CN109785289B - Transmission line defect detection method and system and electronic equipment - Google Patents
Transmission line defect detection method and system and electronic equipment Download PDFInfo
- Publication number
- CN109785289B CN109785289B CN201811547406.6A CN201811547406A CN109785289B CN 109785289 B CN109785289 B CN 109785289B CN 201811547406 A CN201811547406 A CN 201811547406A CN 109785289 B CN109785289 B CN 109785289B
- Authority
- CN
- China
- Prior art keywords
- model
- data
- transmission line
- power transmission
- defect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000007547 defect Effects 0.000 title claims abstract description 207
- 230000005540 biological transmission Effects 0.000 title claims abstract description 166
- 238000001514 detection method Methods 0.000 title claims abstract description 131
- 238000000034 method Methods 0.000 claims abstract description 65
- 238000005457 optimization Methods 0.000 claims abstract description 44
- 238000013135 deep learning Methods 0.000 claims abstract description 42
- 238000013526 transfer learning Methods 0.000 claims abstract description 42
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 32
- 230000010354 integration Effects 0.000 claims abstract description 27
- 230000004927 fusion Effects 0.000 claims abstract description 22
- 238000013136 deep learning model Methods 0.000 claims abstract description 18
- 238000002372 labelling Methods 0.000 claims abstract description 18
- 230000005856 abnormality Effects 0.000 claims abstract description 8
- 238000012502 risk assessment Methods 0.000 claims description 59
- 230000002787 reinforcement Effects 0.000 claims description 43
- 238000009826 distribution Methods 0.000 claims description 23
- 238000010219 correlation analysis Methods 0.000 claims description 17
- 238000004458 analytical method Methods 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 15
- 230000015654 memory Effects 0.000 claims description 15
- 230000000694 effects Effects 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 13
- 230000002860 competitive effect Effects 0.000 claims description 12
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 230000007613 environmental effect Effects 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 230000001133 acceleration Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000013138 pruning Methods 0.000 claims description 6
- 238000011002 quantification Methods 0.000 claims description 6
- 238000007418 data mining Methods 0.000 claims description 5
- 230000006872 improvement Effects 0.000 claims description 5
- 238000012423 maintenance Methods 0.000 claims description 5
- 238000005065 mining Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims description 3
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 claims description 2
- YTAHJIFKAKIKAV-XNMGPUDCSA-N [(1R)-3-morpholin-4-yl-1-phenylpropyl] N-[(3S)-2-oxo-5-phenyl-1,3-dihydro-1,4-benzodiazepin-3-yl]carbamate Chemical compound O=C1[C@H](N=C(C2=C(N1)C=CC=C2)C1=CC=CC=C1)NC(O[C@H](CCN1CCOCC1)C1=CC=CC=C1)=O YTAHJIFKAKIKAV-XNMGPUDCSA-N 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims 4
- 238000003062 neural network model Methods 0.000 claims 1
- 238000007689 inspection Methods 0.000 abstract description 12
- 238000012545 processing Methods 0.000 description 6
- 238000003745 diagnosis Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000009977 dual effect Effects 0.000 description 4
- 238000013508 migration Methods 0.000 description 4
- 230000005012 migration Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 239000012212 insulator Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Abstract
The application relates to a method and a system for detecting defects of a power transmission line and electronic equipment. The method comprises the following steps: the method comprises the following steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprising the following steps: step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data; step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning; step c: and training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the power transmission line element on the basis of the target detection of the deep learning. According to the method and the device, deep learning distributed training is carried out on massive image samples, and a power transmission line defect detection model based on visible light, infrared, ultraviolet and other multi-image fusion is established, so that the defect identification accuracy can be improved, and the inspection work efficiency and quality are improved.
Description
Technical Field
The application belongs to the technical field of smart power grids, and particularly relates to a method and a system for detecting defects of a power transmission line and electronic equipment.
Background
At present, in the face of rapid development of industrialization and commercialization, electric power plays an increasingly important role in our life and work, and ensuring smooth operation of electric power equipment without power failure becomes a key guarantee. The transmission line is an important component of the power system and is also one of the locations where faults are likely to occur in the power grid. However, power lines often experience line failures due to weathering and thermal effects, which affect the proper operation of the power lines and even the entire power grid. Therefore, the transmission line is detected in real time, the power line fault is prevented in time, and the stable operation of the whole power system is directly related. The power line has a special structure and a large distribution range, and the geographical environment is extremely complex and changeable and is in an environment with wind, sunshine or extremely high humidity for a long time; because of the requirement of long-distance transmission, the transmission line is in a high-voltage state for a long time, and high current and voltage can cause abnormal heating of a conductor, so that accidents occur.
The traditional power transmission line inspection is mainly to observe, inspect and measure each part of the power transmission line by using other tools and instruments such as eyes or telescopes of workers, grasp the running state of the line, discover equipment defects in time and threaten the safety of the line. Because there are more limitations in the manual work patrols and examines the transmission line, unmanned aerial vehicle has great advantage in patrolling and examining, but at present mainly through the artifical interpretation mode in later stage to obtaining image processing, work load is heavy, to the manual work requirement higher, the precision remains to be improved, can not carry out the defect analysis early warning to the image in real time, actual effect is not ideal yet.
The national standard DLT 1482-2015 technical guideline for unmanned aerial vehicle inspection operation for overhead transmission lines, issued by the national energy agency, specifies an inspection system for inspecting the overhead transmission lines by using unmanned aerial vehicles, inspection operation requirements, preparation before inspection, inspection modes and methods, inspection contents, inspection data arrangement and transfer, abnormal condition handling and the like. And a power transmission line machine inspection defect specification is also established, and the description library comprises fields such as defect elements, defect types, defect phases, defect descriptions, defect representations, defect grades and the like, wherein the defect elements comprise ground wires, towers, hardware fittings, insulator string hardware fitting assembly strings, tower foundations, grounding devices, wire pulling systems, accessory facilities, lightning protection facilities and the like.
