CN114283330B - Online inspection identification method and system based on multi-source data - Google Patents
Online inspection identification method and system based on multi-source data Download PDFInfo
- Publication number
- CN114283330B CN114283330B CN202111358855.8A CN202111358855A CN114283330B CN 114283330 B CN114283330 B CN 114283330B CN 202111358855 A CN202111358855 A CN 202111358855A CN 114283330 B CN114283330 B CN 114283330B
- Authority
- CN
- China
- Prior art keywords
- health
- primary equipment
- recognition
- visible light
- abnormal
- 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
- 238000007689 inspection Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000036541 health Effects 0.000 claims abstract description 209
- 238000005259 measurement Methods 0.000 claims abstract description 31
- 230000002159 abnormal effect Effects 0.000 claims description 67
- 230000008447 perception Effects 0.000 claims description 16
- 230000003862 health status Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 abstract description 6
- 230000005856 abnormality Effects 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 description 20
- 238000004458 analytical method Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 12
- 238000010438 heat treatment Methods 0.000 description 11
- 239000012212 insulator Substances 0.000 description 11
- 230000007547 defect Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 10
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 8
- 239000003990 capacitor Substances 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000012790 confirmation Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 239000013589 supplement Substances 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 239000002184 metal Substances 0.000 description 3
- 229910052751 metal Inorganic materials 0.000 description 3
- 229910052573 porcelain Inorganic materials 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000001931 thermography Methods 0.000 description 3
- 238000004587 chromatography analysis Methods 0.000 description 2
- 229910052742 iron Inorganic materials 0.000 description 2
- WABPQHHGFIMREM-UHFFFAOYSA-N lead(0) Chemical compound [Pb] WABPQHHGFIMREM-UHFFFAOYSA-N 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000002788 crimping Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000004457 water analysis Methods 0.000 description 1
Landscapes
- Radiation Pyrometers (AREA)
Abstract
The invention discloses an online inspection identification method based on multi-source data, which comprises the following steps: training the recognition model by adopting primary equipment visible light pictures under different weather scenes; performing health recognition on the visible light pictures of the primary equipment acquired on site through the trained recognition model; health identification is carried out on the primary equipment through the collected infrared images of the primary equipment; health identification is carried out on the acquired voice prints of the primary equipment; carrying out health identification on primary equipment through sensor data of a sensing layer; health identification is carried out on primary equipment protection measurement data by collecting the primary equipment protection measurement data; judging the health condition of the primary equipment by carrying out health identification results on the acquired visible light pictures of the primary equipment under different weather scenes; the health condition of the primary equipment is judged by combining the result of the health identification of the visible light picture of the primary equipment with various data source data, and the judgment accuracy of the health abnormality of the primary equipment can be improved.
Description
Technical Field
The invention belongs to the technical field of on-line monitoring of power equipment, and particularly relates to an on-line inspection identification method and system based on multi-source data.
Background
The online inspection is used as an important technical means for supporting operation and maintenance of operation and inspection service, and the reasons include as follows:
1) Is suitable for the development requirement of an unattended transformer substation
With the continuous improvement of the production automation level of the power system, unmanned substations are generated, and at present, a remote viewing function is added to a plurality of substations on the basis of the functions of remote measurement, remote signaling, remote control and remote adjustment, so that the safety and reliability of operation of the unmanned substations are greatly improved by realizing the limitation of fire disaster, personnel intrusion and false intrusion or dangerous areas and introducing video images into the remote monitoring of the substations.
2) Manual omission of transformer substation on duty is avoided
The traditional transformer substation monitoring system requires the staff to watch on the screen all the time to monitor scene activities, which increases the workload of the monitoring staff and is easy to generate fatigue state, meanwhile, the monitoring staff inevitably has limited focusing time or is interfered by other things outside, and useful information is easy to miss. In addition, the subjectivity exists in the manual participation judgment, and the further development of the automation degree of the power system is seriously hindered.
3) Development of image video technology
The technologies such as image recognition and artificial intelligence are widely applied to public security, traffic and the like, but the video monitoring value in the power grid operation and inspection service is only expressed in terms of meeting manual remote monitoring, and the service requirement of automatically finding equipment/environment/personnel abnormality through image recognition and further realizing automatic inspection cannot be met.
The appearance and application of the image recognition function in the intelligent video monitoring system enable the intelligent video monitoring system to be more intelligent, and the image recognition function is also an important sign of development of the video monitoring system. The implementation and selection of the image recognition algorithm are key points in the intelligent video monitoring technology. The image recognition technology is a technology for processing, analyzing and understanding images by using a computer to recognize targets and objects in various different modes, and along with the mature development of the image recognition technology and the urgent need of intelligent construction of a power grid, the application of the image recognition technology to the analysis and recognition of monitoring images of a transformer substation has become an urgent need.
