection control method and its device |
| CN109005391A (en) * | 2018-09-14 | 2018-12-14 | 四川能投云电科技有限公司 | A kind of camera moving method of detection failure |
| CN109615086A (en) * | 2018-10-11 | 2019-04-12 | 国网浙江省电力有限公司电力科学研究院 | A kind of generation method and system of O&M assisted tag |
| CN109725233A (en) * | 2018-12-12 | 2019-05-07 | 杭州申昊科技股份有限公司 | A kind of intelligent substation inspection system and its method for inspecting |
| CN109725233B (en) * | 2018-12-12 | 2020-12-08 | 杭州申昊科技股份有限公司 | Intelligent substation inspection system and inspection method thereof |
| CN110266938B (en) * | 2018-12-27 | 2021-11-09 | 全球能源互联网研究院有限公司 | Transformer substation equipment intelligent shooting method and device based on deep learning |
| CN110266938A (en) * | 2018-12-27 | 2019-09-20 | 全球能源互联网研究院有限公司 | Substation equipment intelligence image pickup method and device based on deep learning |
| WO2020178237A1 (en) * | 2019-03-07 | 2020-09-10 | Abb Schweiz Ag | Apparatus for monitoring a switchgear |
| WO2020178253A1 (en) * | 2019-03-07 | 2020-09-10 | Abb Schweiz Ag | Apparatus for monitoring a switchgear |
| CN113544922A (en) * | 2019-03-07 | 2021-10-22 | Abb瑞士股份有限公司 | Device for monitoring a switchgear |
| CN113544923A (en) * | 2019-03-07 | 2021-10-22 | Abb瑞士股份有限公司 | Device for monitoring a switchgear |
| CN113508503A (en) * | 2019-03-07 | 2021-10-15 | Abb瑞士股份有限公司 | Device for monitoring a switchgear |
| EP3706268A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706267A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706266A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706270A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| EP3706269A1 (en) * | 2019-03-07 | 2020-09-09 | ABB Schweiz AG | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear |
| WO2020178255A1 (en) * | 2019-03-07 | 2020-09-10 | Abb Schweiz Ag | Apparatus for monitoring a switchgear |
| CN110246175A (en) * | 2019-05-24 | 2019-09-17 | 国网安徽省电力有限公司检修分公司 | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera |
| CN110142785A (en) * | 2019-06-25 | 2019-08-20 | 山东沐点智能科技有限公司 | A kind of crusing robot visual servo method based on target detection |
| CN110351527A (en) * | 2019-07-25 | 2019-10-18 | 友谊国际工程咨询有限公司 | The other robot system of standing of engineering construction intelligence |
| CN110610136A (en) * | 2019-08-21 | 2019-12-24 | 广西电网有限责任公司电力科学研究院 | Transformer substation equipment identification module and identification method based on deep learning |
| CN110610136B (en) * | 2019-08-21 | 2023-07-18 | 广西电网有限责任公司电力科学研究院 | Substation equipment identification module and method based on deep learning |
| CN111123916A (en) * | 2019-12-05 | 2020-05-08 | 国网浙江省电力有限公司温州供电公司 | Method and device for power inspection robot |
| CN111046943A (en) * | 2019-12-09 | 2020-04-21 | 国网智能科技股份有限公司 | Method and system for automatically identifying state of isolation switch of transformer substation |
| CN111157531A (en) * | 2019-12-25 | 2020-05-15 | 南京悠阔电气科技有限公司 | Transformer substation equipment defect identification method based on deep learning |
| CN111239768A (en) * | 2020-01-13 | 2020-06-05 | 南京七宝机器人技术有限公司 | Method for automatically constructing map and searching inspection target by electric power inspection robot |
| CN111682440A (en) * | 2020-06-17 | 2020-09-18 | 贵州电网有限责任公司 | Digital substation inspection method |
| CN111682440B (en) * | 2020-06-17 | 2022-07-01 | 贵州电网有限责任公司 | Digital substation inspection method |
| CN111917178A (en) * | 2020-06-23 | 2020-11-10 | 国网浙江乐清市供电有限公司 | Patrol method of operation and maintenance integrated transformer substation |
| CN111917178B (en) * | 2020-06-23 | 2023-07-14 | 国网浙江乐清市供电有限公司 | Inspection method of operation and maintenance integrated transformer substation |
| CN112004012A (en) * | 2020-08-26 | 2020-11-27 | 西安咏圣达电子科技有限公司 | Switch cabinet running state management and detection system and method and computer equipment |
| CN112383698A (en) * | 2020-10-09 | 2021-02-19 | 西安星闪数创智能科技有限公司 | Transformer platform district terminal intelligent monitoring system |
| CN112383698B (en) * | 2020-10-09 | 2022-05-06 | 西安星闪世图科技有限公司 | Transformer platform district terminal intelligent monitoring system |