Target detection is a classic topic in the field of image processing and computer vision, and has wide application in the aspects of traffic monitoring, image retrieval, human-computer interaction and the like. In the traditional target detection algorithm, feature extraction and classification decision are carried out separately, the requirement on feature selection is strict, and an ideal effect is difficult to obtain when a complex scene is faced. Since deep learning has become a research hotspot, more and more deep learning-based models are proposed, which are not identical, but mostly adopt convolutional neural networks to deal with the target detection problem. Compared with the traditional target detection algorithm, the feature extraction and the mode classification in the convolutional neural network are carried out in parallel, and the complex scene can be better processed along with the increase of the number of layers. Reinforcement learning is widely and deeply applied in the field of artificial intelligence and becomes a key machine learning method which breaks through human-like intelligence at present. Deep learning mainly analyzes environmental information and extracts features from the environmental information; the reinforcement learning will further analyze the environmental characteristics based on these characteristics and select the corresponding action to achieve the target return. Deep reinforcement learning is a learning method which combines deep learning and reinforcement learning to realize one-to-one correspondence from perception to action. For the complex decision problem of power grid disaster evaluation, an artificial intelligence method is considered to be introduced into decision control, effective information is extracted from a power grid operation environment, and a decision evaluation mode is determined by combining environment information and a power grid operation state, so that the method has important significance and feasibility.
Disclosure of Invention
The application provides a method, a system and an electronic device for detecting defects of a power transmission line, and aims to solve at least one of the technical problems in the prior art to a certain extent.
In order to solve the above problems, the present application provides the following technical solutions:
a method for detecting defects of a power transmission line is characterized by constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprises the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: and training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the power transmission line element on the basis of the target detection of the deep learning.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the image sample data of the power transmission line element comprises visible light, infrared and ultraviolet sample data, and further comprises a normal labeling image and a defect labeling image.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in the step c, the deep learning model includes: and respectively constructing a defect target detection model, a multi-model ensemble learning model and a fusion detection model aiming at the visible light, infrared and ultraviolet image sample data, completing model training and parameter adjustment optimization, outputting a power transmission line defect detection model, and identifying the defects of the power transmission line elements.
The technical scheme adopted by the embodiment of the application further comprises the following steps: constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, and performing strategy learning of disaster information, disaster multi-source information fusion and analysis and risk assessment disaster early warning; the method for constructing the risk assessment early warning model based on the deep reinforcement learning and the space-time weak correlation analysis specifically comprises the following steps:
step d: constructing a power transmission line fault space-time weak correlation analysis model;
step e: constructing a risk assessment and fault prediction model of the hidden danger of the line defect;
step f: and establishing a reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm.
The technical scheme adopted by the embodiment of the application further comprises the following steps: constructing a defect detection model distribution deployment system and a model optimization system based on deep learning according to the power transmission line defect detection model and the risk assessment early warning model, and applying model mobile deployment, defect identification and risk evaluation; the method specifically comprises the following steps:
step g: deploying a mobile terminal of the terminal based on a Caffe2 framework, wherein the deployment comprises weight pruning, weight quantification, model storage, forward acceleration calculation improvement and application, equipment identification and a power transmission line defect detection model are deployed at the mobile terminal;
step h: performing online defect detection on the mobile terminal by using a trained and tested defect detection model, and identifying and marking a defect element in a detection image;
step i: constructing a multi-source heterogeneous mass big data platform for defects and risks of the power transmission line, and acquiring and accessing mass data;
step j: interface specifications are formulated, real-time acquisition, access verification and reinforcement learning model optimization adjustment of various streaming data are realized at a server side, distribution and deployment are realized through the interface specifications aiming at an optimized model, and a terminal receives the data to update the model.
Another technical scheme adopted by the embodiment of the application is as follows: a transmission line defect detection system, comprising:
a defect detection model construction module: the method comprises the steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements; the defect detection model building module comprises:
virtual and real integrated units: the virtual sample generation and labeling model is used for constructing a virtual-real integrated model, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
a transfer learning unit: the method is used for constructing a deep learning transfer learning model and finishing model optimization based on transfer learning;
a model training unit: the method is used for training a deep learning model based on a target detection algorithm and diagnosing the abnormity of the power transmission line element on the basis of the target detection of the deep learning.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the image sample data of the power transmission line element comprises visible light, infrared and ultraviolet sample data, and further comprises a normal labeling image and a defect labeling image.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the deep learning model comprises: and respectively constructing a defect target detection model, a multi-model ensemble learning model and a fusion detection model aiming at the visible light, infrared and ultraviolet image sample data, completing model training and parameter adjustment optimization, outputting a power transmission line defect detection model, and identifying the defects of the power transmission line elements.
The technical scheme adopted by the embodiment of the application further comprises a risk assessment early warning model construction module: the risk assessment early warning model is used for constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, and is used for carrying out strategy learning of disaster information, disaster multi-source information fusion and analysis and risk assessment disaster early warning; the risk assessment early warning model building module specifically comprises:
a first model construction unit: the method is used for constructing a power transmission line fault space-time weak correlation analysis model;
a second model construction unit: the method is used for constructing a risk assessment and fault prediction model of the hidden danger of the line defect;
a third model construction unit: the method is used for establishing the reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm.
The technical scheme adopted by the embodiment of the application further comprises a deployment and application module: the system comprises a fault detection model distribution and deployment system and a model optimization system, wherein the fault detection model distribution and deployment system and the model optimization system are used for constructing a fault detection model distribution and deployment system and a model optimization system based on deep learning according to the power transmission line fault detection model and the risk assessment early warning model and are used for model mobile deployment, fault identification and risk evaluation application; the deployment and application module specifically includes:
a model deployment unit: the method is used for deploying the mobile terminal of the terminal based on the Caffe2 framework, and comprises weight pruning, weight quantification, model storage, forward acceleration calculation improvement and application, equipment identification and a power transmission line defect detection model are deployed at the mobile terminal;
a defect identification unit: the defect detection module is used for carrying out online defect detection on the mobile terminal by using the trained and tested defect detection model, and identifying and marking the defect elements in the detected image;
a risk evaluation unit: the multi-source heterogeneous mass big data platform is used for constructing defects and risks of the power transmission line and is used for acquiring and accessing mass data;
an optimization updating unit: the method is used for formulating interface specifications, realizing real-time acquisition, access verification and reinforcement learning model optimization adjustment of various streaming data at a server end, realizing distribution and deployment through the interface specifications aiming at an optimized model, and updating the model after receiving by a terminal.
The embodiment of the application adopts another technical scheme that: an electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the following operations of the transmission line defect detection method described above:
the method comprises the following steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprising the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: and training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the power transmission line element on the basis of the target detection of the deep learning.