Various methods for improving the accuracy of online inspection through images and videos exist, for example, an intelligent diagnosis technology of the heating defects of the power transmission line based on unmanned aerial vehicle infrared videos is provided, a main direction-based power transmission line heating defect recognition algorithm is provided, and the main direction of the power transmission line is extracted by calculating the inter-frame difference of the infrared video images, so that the fault position of the line is determined. The system for remote monitoring, detection and automatic alarm of the substation equipment based on the image recognition technology, which is used for eliminating the interference influence of high voltage and strong light direct irradiation on the image to improve the accuracy of image recognition, is applied to wavelet analysis and a median filtering algorithm.
In summary, the existing substation monitoring image recognition methods all adopt a method of performing complex preprocessing on an image, manually designing features, calculating manual features and machine learning, but the recognition rate of part of equipment under part of weather conditions is low due to the problems of substation monitoring image training samples, modeling problems of weather and meteorological influences, feature extraction of different equipment and the like.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an online inspection identification method based on multi-source data, which can improve the abnormality judgment accuracy of primary equipment.
The technical problems to be solved by the invention are realized by the following technical scheme:
in a first aspect, an online inspection identification method based on multi-source data is provided, including:
aiming at primary equipment to be detected, performing visible light health recognition on a visible light picture of the primary equipment through a trained recognition model to obtain a visible light recognition result; respectively obtaining the infrared image health recognition result of the primary equipment,
Voiceprint health recognition result,
Sensing sensor data health recognition result,
Protecting the health recognition result of the measured data;
judging the health condition of primary equipment according to the visible light health identification result;
or judging the health condition of the primary equipment according to the visible light health recognition result, the infrared image health recognition result, the voiceprint health recognition result, the perception sensor data health recognition result and the protection measurement data health recognition result.
With reference to the first aspect, further, the determining, according to the visible light health recognition result, the health condition of the primary device includes one of the following:
when the correct recognition rate of the abnormal health state occurrence in the visible light health recognition result of the primary equipment in the specific weather scene is higher than a first threshold value, determining that the abnormal health state occurrence of the primary equipment;
the correct recognition rate of the abnormal health state occurrence in at least two weather scenes is higher than a second threshold value, and the abnormal health state occurrence of the primary equipment is determined;
and if the correct recognition rates of the abnormal health states in at least three weather scenes are higher than a third threshold value, determining that the abnormal health states of the primary equipment occur.
With reference to the first aspect, further, determining the health condition of the primary device according to the visible light health recognition result, the infrared image health recognition result, the voiceprint health recognition result, the perception sensor data health recognition result, and the protection measurement data health recognition result includes:
when the visible light health recognition is carried out on the primary equipment to be detected, the correct recognition rate of the abnormal state of the primary equipment in a certain weather scene is higher than a second threshold value, and meanwhile, the abnormal state of the primary equipment is recognized in one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition, the abnormal state of the primary equipment is judged, and an abnormal alarm is given.
With reference to the first aspect, further includes: a multi-data source health dual-acknowledgement determination, comprising: and judging the health condition of the primary equipment as abnormal according to the visible light health recognition results of the primary equipment in different weather scenes, and judging that the health condition of the primary equipment is abnormal by one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition junction.
In a second aspect, an online inspection identification system based on multi-source data is provided, including:
health identification module: the method comprises the steps of performing visible light health recognition on visible light pictures of primary equipment to be detected through a trained recognition model to obtain visible light recognition results; and
respectively obtaining the infrared image health recognition result of the primary equipment,
Voiceprint health recognition result,
Sensing sensor data health recognition result
Protecting the health recognition result of the measured data;
one of a visible light health status determination module or a multiple data source health status determination module;
the visible light health condition determining module is used for judging the health condition of the primary equipment according to the visible light health identification result;
the multi-data source health condition determining module is used for judging the health condition of the primary equipment according to the visible light health recognition result, the infrared image health recognition result, the voiceprint health recognition result, the perception sensor data health recognition result and the protection measurement data health recognition result.
With reference to the second aspect, further, the operation performed by the visible light health status determining module includes one of the following:
when the correct recognition rate of the abnormal health state occurrence in the visible light health recognition result of the primary equipment in the specific weather scene is higher than a first threshold value, determining that the abnormal health state occurrence of the primary equipment;
the correct recognition rate of the abnormal health state occurrence in at least two weather scenes is higher than a second threshold value, and the abnormal health state occurrence of the primary equipment is determined;
and if the correct recognition rates of the abnormal health states in at least three weather scenes are higher than a third threshold value, determining that the abnormal health states of the primary equipment occur.
