CN114694050A - Power equipment running state detection method based on infrared image - Google Patents
Power equipment running state detection method based on infrared image Download PDFInfo
- Publication number
- CN114694050A CN114694050A CN202210223869.7A CN202210223869A CN114694050A CN 114694050 A CN114694050 A CN 114694050A CN 202210223869 A CN202210223869 A CN 202210223869A CN 114694050 A CN114694050 A CN 114694050A
- Authority
- CN
- China
- Prior art keywords
- power equipment
- infrared image
- state
- network
- temperature
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method for detecting the running state of electric power equipment based on an infrared image, and belongs to the field of automation. The method comprises the following steps: s1: collecting and integrating data of each inspection platform of the power grid, and performing uniform preprocessing for subsequent algorithms; s2: identifying and positioning the electric power equipment based on a fast _ RCNN algorithm, and making a response strategy of the corresponding state of the electric power equipment; s3: and (4) assisting in judging the state detection of the power equipment by assisting in a temperature threshold method, and implementing a corresponding response strategy. The existing scheme for detecting and identifying the infrared image state of the power equipment is improved, the scheme can be used for detecting, identifying and positioning the mark of the infrared image state of the power equipment, the infrared image information is automatically processed through an algorithm model, manual judgment is omitted or a traditional machine identification method is used, and a more efficient, more accurate and more generalized state detection model is established.
Description
Technical Field
The invention belongs to the field of automation, and relates to a power equipment running state detection method based on an infrared image.
Background
The current infrared detection and identification technology based on the state of the power equipment mainly comprises a first image feature comparison method. And observing abnormal red and white areas in the acquired infrared images by using the eyes of a human body, and then comparing the acquired infrared images with similar equipment and comparing the acquired infrared images with the acquired infrared images. And (2) a threshold value discrimination method. The software sets the maximum allowable temperature, and when the temperature of the equipment is detected to be close to or reach the set threshold value, the abnormal state of the equipment is represented. And thirdly, calculating the temperature difference. And setting a detection standard, calculating a temperature rise ratio according to the standard, and paying attention when the temperature rise is too large. And fourthly, an archive analysis method. And performing comparative analysis according to the historical data of the transformer substation and the existing test data. The methods have respective advantages and can be applied to infrared detection of the state of the power equipment, but the efficiency is not high, the intelligence is not high, and a large amount of infrared images are difficult to process under the condition that the power equipment is increased by manpower. The method mainly applies an image characteristic comparison method, replaces human eyes for judgment by a detection algorithm, realizes automatic detection, and is used as a one-step supplement for other threshold methods and temperature difference methods
The existing identification and positioning aiming at the infrared image state of the power equipment still has the following problems and difficulties, (1) the situations of unclear infrared image, angle deviation and unobvious equipment characteristics of the power equipment can occur under the influence of shooting angle quality, environment and the like. The current transformer substation is made to face the problems of complex background, difficult image identification and the like; (2) the intelligent inspection robot has the advantages that the image processing speed of mass equipment generated by inspection is slow, the intelligence is lack, and the judgment of human eyes is relatively relied; (3) the condition of missing detection of the power equipment exists, and the detection real-time performance is not effectively realized.
Aiming at the problem that the state of an infrared image of the traditional substation power equipment cannot be identified and positioned efficiently and intelligently, a set of power equipment running state detection and identification system which is efficient and high in identification accuracy is designed for a power system based on deep learning. The system trains and tests the model by applying the classic fault characteristic diagram of the infrared images of mass power equipment, so that the model can identify, classify and position the defects of the input infrared images, adopts a target detection method based on fast RCNN and is assisted with a temperature threshold method, achieves the state identification and positioning of the power equipment, greatly reduces the omission factor, improves the detection efficiency and has high identification accuracy.
Disclosure of Invention
In view of the above, the present invention provides a method for detecting an operating state of an electrical device based on an infrared image. The existing power equipment state identification scheme based on the infrared images is improved, and the functions of automatic learning and automatic identification of the infrared image characteristics of the power equipment, high-efficiency processing of mass infrared images, real-time detection and positioning of defects of the power equipment and the like are realized.
In order to achieve the purpose, the invention provides the following technical scheme:
an electric power equipment operation state detection method based on infrared images comprises the following steps:
s1: collecting and integrating data of each inspection platform of the power grid, and performing uniform preprocessing for subsequent algorithms;
s2: identifying and positioning the electric power equipment based on a fast _ RCNN algorithm, and making a response strategy of the corresponding state of the electric power equipment;
s3: and (4) assisting in judging the state detection of the power equipment by assisting in a temperature threshold method, and implementing a corresponding response strategy.