Compared with the prior art, the embodiment of the application has the advantages that: according to the method, the system and the electronic equipment for detecting the defects of the power transmission line, a power transmission line defect detection model based on visible light, infrared, ultraviolet and other multi-image fusion is established by performing deep learning distributed training on a mass of image samples; aiming at the characteristics of multi-dimension and multi-source isomerism of disaster information, a real-time high-efficiency processing big data platform of multi-mode isomerism mass data is constructed, a reinforcement learning risk assessment model is established, strategy learning of disaster information is carried out, and disaster multi-source information fusion and analysis and risk assessment disaster early warning are achieved. Compared with the prior art, the defect identification accuracy can be improved, and the inspection work efficiency and quality are improved.
Drawings
Fig. 1 is a flowchart of a method for detecting defects of a power transmission line according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a power transmission line defect detection system according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of hardware equipment of the power transmission line defect detection method provided in the embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Please refer to fig. 1, which is a flowchart of a method for detecting defects of a power transmission line according to an embodiment of the present application. The method for detecting the defects of the power transmission line comprises the following steps:
step 100: constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of the power transmission line element, and performing abnormity diagnosis on the power transmission line element;
in step 100, image sample data of the power transmission line elements comprise sample data of visible light, infrared, ultraviolet and the like, and comprise normal annotation images and defect annotation images, wherein the image sample data comprise multi-source images of all types of elements, and the number of the images of each type of elements meets the requirements of deep learning network training. The power transmission line defect detection model based on virtual and real sample integration and transfer learning can realize a learning mode that a source task of target recognition in a virtual scene is transferred to a target task of target recognition in a real scene, realize the fusion of an element automatic marking sample and an artificial marking sample and iterative incremental learning, reduce the workload of an artificial marking training data set and improve overfitting caused by too little defect data. Specifically, the method for constructing the power transmission line defect detection model based on virtual and real sample integration and transfer learning mainly comprises the following steps:
step 101: constructing a virtual sample generation and labeling model of virtual-real integration, which is used for integrating abundant ground feature information in virtual data and real image data to realize effective fusion of the virtual data and the real data;
step 102: constructing a deep learning transfer learning model, and realizing model optimization based on transfer learning;
in step 102, after a virtual sample generation and labeling model of virtual-real integration is constructed, the model is migrated to actual data training, migration learning is realized based on a deep learning framework, a source task of a virtual data learning model is firstly performed, a target task of an actual data learning optimization model is then performed, and model optimization based on migration learning is realized.
Step 103: training a deep learning model based on target detection algorithms such as Spark, TensorFlow, Caffe framework, FasterR-CNN, YOLO, SSD and the like, and carrying out abnormity diagnosis on the power transmission line element on the basis of target detection of deep learning;
in step 103, the deep learning model includes: and respectively constructing a defect target detection model, a multi-model integrated learning model and a fusion detection model aiming at sample data of visible light, infrared and ultraviolet images, completing model training and parameter adjustment optimization, outputting a power transmission line defect detection model, and realizing online intelligent identification of the element defects of the power transmission line.
Step 200: constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, performing strategy learning of disaster information, and realizing disaster multi-source information fusion and analysis and risk assessment disaster early warning;
in step 200, constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis mainly comprises the following steps:
step 201: constructing a power transmission line fault space-time weak correlation analysis model; based on deep mining of historical fault data of the overhead line, researching the space-time characteristics of state characteristics of the overhead line, analyzing the incidence relation between the defect development of the overhead line and the influence of a fault chain effect on the environment, screening different types of fault characteristics from multidimensional state quantities by adopting a characteristic matching algorithm, resolving the intrinsic incidence relation between the fault characteristics of the power transmission line and multidimensional data parameters such as environment information, meteorological information and body characteristic information, constructing a power transmission line fault space-time weak correlation analysis model, and realizing the risk assessment of the overhead line based on the space-time correlation characteristics;
step 202: constructing a risk assessment and fault prediction model of the hidden danger of the line defect;
in step 202, the environments of the transmission line are complex and various, and the wires, insulators, hardware fittings, towers and grounding devices are exposed outside, so that the transmission line is greatly damaged along with the alternation of seasons, the change of climate and the occurrence of extreme natural disasters. In the embodiment of the application, historical operation and maintenance data of the overhead transmission line are rooted on the basis of historical defect monitoring data, meteorological data, external hidden dangers and big data information of body defect conditions according to the environmental characteristics of a transmission corridor and the characteristic information of a tower body, the correlation characteristics of the defects of the transmission line and the meteorological data are analyzed, and the correlation relationship between meteorological factors and the defects and hidden dangers of the transmission line is analyzed by using a time-space data mining algorithm; analyzing the defect occurrence probability of different regions by researching the spatial characteristics of different defect types; and analyzing the occurrence probability of different defects of the overhead transmission line by combining the space-time correlation result and the overhead line defect space distribution characteristic, and constructing line defect hidden danger risk assessment and fault prediction.
Step 203: establishing a reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm;
in step 203, the data preprocessing divides the input information into operation environment data and operation state data, the deep reinforcement learning algorithm adopts a mode of combining a competitive Q network and a dual Q network, the competitive Q network divides a return value into an operation environment return value and an operation state return value, and the dual Q network is responsible for selecting the operation state data and evaluating the operation state effect. The front part of the deep convolutional neural network model framework is used for learning the operation environment data by partial convolutional kernels, and the rest convolutional kernels are used for learning the operation state information data, so that the training efficiency is improved, a uniform standard environment is provided for disaster evaluation comparison under the same condition, and the result is more reasonable. In addition, the operation environment information can be correspondingly changed in different operation modes, and the competitive Q network can obtain different environment return values aiming at different operation modes, so that the generalization capability of the model is enhanced, and the algorithm performance is improved. The risk assessment method starts from power grid operation data directly, and does not need to adjust models for analysis of different defect faults, defect fault types and power grid operation modes, while the traditional method needs to adjust mathematical models according to different operation modes, fault types and topological results. In addition, the influence of various factors of the air environment such as weather and the like is considered at the same time, and more comprehensive auxiliary decision can be provided.
Step 300: a defect detection model distribution deployment system and a model optimization system based on deep learning are constructed according to the power transmission line defect detection model and the risk assessment early warning model, and model mobile deployment, defect identification and risk evaluation application are realized;
in step 300, constructing a defect detection model distribution deployment system and a model optimization system based on deep learning mainly comprises the following steps:
step 301: deploying a model; mobile terminal deployment of various types of terminals is realized based on a Caffe2 framework, and technologies such as weight pruning, weight quantification, model storage, forward acceleration calculation and the like are improved and applied, and equipment identification and a power transmission line defect detection model are deployed at the mobile terminal;
step 302: identifying defects; performing online defect detection on the mobile terminal by using a trained and tested defect detection model, and identifying and marking a defect element in a detection image;
step 303: evaluating the risk; constructing a multi-source heterogeneous mass big data platform of the defects and risks of the power transmission line, and realizing mass data acquisition access;
step 304: a data acquisition access verification and model optimization updating mechanism; the method comprises the steps of formulating interface specifications, realizing real-time acquisition, access verification and reinforcement learning model optimization adjustment of various streaming data at a server end, realizing distribution and deployment through the interface specifications aiming at an optimized model, updating the model after receiving the model by a terminal, and realizing model mobile deployment, defect identification and risk evaluation application.