With reference to the second aspect, further operations performed by the multiple data source health determination module include: when the visible light health recognition is carried out on the primary equipment to be detected, the correct recognition rate of the abnormal state of the primary equipment in a certain weather scene is higher than a second threshold value, and meanwhile, the abnormal state of the primary equipment is recognized in one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition, the abnormal state of the primary equipment is judged, and an abnormal alarm is given.
In combination with the second aspect, the system further comprises a multi-data-source health condition dual-determination module, wherein the health condition of the primary equipment is judged to be abnormal according to the visible light health recognition results of the primary equipment under different weather scenes, and when the health condition of the primary equipment is judged to be abnormal through one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition, the health condition of the primary equipment is judged to be abnormal. .
The invention has the beneficial effects that: the invention can confirm the health condition of the primary equipment by visible light under various scenes based on multi-source data, and further can confirm the health condition of the monitoring multi-data source primary equipment by combining infrared temperature imaging, voiceprint information data, three/four-area equipment state sensing data, one/two-area protection measurement and the like, thereby improving the recognition rate of online inspection.
For some special cases, the invention can further confirm the health condition of the primary equipment by a double-confirmation mode of visible light identification and multiple data source identification, thereby further improving the inspection recognition rate.
The method is suitable for primary equipment health state inspection and identification of transformer substations of transmission and transformation power distribution networks and centralized control stations, primary equipment health state inspection and identification of traditional power generation booster stations such as thermal power stations, hydropower stations and nuclear power stations, and primary equipment health state inspection and identification of new energy booster stations for wind power generation and photovoltaic power generation.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is one of the logic diagrams of the anomaly results of the image recognition device of the electrical primary device of the present invention;
FIG. 3 is a second logic diagram of the anomaly result of the image recognition device of the electrical primary device according to the present invention;
FIG. 4 is a logic diagram of the abnormal result of the image recognition device of the electrical primary device according to the present invention;
FIG. 5 is a logic diagram of an electrical primary device image criterion complementation decision device anomaly in the present invention;
FIG. 6 is a logic diagram of an electrical primary device image and other data source complementarity determining device anomaly in the present invention;
fig. 7 is a logic diagram of an electrical primary device double-confirmation judging device abnormality in the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order to better understand the present invention, the following describes related technologies in the technical solution of the present invention.
As shown in fig. 1-7, the online inspection identification method based on multi-source data of the invention comprises the following steps:
step 1, training an identification model by adopting primary equipment visible light pictures in different weather scenes
Modeling is carried out on various electric primary equipment according to different weather scenes such as sunny days, overcast days, rainy days, snowy days, foggy days, sand storm, hail and the like. To achieve a more accurate recognition effect, the model can be built up to specific parts of the primary device.
For example, circuit breaker equipment components modeled for sunny conditions include body appearance, oil level gauges, bushing current transformers, pressure gauges, etc., modeled separately from the different components of the primary equipment.
The modeled disconnector device components include contacts, electrical arms, position indicators, insulator appearance, leads, joints, etc., modeled separately from the different components of the primary device.
The modeled current transformer equipment parts comprise a body appearance, a tail screen, a pressure gauge, an oil level gauge, a lead wire, a connector and the like, and are modeled according to different parts of primary equipment.
The modeled voltage transformer equipment parts comprise a body appearance, a tail screen, a pressure gauge, an oil level gauge, a lead wire, a connector and the like, and are modeled according to different parts of primary equipment.
The modeled lightning arrester equipment parts comprise a grading ring, a body appearance, a grounding down-lead, a lightning arrester meter, a lead, a connector and the like, and are respectively modeled according to different parts of primary equipment.
The modeled parallel capacitor bank equipment components comprise capacitor appearance, reactor appearance, insulator appearance, discharge coil appearance and the like, and are respectively modeled according to different components of the primary equipment.
The modeled bus bar components comprise bus bar appearance, insulator appearance, hardware appearance, wire clamps, joints and the like, and are modeled according to different components of the primary equipment.
The abnormal characteristics of different devices of the listed transformer substation are different, training is performed according to abnormal picture information of different devices such as a transformer, a circuit breaker, an isolating switch, a lightning arrester, a current transformer, a voltage transformer, a parallel capacitor bank, a bus and the like, the abnormal characteristics of various components of various devices are respectively extracted, and targeted model training is realized. The training of other weather scenarios is the same.