Optionally, the method specifically includes:
1. preparing infrared image data set of electric power equipment
Collecting infrared images of a transformer, a circuit breaker, a current transformer, a lightning arrester and a voltage transformer from a classic fault gallery of the power equipment, and carrying out unified specification and marking processing on the collected infrared images;
2. power equipment identification and positioning by fast RCNN algorithm
(ii) fast RCNN Algorithm Structure
The Fast RCNN algorithm is regarded as the synthesis of two modules, RPN region suggestion and a Fast RCNN detection module, and the specific network structure is divided into three parts:
1) extracting characteristics; selecting a deep residual error network ResNet101 as a base network to extract features, wherein the accuracy of a training set is not reduced along with the deepening of the network;
2) a region suggestion; inputting a feature map extracted by a base network, performing sliding scanning on the feature map by using a sliding window, setting detection targets in a plurality of region coverage maps with different sizes and aspect ratios, connecting the extracted features to a classification layer and an edge regression layer, wherein the edge regression layer enables a region suggestion obtained in a region outside a reference window to be very close to the targets, and obtaining the category and the probability of the input feature through the classification layer;
3) detecting; the detection part comprises a core algorithm ROI Pooling layer of fast RCNN, candidate areas obtained by the network part are suggested in the areas and are different in size, the ROI Pooling layer extracts features of fixed scale by setting ROI Pooling of different scales, the classification of the target area is carried out through a full connection layer and a softmax layer, and a positioning coordinate frame is obtained by regression with a bounding box,
fast RCNN loss function
The index i represents a certain reference window in a minimum training step; p is a radical ofiIs the probability that the ith reference window is the model detection target,
is a true border label, if the reference window is a negative example, then
If the reference window is a positive sample, then
LregThe (·) term is activated only if the reference sample is a positive sample; f. ofi＝{fx,fy,fw,fhIs a coordinate vector used to predict the coordinates of the center point of the frame, as well as the width and height; f. ofi *Is the coordinate vector of the real frame, the target loss function is divided into two parts, one is the classification loss Lcls(. II) regression loss Lreg(. to) default value λ 10, NclsAnd NregRespectively for standardizing Lcls(. and L)reg(. in which N isclsDefaults to 256, N minimum training step sizeregThe size of the feature map is taken to be about 2400; (initialized with a range of reference window position values to 2400) a loss function;
③ fast RCNN model training
The RPN area suggestion network, the detection network and the Fast CNN module are independently trained, and the characteristics are not shared; enabling two networks which are not shared to share the characteristics by an alternate training method so as to obtain a classification result and an optimal positioning frame;
3. temperature threshold method for determining state of power equipment
Setting three upper temperature limit values as judgment standard temperatures, wherein the temperatures are respectively 50 ℃, 80 ℃ and 110 ℃, the temperature of 50 ℃ is a common defect, which indicates that the power equipment has no safety threat for the moment in operation but needs to be enhanced for subsequent monitoring, the temperature of 80 ℃ is a serious defect, which indicates that the power equipment has the safety threat possibly and needs to be processed as soon as possible, and the temperature of 110 ℃ is a dangerous defect, which indicates that the power equipment has the very serious safety threat and needs to be processed immediately;
the method for recognizing and positioning the state of the electric power equipment based on the fast RCNN sometimes has the condition that the electric power equipment in the infrared image is missed to be detected, if potential safety hazards exist in the missed equipment, huge loss is possibly caused, the temperature threshold value of the whole infrared image is judged, and when the detected temperature exceeds three set temperature standards, corresponding alarm signals are sent out automatically so that a worker can take corresponding measures in time;
the method comprises the steps of selecting a fast RCNN algorithm of a ResNet101 network to detect and identify the state of an infrared image of the power equipment, and assisting in carrying out temperature judgment on the whole infrared image by a temperature threshold method.