Please refer to fig. 2, which is a schematic structural diagram of a power transmission line defect detection system according to an embodiment of the present application. The power transmission line defect detection system comprises a defect detection model building module, a risk assessment early warning model building module and a deployment and application module.
A defect detection model construction module: constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of the power transmission line element, and performing abnormity diagnosis on the power transmission line element; the image sample data of the power transmission line elements comprises sample data of visible light, infrared, ultraviolet and the like, and comprises normal annotation images and defect annotation images, wherein the image sample data comprises multi-source images of all types of elements, and the number of the images of each type of elements meets the requirement of deep learning network training. The power transmission line defect detection model based on virtual and real sample integration and transfer learning can realize a learning mode that a source task of target recognition in a virtual scene is transferred to a target task of target recognition in a real scene, realize the fusion of an element automatic marking sample and an artificial marking sample and iterative incremental learning, reduce the workload of an artificial marking training data set and improve overfitting caused by too little defect data.
Specifically, the defect detection model building module comprises:
virtual and real integrated units: the virtual sample generation and labeling model is used for constructing virtual and real integration, and is used for integrating abundant ground feature information in virtual data and real image data to realize effective fusion of the virtual data and the real data;
a transfer learning unit: the method is used for constructing a deep learning transfer learning model and realizing model optimization based on transfer learning; in the embodiment of the application, after a virtual sample generation and labeling model of virtual-real integration is constructed, the model is migrated to actual data training, migration learning is realized based on a deep learning framework, a source task of a virtual data learning model is firstly carried out, a target task of an actual data learning optimization model is then carried out, and model optimization based on the migration learning is realized.
A model training unit: the method is used for training a deep learning model based on Spark, TensorFlow, Caffe framework, FasterR-CNN, YOLO, SSD and other target detection algorithms, and carrying out abnormity diagnosis on the power transmission line element on the basis of target detection of deep learning; wherein, the deep learning model includes: and respectively constructing a defect target detection model, a multi-model integrated learning model and a fusion detection model aiming at sample data of visible light, infrared and ultraviolet images, completing model training and parameter adjustment optimization, outputting a power transmission line defect detection model, and realizing online intelligent identification of the element defects of the power transmission line.
A risk assessment early warning model construction module: the risk assessment early warning system is used for constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, performing strategy learning of disaster information, and achieving disaster multi-source information fusion and analysis and risk assessment disaster early warning;
specifically, the risk assessment early warning model building module comprises:
a first model construction unit: the method is used for constructing a power transmission line fault space-time weak correlation analysis model; based on deep mining of historical fault data of the overhead line, researching the space-time characteristics of state characteristics of the overhead line, analyzing the incidence relation between the defect development of the overhead line and the influence of a fault chain effect on the environment, screening different types of fault characteristics from multidimensional state quantities by adopting a characteristic matching algorithm, resolving the intrinsic incidence relation between the fault characteristics of the power transmission line and multidimensional data parameters such as environment information, meteorological information and body characteristic information, constructing a power transmission line fault space-time weak correlation analysis model, and realizing the risk assessment of the overhead line based on the space-time correlation characteristics;
a second model construction unit: the method is used for constructing a risk assessment and fault prediction model of the hidden danger of the line defect; the transmission line is complex and diverse in environment, wires, insulators, hardware fittings, towers and grounding devices are exposed outside, and great damage is caused to the transmission line along with seasonal alternation, climate change and extreme natural disasters. In the embodiment of the application, historical operation and maintenance data of the overhead transmission line are rooted on the basis of historical defect monitoring data, meteorological data, external hidden dangers and big data information of body defect conditions according to the environmental characteristics of a transmission corridor and the characteristic information of a tower body, the correlation characteristics of the defects of the transmission line and the meteorological data are analyzed, and the correlation relationship between meteorological factors and the defects and hidden dangers of the transmission line is analyzed by using a time-space data mining algorithm; analyzing the defect occurrence probability of different regions by researching the spatial characteristics of different defect types; and analyzing the occurrence probability of different defects of the overhead transmission line by combining the space-time correlation result and the overhead line defect space distribution characteristic, and constructing line defect hidden danger risk assessment and fault prediction.
A third model construction unit: the system is used for establishing a reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm; the data preprocessing divides input information into operation environment data and operation state data, the deep reinforcement learning algorithm adopts a mode of combining a competitive Q network and a dual Q network, the competitive Q network divides return values into operation environment return values and operation state return values, and the dual Q network is responsible for selecting the operation state data and evaluating operation state effects. The front part of the deep convolutional neural network model framework is used for learning the operation environment data by partial convolutional kernels, and the rest convolutional kernels are used for learning the operation state information data, so that the training efficiency is improved, a uniform standard environment is provided for disaster evaluation comparison under the same condition, and the result is more reasonable. In addition, the operation environment information can be correspondingly changed in different operation modes, and the competitive Q network can obtain different environment return values aiming at different operation modes, so that the generalization capability of the model is enhanced, and the algorithm performance is improved. The risk assessment method starts from power grid operation data directly, and does not need to adjust models for analysis of different defect faults, defect fault types and power grid operation modes, while the traditional method needs to adjust mathematical models according to different operation modes, fault types and topological results. In addition, the influence of various factors of the air environment such as weather and the like is considered at the same time, and more comprehensive auxiliary decision can be provided.