| CN112526254A (en) * | 2020-11-10 | 2021-03-19 | 国网内蒙古东部电力有限公司 | High-precision positioning inspection monitoring method capable of resisting strong electromagnetic interference |
| CN115223384A (en) * | 2021-03-29 | 2022-10-21 | 东风汽车集团股份有限公司 | Vehicle data display method and device, electronic equipment and storage medium |
| CN115223384B (en) * | 2021-03-29 | 2024-01-16 | 东风汽车集团股份有限公司 | Vehicle data display method and device, electronic equipment and storage medium |
| CN113131378A (en) * | 2021-04-13 | 2021-07-16 | 国网黑龙江省电力有限公司鹤岗供电公司 | Intelligent inspection system and method for secondary system of transformer substation based on image recognition |
| CN113205177B (en) * | 2021-04-25 | 2022-03-25 | 广西大学 | Electric power terminal identification method based on incremental collaborative attention mobile convolution |
| CN113205177A (en) * | 2021-04-25 | 2021-08-03 | 广西大学 | Electric power terminal identification method based on incremental collaborative attention mobile convolution |
| CN114665608A (en) * | 2022-05-19 | 2022-06-24 | 国网智能科技股份有限公司 | Intelligent sensing inspection system and method for transformer substation |
| CN115150559A (en) * | 2022-09-06 | 2022-10-04 | 国网天津市电力公司高压分公司 | Remote vision system with acquisition self-adjustment calculation compensation and calculation compensation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107680195A (en) | A kind of transformer station intelligent robot inspection Computer Aided Analysis System and method | |
| CN111307823B (en) | System and method for detecting typical visual defects of power transformation equipment based on edge cloud cooperation | |
| CN108734117A (en) | Cable machinery external corrosion failure evaluation method based on YOLO | |
| CN105426908B (en) | A kind of substation's attributive classification method based on convolutional neural networks | |
| CN110427882A (en) | For the intelligent analysis method of tour, device, equipment and its storage medium | |
| CN113746663B (en) | Performance degradation fault root cause positioning method combining mechanism data and dual drives | |
| WO2018114166A1 (en) | Method, control device, and system for determining state values for describing operating states in a partial network of an energy supply network | |
| CN105958644A (en) | Pressing plate information management system for intelligent transformer station | |
| CN104218679A (en) | Remote automatic transformer substation polling method based on remote viewing system | |
| US20230290122A1 (en) | Unmanned aerial vehicle based system to track solar panel system construction and commissioning | |
| CN113205039A (en) | Power equipment fault image identification and disaster investigation system and method based on multiple DCNNs | |
| CN106339722A (en) | Line knife switch state monitoring method and device | |
| CN114326648B (en) | Remote cooperative electrolysis control method and system | |
| CN105005814A (en) | Online comprehensive service system for electric automobiles | |
| CN109002633A (en) | Device network modeling method based on separate space | |
| US20230045188A1 (en) | Meter recognition apparatus, meter monitoring system, and monitoring method therefor | |
| CN112873218A (en) | Transformer substation inspection robot and inspection method | |
| CN108120475A (en) | The data monitoring method and device of a kind of piping lane | |
| CN116433645A (en) | Belt bulge detection method and detection system | |
| CN113284103B (en) | Substation equipment defect online detection method based on space transformation fast R-CNN model | |
| CN114445411A (en) | Unmanned aerial vehicle line patrol defect identification system and control method | |
| CN115171214A (en) | Construction site abnormal behavior detection method and system based on FCOS target detection | |
| Jiang et al. | Application of power ubiquitous Internet of Things technology in intelligent inspection of unattended substation | |
| CN114021906A (en) | Unattended wind power generation operation and maintenance management method and system | |
| CN113920610A (en) | AR technology-based power equipment inspection method and system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180209 |
|
| RJ01 | Rejection of invention patent application after publication |