The invention has the beneficial effects that: the existing power equipment infrared image state detection and identification scheme is improved, the method can be used for power equipment infrared image state detection and identification and positioning marks, infrared image information is automatically processed through an algorithm model, manual judgment is omitted or a traditional machine identification method is used, and a more efficient, more accurate and more generalized state detection model is established.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a step of target detection based on fast RCNN;
FIG. 2 illustrates power device infrared image acquisition;
FIG. 3 is a diagram of the fast RCNN algorithm structure;
FIG. 4 is a flow chart of power device status identification;
FIG. 5 is a flow chart of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
Referring to fig. 1 to 5, the present invention includes the following components:
(1) collecting and integrating data of each inspection platform of the power grid, and performing uniform preprocessing for subsequent algorithms;
(2) identifying and positioning the electric power equipment based on a fast _ RCNN algorithm, and making a response strategy of the corresponding state of the electric power equipment;
(3) the method is characterized in that a temperature threshold value method is used for assisting in judging the state detection of the power equipment, and a corresponding response strategy is implemented;
1. preparing an electrical device infrared image dataset
Collecting infrared images of equipment such as a transformer, a circuit breaker, a current transformer, a lightning arrester, a voltage transformer and the like from a classical fault map library of power equipment of a certain provincial power grid company, and carrying out unified specification and marking processing on the collected infrared images;
2. power equipment identification and positioning by fast RCNN algorithm
(ii) fast RCNN Algorithm Structure
The Fast RCNN algorithm can be mainly regarded as the synthesis of two modules, RPN region suggestion and Fast RCNN detection module, and the specific network structure can be divided into three parts:
1) and (5) feature extraction. A deep residual error network ResNet101 is selected as a base network for extracting features, and the accuracy of a training set is not reduced along with the deepening of the network.
2) And (4) area suggestion. Inputting a feature map extracted by a base network, setting a detection target in a coverage map which has different sizes and aspect ratios and can be provided with a plurality of regional suggestions more comprehensively by using sliding window sliding scanning on the feature map, connecting the extracted features to a classification layer and an edge regression layer, wherein the edge regression layer enables the regional suggestions obtained in the region outside a reference window to be very close to the target, and obtaining the category and the probability of the input feature through the classification layer;
3) and (6) detecting. The detection part comprises a core algorithm ROI Pooling layer of fast RCNN, candidate areas obtained by the area suggestion network part are different in size, the ROI Pooling layer extracts features of fixed scale by setting ROI Pooling of different scales, the classification of the target area is carried out through a full connection layer and a softmax layer, a positioning coordinate frame is obtained by using bounding box regression,
fast RCNN loss function
The gradual convergence of the loss function represents that the model training successfully lays a foundation.
The index i represents a certain reference window in a minimum training step; p is a radical ofiIs the probability that the ith reference window is the model detection target,
is a true border label, if the reference window is a negative example, then
If the reference window is a positive sample, then
LregThe (·) term is activated only if the reference sample is a positive sample; f. ofi＝{fx,fy,fw,fhIs a coordinate vector used to predict the coordinates of the center point of the frame, as well as the width and height; f. ofi *Is the coordinate vector of the real frame, the target loss function is divided into two parts, one is the classification loss Lcls(. II) regression loss Lreg(. cndot.), default value of λ 10, NclsAnd NregRespectively for standardizing Lcls(. and L)reg(. in) wherein N isclsDefaults to a minimum training step size of 256, NregThe size of the feature map is taken to be about 2400. (initialized with a range of reference window position values to 2400) a penalty function.
③ fast RCNN model training
The RPN region suggests that the network and detection network and Fast CNN modules are trained independently and features are not shared. Enabling two networks which are not shared to share the characteristics by an alternate training method so as to obtain a classification result and an optimal positioning frame;
3. temperature threshold method for determining state of power equipment
The surface temperature judgment is a very simple and effective method, and can quickly detect whether the running state of the power equipment is abnormal or not. According to literature and experience of professional workers of a power grid, three upper temperature limit values can be set as judgment standard temperatures, the temperatures are 50 ℃, 80 ℃ and 110 ℃, the common defects of 50 ℃ and 50 ℃ respectively indicate that the power equipment has no safety threat during operation but needs to be enhanced for subsequent monitoring, the serious defect of 80 ℃ indicates that the power equipment is very likely to have the safety threat and needs to be processed as soon as possible, and the dangerous defect of 110 ℃ indicates that the power equipment has the very serious safety threat and needs to be processed immediately;
the method for recognizing and positioning the state of the electric power equipment based on the fast RCNN can sometimes cause the condition of missed detection of the electric power equipment in the infrared image, if the potential safety hazard of the missed detection equipment is not found in time, huge loss can be caused, the safe operation of an electric power system is threatened, and the construction of a strong intelligent power grid is not facilitated.
In conclusion, the fast RCNN algorithm of the ResNet101 network is selected to perform state detection and identification on the infrared images of the electric power equipment, the temperature judgment of the whole infrared images is assisted by a temperature threshold method, the missing detection situation is reduced, and the functions of intelligently and efficiently processing mass infrared images and performing state identification on the electric power equipment are realized.