Deployment and application module: the method is used for constructing a defect detection model distribution deployment system and a model optimization system based on deep learning according to a defect detection model and a risk assessment early warning model of the power transmission line, and realizing model mobile deployment, defect identification and risk evaluation application;
specifically, the deployment and application module includes:
a model deployment unit: the method is used for realizing mobile terminal deployment of various types of terminals based on a Caffe2 framework, and comprises the improvement and application of technologies such as weight pruning, weight quantification, model storage, forward acceleration calculation and the like, and equipment identification and a power transmission line defect detection model are deployed at a mobile terminal;
a defect identification unit: the defect detection module is used for carrying out online defect detection on the mobile terminal by using the trained and tested defect detection model, and identifying and marking the defect elements in the detected image;
a risk evaluation unit: the multi-source heterogeneous mass big data platform is used for constructing defects and risks of the power transmission line, and mass data acquisition access is realized;
an optimization updating unit: the system is used for realizing data acquisition access verification and model optimization updating mechanism; the method comprises the steps of formulating interface specifications, realizing real-time acquisition, access verification and reinforcement learning model optimization adjustment of various streaming data at a server end, realizing distribution and deployment through the interface specifications aiming at an optimized model, updating the model after receiving the model by a terminal, and realizing model mobile deployment, defect identification and risk evaluation application.
Fig. 3 is a schematic structural diagram of hardware equipment of the power transmission line defect detection method provided in the embodiment of the present application. As shown in fig. 3, the device includes one or more processors and memory. Taking a processor as an example, the apparatus may further include: an input system and an output system.
The processor, memory, input system, and output system may be connected by a bus or other means, as exemplified by the bus connection in fig. 3.
The memory, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules. The processor executes various functional applications and data processing of the electronic device, i.e., implements the processing method of the above-described method embodiment, by executing the non-transitory software program, instructions and modules stored in the memory.
The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data and the like. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and these remote memories may be connected to the processing system over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input system may receive input numeric or character information and generate a signal input. The output system may include a display device such as a display screen.
The one or more modules are stored in the memory and, when executed by the one or more processors, perform the following for any of the above method embodiments:
the method comprises the following steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprising the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: and training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the power transmission line element on the basis of the target detection of the deep learning.
The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the methods provided in the embodiments of the present application.
Embodiments of the present application provide a non-transitory (non-volatile) computer storage medium having stored thereon computer-executable instructions that may perform the following operations:
the method comprises the following steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprising the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: and training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the power transmission line element on the basis of the target detection of the deep learning.
Embodiments of the present application provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions that, when executed by a computer, cause the computer to perform the following:
the method comprises the following steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprising the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: and training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the power transmission line element on the basis of the target detection of the deep learning.
According to the method, the system and the electronic equipment for detecting the defects of the power transmission line, a power transmission line defect detection model based on visible light, infrared, ultraviolet and other multi-image fusion is established by performing deep learning distributed training on a mass of image samples; aiming at the characteristics of multi-dimension and multi-source isomerism of disaster information, a real-time high-efficiency processing big data platform of multi-mode isomerism mass data is constructed, a reinforcement learning risk assessment model is established, strategy learning of disaster information is carried out, and disaster multi-source information fusion and analysis and risk assessment disaster early warning are achieved. Compared with the prior art, the defect identification accuracy can be improved, and the inspection work efficiency and quality are improved.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. The method for detecting the defects of the power transmission line is characterized by constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of elements of the power transmission line, and specifically comprises the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the transmission line element on the basis of the target detection of the deep learning;
the method for detecting the defects of the power transmission line further comprises the following steps: constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, and performing strategy learning of disaster information, disaster multi-source information fusion and analysis and risk assessment disaster early warning; the method for constructing the risk assessment early warning model based on the deep reinforcement learning and the space-time weak correlation analysis specifically comprises the following steps:
step d: based on deep mining of historical fault data of the overhead line, researching the space-time characteristics of state characteristics of the overhead line, analyzing the incidence relation between the defect development of the overhead line and the influence of a fault chain effect on the environment, screening different types of fault characteristics from multidimensional state quantities by adopting a characteristic matching algorithm, resolving the intrinsic incidence relation between the fault characteristics of the power transmission line and multidimensional data parameters of environment information, meteorological information and body characteristic information, and constructing a power transmission line fault space-time weak incidence analysis model;
step e: analyzing the correlation characteristics of the defects of the power transmission line and the meteorological data based on historical defect monitoring data, meteorological data, external hidden dangers and big data information of body defect conditions, and according to the environmental characteristics of the power transmission corridor, the body characteristic information of the tower and historical operation and maintenance data rooted in the overhead power transmission line, and analyzing the correlation relationship between meteorological factors and the defects and hidden dangers of the power transmission line by utilizing a time-space data mining algorithm; analyzing the defect occurrence probability of different regions by researching the spatial characteristics of different defect types; analyzing the occurrence probability of different defects of the overhead transmission line by combining the time-space correlation result and the overhead line defect space distribution characteristic, and constructing a line defect hidden danger risk assessment and fault prediction model;
step f: the method comprises the steps of establishing a reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm, dividing input information into operation environment data and operation state data by data preprocessing, adopting a mode of combining a competitive Q network and a double Q network by the deep reinforcement learning algorithm, dividing a return value into an operation environment return value and an operation state return value by the competitive Q network, selecting the operation state data by the double Q network, evaluating the operation state effect, rolling the operation environment data of kernel learning in the front part of a deep convolutional neural network model frame, and rolling the operation state data of kernel learning in the rest.
2. The method according to claim 1, wherein the image sample data of the transmission line element includes visible light, infrared and ultraviolet sample data, and includes a normal annotation image and a defect annotation image.
3. The method for detecting defects of power transmission lines according to claim 2, wherein in the step c, the deep learning model comprises: and respectively constructing a defect target detection model, a multi-model ensemble learning model and a fusion detection model aiming at the visible light, infrared and ultraviolet image sample data, completing model training and parameter adjustment optimization, outputting a power transmission line defect detection model, and identifying the defects of the power transmission line elements.
4. The method for detecting the defects of the power transmission line according to claim 1, further comprising: constructing a defect detection model distribution deployment system and a model optimization system based on deep learning according to the power transmission line defect detection model and the risk assessment early warning model, and applying model mobile deployment, defect identification and risk evaluation; the method specifically comprises the following steps:
step g: deploying a mobile terminal of the terminal based on a Caffe2 framework, wherein the deployment comprises weight pruning, weight quantification, model storage, forward acceleration calculation improvement and application, equipment identification and a power transmission line defect detection model are deployed at the mobile terminal;
step h: performing online defect detection on the mobile terminal by using a trained and tested defect detection model, and identifying and marking a defect element in a detection image;
step i: constructing a multi-source heterogeneous mass big data platform for defects and risks of the power transmission line, and acquiring and accessing mass data;
step j: interface specifications are formulated, real-time acquisition, access verification and reinforcement learning model optimization adjustment of various streaming data are realized at a server side, distribution and deployment are realized through the interface specifications aiming at an optimized model, and a terminal receives the data to update the model.