Based on the method for detecting, identifying and positioning the state of the electrical equipment based on the fast RCNN algorithm and assisted by the temperature threshold method, firstly, the inspection robot is bound with the infrared imager to obtain the infrared image data of the electrical equipment, and then:
(1) unifying the data specification, marking the points with abnormal temperature in the infrared image through data marking software, and arranging files as labels for infrared image training;
(2) extracting the characteristics of the preprocessed pictures, calculating training samples and testing samples by using an algorithm model, and detecting, identifying and positioning the states of the power equipment according to the set temperature grade;
(3) the temperature judgment is carried out on the whole infrared image by a temperature threshold method in an auxiliary manner so as to reduce the possible missing detection condition of the algorithm, and finally, a corresponding subsequent response program is started for the obtained result;
the invention divides the temperature of the power equipment into 3 grades, 50 ℃, 80 ℃ and 110 ℃, which respectively correspond to ordinary, serious and dangerous dangers.
Finally, the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all that should be covered by the claims of the present invention.
Claims (2)
1. A method for detecting the running state of electric equipment based on infrared images is characterized in that: the method comprises the following steps:
s1: collecting and integrating data of each inspection platform of the power grid, and performing uniform preprocessing for subsequent algorithms;
s2: identifying and positioning the electric power equipment based on a fast _ RCNN algorithm, and making a response strategy of the corresponding state of the electric power equipment;
s3: and (4) assisting in judging the state detection of the power equipment by assisting in a temperature threshold method, and implementing a corresponding response strategy.
2. The method for detecting the operating state of the electrical equipment based on the infrared image as claimed in claim 1, wherein: the method specifically comprises the following steps:
s11: preparing an electrical device infrared image dataset
Collecting infrared images of a transformer, a circuit breaker, a current transformer, a lightning arrester and a voltage transformer from a classic fault gallery of the power equipment, and carrying out unified specification and marking processing on the collected infrared images;
s12: power equipment identification and positioning by fast RCNN algorithm
(ii) fast RCNN Algorithm Structure
The Fast RCNN algorithm is regarded as the synthesis of two modules, RPN region suggestion and a Fast RCNN detection module, and the specific network structure is divided into three parts:
1) extracting characteristics; selecting a deep residual error network ResNet101 as a base network to extract features, wherein the accuracy of a training set is not reduced along with the deepening of the network;
2) a region suggestion; inputting a feature map extracted by a base network, performing sliding scanning on the feature map by using a sliding window, setting detection targets in a plurality of region coverage maps with different sizes and aspect ratios, connecting the extracted features to a classification layer and an edge regression layer, wherein the edge regression layer enables region suggestions obtained in regions outside a reference window to be very close to the targets, and obtaining the category and the probability of the input features through the classification layer;
3) detecting; the detection part comprises a core algorithm ROI Pooling layer of fast RCNN, candidate areas obtained by the network part are suggested in the areas and are different in size, the ROI Pooling layer extracts features of fixed scale by setting ROI Pooling of different scales, the classification of the target area is carried out through a full connection layer and a softmax layer, a positioning coordinate frame is obtained by regression through a bounding box,
fast RCNN loss function
The index i represents a certain reference window in a minimum training step; p is a radical ofiIs the probability that the ith reference window is the model detection target,
is a true border label, if the reference window is a negative example, then
If the reference window is a positive sample, then
LregThe (·) term is activated only if the reference sample is a positive sample; f. ofi＝{fx,fy,fw,fhIs a coordinate vector used to predict the coordinates of the center point of the frame, as well as the width and height; f. ofi *Is the coordinate vector of the real frame, the target loss function is divided into two parts, one is the classification loss Lcls(. C), which is the regression loss Lreg(. to) default value λ 10, NclsAnd NregRespectively for standardizing Lcls(. and L)reg(. in which N isclsDefaults to 256, N minimum training step sizeregThe size of the feature map is taken to be about 2400; (initialized with a range of reference window position values to 2400) a loss function;
(iii) FasterRCNN model training
The RPN area suggestion network, the detection network and the Fast CNN module are independently trained, and the characteristics are not shared; enabling two networks which are not shared to share the characteristics by an alternate training method so as to obtain a classification result and an optimal positioning frame;
s13: temperature threshold method for determining state of power equipment