5. A transmission line defect detection system, comprising:
a defect detection model construction module: the method comprises the steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements; the defect detection model building module comprises:
virtual and real integrated units: the virtual sample generation and labeling model is used for constructing a virtual-real integrated model, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
a transfer learning unit: the method is used for constructing a deep learning transfer learning model and finishing model optimization based on transfer learning;
a model training unit: the system is used for training a deep learning model based on a target detection algorithm and diagnosing the abnormity of the power transmission line element on the basis of the target detection of the deep learning;
the power transmission line defect detection system further comprises a risk assessment early warning model construction module: the risk assessment early warning model is used for constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, and is used for carrying out strategy learning of disaster information, disaster multi-source information fusion and analysis and risk assessment disaster early warning; the risk assessment early warning model building module specifically comprises:
a first model construction unit: the method is used for deeply mining based on historical fault data of the overhead line, researching the space-time characteristics of the state characteristics of the overhead line, analyzing the incidence relation between the defect development of the overhead line and the influence of a fault chain effect on the environment, screening different types of fault characteristics from multidimensional state quantity by adopting a characteristic matching algorithm, resolving the intrinsic incidence relation between the fault characteristics of the power transmission line and multidimensional data parameters of environment information, meteorological information and body characteristic information, and constructing a power transmission line fault space-time weak incidence analysis model;
a second model construction unit: the system is used for analyzing the correlation characteristics of the defects of the power transmission line and the meteorological data based on historical defect monitoring data, meteorological data, external hidden dangers and big data information of body defect conditions, and analyzing the correlation relationship between meteorological factors and the defects and hidden dangers of the power transmission line by using a time-space data mining algorithm according to the environmental characteristics of a power transmission corridor, the body characteristic information of a tower and historical operation and maintenance data rooted in the overhead power transmission line; analyzing the defect occurrence probability of different regions by researching the spatial characteristics of different defect types; analyzing the occurrence probability of different defects of the overhead transmission line by combining the time-space correlation result and the overhead line defect space distribution characteristic, and constructing a line defect hidden danger risk assessment and fault prediction model;
a third model construction unit: the third model building unit is used for building a reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm, and the specific mode of building the reinforcement learning risk assessment early warning model is as follows: the data preprocessing divides input information into operation environment data and operation state data, a competitive Q network and a double Q network are combined in a deep reinforcement learning algorithm, the competitive Q network divides return values into operation environment return values and operation state return values, the double Q network is responsible for selecting the operation state data and evaluating operation state effects, the front part of a deep convolution neural network model frame is rolled up to learn the operation environment data, and the rest of convolution kernels learn the operation state data.
6. The system of claim 5, wherein the image sample data of the transmission line component includes visible light, infrared and ultraviolet sample data, and includes a normal annotation image and a defect annotation image.
7. The system of claim 6, wherein the deep learning model comprises: and respectively constructing a defect target detection model, a multi-model ensemble learning model and a fusion detection model aiming at the visible light, infrared and ultraviolet image sample data, completing model training and parameter adjustment optimization, outputting a power transmission line defect detection model, and identifying the defects of the power transmission line elements.
8. The system of claim 5, further comprising a deployment and application module: the system comprises a fault detection model distribution and deployment system and a model optimization system, wherein the fault detection model distribution and deployment system and the model optimization system are used for constructing a fault detection model distribution and deployment system and a model optimization system based on deep learning according to the power transmission line fault detection model and the risk assessment early warning model and are used for model mobile deployment, fault identification and risk evaluation application; the deployment and application module specifically includes:
a model deployment unit: the method is used for deploying the mobile terminal of the terminal based on the Caffe2 framework, and comprises weight pruning, weight quantification, model storage, forward acceleration calculation improvement and application, equipment identification and a power transmission line defect detection model are deployed at the mobile terminal;
a defect identification unit: the defect detection module is used for carrying out online defect detection on the mobile terminal by using the trained and tested defect detection model, and identifying and marking the defect elements in the detected image;
a risk evaluation unit: the multi-source heterogeneous mass big data platform is used for constructing defects and risks of the power transmission line and is used for acquiring and accessing mass data;
an optimization updating unit: the method is used for formulating interface specifications, realizing real-time acquisition, access verification and reinforcement learning model optimization adjustment of various streaming data at a server end, realizing distribution and deployment through the interface specifications aiming at an optimized model, and updating the model after receiving by a terminal.
9. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the following operations of the transmission line defect detection method of any one of claims 1 to 4:
the method comprises the following steps of constructing a power transmission line defect detection model based on virtual and real sample integration and transfer learning according to image sample data of power transmission line elements, and specifically comprising the following steps:
step a: constructing a virtual sample generation and labeling model of virtual-real integration, integrating abundant ground feature information in virtual data and real image data, and fusing the virtual data and the real data;
step b: constructing a deep learning transfer learning model, and completing model optimization based on transfer learning;
step c: training a deep learning model based on a target detection algorithm, and diagnosing the abnormality of the transmission line element on the basis of the target detection of the deep learning;
the method for detecting the defects of the power transmission line further comprises the following steps: constructing a risk assessment early warning model based on deep reinforcement learning and space-time weak correlation analysis according to the power transmission line defect detection model, and performing strategy learning of disaster information, disaster multi-source information fusion and analysis and risk assessment disaster early warning; the method for constructing the risk assessment early warning model based on the deep reinforcement learning and the space-time weak correlation analysis specifically comprises the following steps:
step d: based on deep mining of historical fault data of the overhead line, researching the space-time characteristics of state characteristics of the overhead line, analyzing the incidence relation between the defect development of the overhead line and the influence of a fault chain effect on the environment, screening different types of fault characteristics from multidimensional state quantities by adopting a characteristic matching algorithm, resolving the intrinsic incidence relation between the fault characteristics of the power transmission line and multidimensional data parameters of environment information, meteorological information and body characteristic information, and constructing a power transmission line fault space-time weak incidence analysis model;
step e: analyzing the correlation characteristics of the defects of the power transmission line and the meteorological data based on historical defect monitoring data, meteorological data, external hidden dangers and big data information of body defect conditions, and according to the environmental characteristics of the power transmission corridor, the body characteristic information of the tower and historical operation and maintenance data rooted in the overhead power transmission line, and analyzing the correlation relationship between meteorological factors and the defects and hidden dangers of the power transmission line by utilizing a time-space data mining algorithm; analyzing the defect occurrence probability of different regions by researching the spatial characteristics of different defect types; analyzing the occurrence probability of different defects of the overhead transmission line by combining the time-space correlation result and the overhead line defect space distribution characteristic, and constructing a line defect hidden danger risk assessment and fault prediction model;
step f: the method comprises the steps of establishing a reinforcement learning risk assessment early warning model by adopting a deep reinforcement learning algorithm, dividing input information into operation environment data and operation state data by data preprocessing, adopting a mode of combining a competitive Q network and a double Q network by the deep reinforcement learning algorithm, dividing a return value into an operation environment return value and an operation state return value by the competitive Q network, selecting the operation state data by the double Q network, evaluating the operation state effect, rolling the operation environment data of kernel learning in the front part of a deep convolutional neural network model frame, and rolling the operation state data of kernel learning in the rest.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811547406.6A CN109785289B (en) | 2018-12-18 | 2018-12-18 | Transmission line defect detection method and system and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811547406.6A CN109785289B (en) | 2018-12-18 | 2018-12-18 | Transmission line defect detection method and system and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109785289A CN109785289A (en) | 2019-05-21 |
| CN109785289B true CN109785289B (en) | 2021-07-20 |
Family
ID=66497070
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811547406.6A Active CN109785289B (en) | 2018-12-18 | 2018-12-18 | Transmission line defect detection method and system and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109785289B (en) |
Families Citing this family (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288578A (en) * | 2019-06-24 | 2019-09-27 | 国网上海市电力公司 | A kind of power equipments defect infrared image recognizing system of high discrimination |
| CN110473178A (en) * | 2019-07-30 | 2019-11-19 | 上海深视信息科技有限公司 | A kind of open defect detection method and system based on multiple light courcess fusion |
| CN110599460A (en) * | 2019-08-14 | 2019-12-20 | 深圳市勘察研究院有限公司 | Underground pipe network detection and evaluation cloud system based on hybrid convolutional neural network |
| CN110599459A (en) * | 2019-08-14 | 2019-12-20 | 深圳市勘察研究院有限公司 | Underground pipe network risk assessment cloud system based on deep learning |
| CN110503644B (en) * | 2019-08-27 | 2023-07-25 | 广东工业大学 | Defect detection implementation method based on mobile platform, defect detection method and related equipment |
| CN110502034B (en) * | 2019-09-04 | 2022-08-09 | 中国人民解放军国防科技大学 | Fixed-wing unmanned aerial vehicle cluster control method based on deep reinforcement learning |
| CN110751642A (en) * | 2019-10-18 | 2020-02-04 | 国网黑龙江省电力有限公司大庆供电公司 | Insulator crack detection method and system |
| CN111027402B (en) * | 2019-11-15 | 2022-09-02 | 国网江西省电力有限公司电力科学研究院 | Intelligent detection and evaluation method for natural disaster risk |
| CN111062529A (en) * | 2019-12-10 | 2020-04-24 | 国网北京市电力公司 | Power line processing method and device, storage medium and processor |
| CN111199213B (en) * | 2020-01-03 | 2023-09-26 | 云南电网有限责任公司电力科学研究院 | Method and device for detecting defects of equipment for transformer substation |
| CN111274880A (en) * | 2020-01-10 | 2020-06-12 | 丽水正阳电力建设有限公司 | Video intelligent analysis auxiliary inspection and abnormity warning method |
| CN111310785A (en) * | 2020-01-15 | 2020-06-19 | 杭州华网信息技术有限公司 | National power grid mechanical external damage prediction method |
| CN111311967A (en) * | 2020-03-31 | 2020-06-19 | 普宙飞行器科技（深圳）有限公司 | Unmanned aerial vehicle-based power line inspection system and method |
| CN111583196B (en) * | 2020-04-22 | 2021-09-07 | 北京智芯微电子科技有限公司 | Monitoring system and monitoring method for power transmission line |
| CN111598138A (en) * | 2020-04-24 | 2020-08-28 | 山东易华录信息技术有限公司 | Neural network learning image identification method and device |
| CN111784692A (en) * | 2020-08-11 | 2020-10-16 | 国网内蒙古东部电力有限公司 | Method and device for detecting insulator defects in power system and electronic equipment |
| CN112036450B (en) * | 2020-08-12 | 2024-02-23 | 国家电网有限公司 | High-voltage cable partial discharge mode identification method and system based on transfer learning |
| CN112132065B (en) * | 2020-09-25 | 2021-08-20 | 智洋创新科技股份有限公司 | Alarm strategy method based on power transmission line channel visual continuous alarm |
| CN112734692B (en) * | 2020-12-17 | 2023-12-22 | 国网信息通信产业集团有限公司 | Defect identification method and device for power transformation equipment |
| CN112633677A (en) * | 2020-12-19 | 2021-04-09 | 大秦铁路股份有限公司 | Method for evaluating line quality based on railway power supply specialty |
| CN114764112B (en) * | 2021-01-14 | 2024-03-22 | 广州中国科学院先进技术研究所 | Non-access type machine fault prediction method |
| CN112785138A (en) * | 2021-01-18 | 2021-05-11 | 内蒙古电力（集团）有限责任公司呼和浩特供电局 | Method for carrying out three-span line monitoring analysis early warning based on numerical weather |
| CN113052296A (en) * | 2021-03-02 | 2021-06-29 | 贵州电网有限责任公司 | Power grid power transmission defect intelligent management system based on deep learning convolutional neural network technology |