Setting three upper temperature limit values as judgment standard temperatures, wherein the temperatures are respectively 50 ℃, 80 ℃ and 110 ℃, the temperature of 50 ℃ is a common defect, which indicates that the power equipment has no safety threat for the moment in operation but needs to be enhanced for subsequent monitoring, the temperature of 80 ℃ is a serious defect, which indicates that the power equipment has the high possibility of having the safety threat and needs to be processed as soon as possible, and the temperature of 110 ℃ is a dangerous defect, which indicates that the power equipment has the high possibility of having the safety threat and needs to be processed immediately;
the method for recognizing and positioning the state of the electric power equipment based on the fast RCNN sometimes has the condition that the electric power equipment in the infrared image is missed, if potential safety hazards exist in the missed equipment, huge loss is possibly caused, the temperature threshold value of the whole infrared image is judged, and when the detected temperature exceeds three set temperature standards, corresponding alarm signals are respectively sent out so that a worker can take corresponding measures in time;
a fast RCNN algorithm of a ResNet101 network is selected to detect and identify the state of the infrared image of the power equipment, and the temperature of the whole infrared image is judged by a temperature threshold method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210223869.7A CN114694050A (en) | 2022-03-09 | 2022-03-09 | Power equipment running state detection method based on infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210223869.7A CN114694050A (en) | 2022-03-09 | 2022-03-09 | Power equipment running state detection method based on infrared image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114694050A true CN114694050A (en) | 2022-07-01 |
Family
ID=82136646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210223869.7A Pending CN114694050A (en) | 2022-03-09 | 2022-03-09 | Power equipment running state detection method based on infrared image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114694050A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115908407A (en) * | 2023-01-05 | 2023-04-04 | 佰聆数据股份有限公司 | Power equipment defect detection method and device based on infrared image temperature value |
| CN117575165A (en) * | 2023-12-05 | 2024-02-20 | 浙江万胜智通科技有限公司 | Intelligent patrol management method and system for digital power distribution network |
-
2022
- 2022-03-09 CN CN202210223869.7A patent/CN114694050A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115908407A (en) * | 2023-01-05 | 2023-04-04 | 佰聆数据股份有限公司 | Power equipment defect detection method and device based on infrared image temperature value |
| CN115908407B (en) * | 2023-01-05 | 2023-05-02 | 佰聆数据股份有限公司 | Power equipment defect detection method and device based on infrared image temperature value |
| CN117575165A (en) * | 2023-12-05 | 2024-02-20 | 浙江万胜智通科技有限公司 | Intelligent patrol management method and system for digital power distribution network |
| CN117575165B (en) * | 2023-12-05 | 2024-05-07 | 浙江万胜智通科技有限公司 | Intelligent patrol management method and system for digital power distribution network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108537154B (en) | Power transmission line bird nest identification method based on HOG characteristics and machine learning | |
| CN108037133B (en) | Intelligent electric power equipment defect identification method and system based on unmanned aerial vehicle inspection image | |
| CN114694050A (en) | Power equipment running state detection method based on infrared image | |
| CN108280855A (en) | A kind of insulator breakdown detection method based on Fast R-CNN | |
| CN108680833B (en) | Composite insulator defect detection system based on unmanned aerial vehicle | |
| CN105654461A (en) | Machine vision detection method for fracture of multi-bundle conductor spacer | |
| CN112001260A (en) | Cable trench fault detection method based on infrared and visible light image fusion | |
| CN109740507A (en) | A kind of method of transformer equipment abnormality sensing | |
| CN112528979B (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN111754737A (en) | Online identification and evaluation device and method for installation acceptance of metering device | |
| CN111639568A (en) | Navigation patrol method for intelligent robot of transformer substation | |
| CN112508911A (en) | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof | |
| CN115690012A (en) | Detection method for wrong connection line of electric energy meter | |
| CN111507189A (en) | Insulator string defect rapid detection method based on image processing technology | |
| CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
| CN108898187A (en) | A kind of method and device of automatic identification power distribution room indicating equipment image | |
| CN112818806A (en) | Transformer substation inspection robot auxiliary navigation method based on deep learning | |
| CN117197700A (en) | Intelligent unmanned inspection contact net defect identification system | |
| CN111968078A (en) | Appearance detection method, device, equipment and storage medium for power transformation equipment | |
| CN111428987A (en) | Artificial intelligence-based image identification method and system for relay protection device | |
| CN114330477A (en) | Power equipment defect detection system and method based on mixed reality equipment | |
| CN110688919A (en) | Intelligent line patrol fault identification method | |
| CN111652055A (en) | Intelligent switch instrument identification method based on two-stage positioning | |
| Rongrong et al. | Fault Diagnosis Method of Distribution Equipment Based on Hybrid Model of Robot and Deep Learning | |
| Jalil et al. | Multimodal image analysis for power line inspection |
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 |