| CN113192017A (en) * | 2021-04-21 | 2021-07-30 | 上海东普信息科技有限公司 | Package defect identification method, device, equipment and storage medium |
| CN113095321B (en) * | 2021-04-22 | 2023-07-11 | 武汉菲舍控制技术有限公司 | Roller bearing temperature measurement and fault early warning method and device for belt conveyor |
| CN113537341A (en) * | 2021-07-14 | 2021-10-22 | 安徽炬视科技有限公司 | Online monitoring device and identification method for line hidden danger based on big data and self-learning |
| CN113610252B (en) * | 2021-08-17 | 2023-09-19 | 浙江捷瑞电力科技有限公司 | System and method for detecting and alarming defect target of power transmission line by meta-learning small sample |
| CN113850302B (en) * | 2021-09-02 | 2023-08-29 | 杭州海康威视数字技术股份有限公司 | Incremental learning method, device and equipment |
| CN114167225A (en) * | 2021-10-07 | 2022-03-11 | 国网山东省电力公司潍坊供电公司 | Ultraviolet light detection device and product for automatically identifying defects of power transmission line |
| CN117314896B (en) * | 2023-11-28 | 2024-02-06 | 国网湖北省电力有限公司 | Power system abnormality detection method and system based on deep learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105631578A (en) * | 2015-12-10 | 2016-06-01 | 浙江大学 | Risk assessment-orientated modeling method of power transmission and transformation equipment failure probability model |
| CN106651188A (en) * | 2016-12-27 | 2017-05-10 | 贵州电网有限责任公司贵阳供电局 | Electric transmission and transformation device multi-source state assessment data processing method and application thereof |
| CN106981063A (en) * | 2017-03-14 | 2017-07-25 | 东北大学 | A kind of grid equipment state monitoring apparatus based on deep learning |
| CN106980922A (en) * | 2017-03-03 | 2017-07-25 | 国网天津市电力公司 | A kind of power transmission and transformation equipment state evaluation method based on big data |
| CN108010030A (en) * | 2018-01-24 | 2018-05-08 | 福州大学 | A kind of Aerial Images insulator real-time detection method based on deep learning |
| CN108174165A (en) * | 2018-01-17 | 2018-06-15 | 重庆览辉信息技术有限公司 | Electric power safety operation and O＆M intelligent monitoring system and method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8713025B2 (en) * | 2005-03-31 | 2014-04-29 | Square Halt Solutions, Limited Liability Company | Complete context search system |
| US8874498B2 (en) * | 2011-09-16 | 2014-10-28 | International Business Machines Corporation | Unsupervised, supervised, and reinforced learning via spiking computation |
| CN105228158A (en) * | 2015-11-11 | 2016-01-06 | 国家电网公司 | Based on the cognition wireless network cooperative node selection method of intensified learning |
| CN106448670B (en) * | 2016-10-21 | 2019-11-19 | 竹间智能科技（上海）有限公司 | Conversational system is automatically replied based on deep learning and intensified learning |
| CN108365615B (en) * | 2018-02-08 | 2021-02-09 | 华中科技大学 | Self-adaptive wide area damping controller and control method |
| CN108520472A (en) * | 2018-02-28 | 2018-09-11 | 北京邮电大学 | A kind of method, apparatus and electronic equipment of processing electric power system data |
-
2018
- 2018-12-18 CN CN201811547406.6A patent/CN109785289B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105631578A (en) * | 2015-12-10 | 2016-06-01 | 浙江大学 | Risk assessment-orientated modeling method of power transmission and transformation equipment failure probability model |
| CN106651188A (en) * | 2016-12-27 | 2017-05-10 | 贵州电网有限责任公司贵阳供电局 | Electric transmission and transformation device multi-source state assessment data processing method and application thereof |
| CN106980922A (en) * | 2017-03-03 | 2017-07-25 | 国网天津市电力公司 | A kind of power transmission and transformation equipment state evaluation method based on big data |
| CN106981063A (en) * | 2017-03-14 | 2017-07-25 | 东北大学 | A kind of grid equipment state monitoring apparatus based on deep learning |
| CN108174165A (en) * | 2018-01-17 | 2018-06-15 | 重庆览辉信息技术有限公司 | Electric power safety operation and O＆M intelligent monitoring system and method |
| CN108010030A (en) * | 2018-01-24 | 2018-05-08 | 福州大学 | A kind of Aerial Images insulator real-time detection method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109785289A (en) | 2019-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109785289B (en) | Transmission line defect detection method and system and electronic equipment | |
| WO2022077605A1 (en) | Wind turbine blade image-based damage detection and localization method | |
| CN110705727B (en) | Photovoltaic power station shadow shielding diagnosis method and system based on random forest algorithm | |
| CN111008641B (en) | Power transmission line tower external force damage detection method based on convolutional neural network | |
| CN116614177B (en) | Optical fiber state multidimensional parameter monitoring system | |
| CN113472079A (en) | Power distribution station operation and maintenance monitoring cloud robot system, background processing and operation task method | |
| CN112906654A (en) | Anti-vibration hammer detection method based on deep learning algorithm | |
| CN114863118A (en) | Self-learning identification system and method based on external hidden danger of power transmission line | |
| CN115393347A (en) | Intelligent power grid inspection method and system based on urban brain | |
| CN113066070A (en) | Multi-source data fusion interaction method in three-dimensional scene | |
| CN115912183B (en) | Ecological measure inspection method and system for high-voltage transmission line and readable storage medium | |
| CN108470141B (en) | Statistical feature and machine learning-based insulator identification method in distribution line | |
| CN115690505A (en) | Photovoltaic module fault detection method and device, computer equipment and storage medium | |
| CN116739963A (en) | Power grid equipment defect detection method based on multi-level multi-scale feature fusion | |
| CN113284103B (en) | Substation equipment defect online detection method based on space transformation fast R-CNN model | |
| CN115236451A (en) | Power distribution network overhead line fault monitoring device and method | |
| CN113033556A (en) | Insulator rapid distinguishing and positioning method and system based on machine vision | |
| Jiang et al. | Research on Lightweight Method of Image Deep Learning Model for Power Equipment | |
| CN112070730A (en) | Anti-vibration hammer falling detection method based on power transmission line inspection image | |
| Lowin et al. | From physical to virtual: leveraging drone imagery to automate photovoltaic system maintenance | |
| Bao et al. | Automatic identification and defect diagnosis of transmission line insulators based on YOLOv3 network | |
| Vahidi et al. | Fault detection and classification in PV arrays using machine learning algorithms in the presence of noisy data | |
| CN113205487B (en) | Cable state detection method based on residual error network fusion heterogeneous data | |
| Hong-Bin et al. | Key fittings monitoring method of transmission line based on cloud-edge cooperation | |
| Zhang et al. | An automatic defect detection method for gas insulated switchgear |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240131 Address after: 519085 101, Building 5, Longyuan Smart Industrial Park, No. 2, Hagongda Road, Tangjiawan Town, High-tech Zone, Zhuhai City, Guangdong Province Patentee after: ZHUHAI INSTITUTE OF ADVANCED TECHNOLOGY CHINESE ACADEMY OF SCIENCES Co.,Ltd. Country or region after: China Address before: 1068 No. 518055 Guangdong city of Shenzhen province Nanshan District Shenzhen University city academy Avenue Patentee before: SHENZHEN INSTITUTES OF ADVANCED TECHNOLOGY CHINESE ACADEMY OF SCIENCES Country or region before: China |