Step 2, carrying out health recognition on the visible light pictures of the primary equipment acquired on site through a trained recognition model;
and acquiring high-definition image data of the electric primary equipment in the transformer substation through the visible light cameras of each point in the transformer substation and the cameras on the inspection robot. And (3) identifying the health condition of the acquired high-definition image of the electric primary equipment through the trained model in the step (1).
Step 3, recognizing the health condition of the electric primary equipment through data of other various data sources except the visible light picture;
comprising the following steps: the infrared thermal imaging is adopted to diagnose the primary equipment with the characteristic of surface temperature subsection caused by current, voltage heating effect or other heating effects, the infrared image information is used as the effective supplement of visible light image identification, and the health status identification of the primary equipment is jointly completed.
The defect properties of the current-induced thermal type device are classified into an emergency defect, a serious defect and a general defect.
And connecting the electrical equipment with the metal part, and taking the wire clamp and the joint as a thermal image.
The connection of the metal parts takes the thermal image with the wire clip and the joint as the center.
A metal wire, a thermal image centered on the wire.
A swivel of the isolating switch, a thermal image taking the swivel as a center; the thermal image of the knife edge of the isolating switch with the knife edge crimping spring as the center.
The dynamic and static contacts of the circuit breaker take a thermal image with a top cap and a lower flange as centers, and the temperature of the top cap is higher than that of the lower flange; and a thermal image taking a lower flange and a top cap as centers of the middle contact of the circuit breaker, wherein the temperature of the lower flange is higher than that of the top cap.
The current transformer is characterized by taking a serial-parallel wire outlet head or a large screw wire outlet clamp as a thermal image with the highest temperature or by heating a top iron anchor.
The sleeve uses the top column cap of the sleeve as the hottest thermal image.
A capacitor fuse, which takes the middle part of the fuse near the capacitor side as the hottest thermal image; and the fuse seat of the capacitor takes the fuse seat as the hottest thermal image.
A transformer tank characterized by overheating of a local surface of the tank.
The iron cores of the dry-type transformer, the grounding transformer, the series reactor and the shunt reactor are characterized in that the local surface of the iron core is overheated.
The windings of the dry-type transformer, the grounding transformer, the series reactor and the shunt reactor are characterized in that the surface of the windings is locally overheated or the positions of the outgoing terminals are overheated.
The voltage-induced thermal device defect properties are classified into two types, namely, serious defects and general defects.
The voltage transformer heats up integrally with the body as the center.
The coupling capacitor is in high overall temperature rise or local overheat, and the heating accords with the gradual decreasing rule from top to bottom.
The thermal imaging of the mobile capacitor generally takes the upper part of the body as the center, the highest temperature of the normal thermal imaging is generally about two-thirds of the height of the vertical bisector of the wide surface, the surface temperature rise is slightly high, and the whole heating or the local heating is realized.
The high-voltage sleeve, the thermal image characteristic is presented by the integral heating thermal image of the sleeve; the thermal image shows a local heating region fault for the corresponding part.
The thermal image of the oil-filled sleeve is characterized in that the oil surface is the thermal image with the highest temperature, and the oil surface has an obvious horizontal dividing line.
The porcelain insulator has the temperature distribution rule of the normal insulator string and the voltage distribution rule, namely, the porcelain insulator presents a saddle type called by army, the temperature difference of adjacent insulators is small, and the porcelain insulator takes an iron cap as a thermal image of a heating center, which is higher than the temperature of the normal insulator.
The composite insulator is locally overheated at the junction of good insulator and insulation deterioration, and the overheated part moves with the extension of time.
After the infrared camera or the robot makes an inspection tour according to a specified route, the infrared temperature measuring map inspection tour data with the physical identity mark is uploaded to an inspection tour host, and defect identification is completed on the inspection tour host.
And collecting and analyzing voiceprint information data of the electric primary equipment, wherein the voiceprint information is used as an effective supplement for visible light image identification, and the health state identification of the primary equipment is jointly completed.
Voiceprint information acquisition is commonly used currently, and signal analysis of corresponding frequencies is extracted through an ultrahigh frequency signal sensor and an ultrasonic signal sensor, so that a judgment conclusion is given.
The characteristic quantity recognition method of the voiceprint device is characterized in that an ultrahigh frequency signal sensor is arranged to receive ultrahigh frequency electromagnetic waves radiated in the partial discharge process, electromagnetic wave signals generated by partial discharge are extracted in the ultrahigh frequency range of 300-3000 MHz, and the on-line monitoring of the partial discharge of primary equipment is realized.
According to the characteristic quantity recognition method of the voiceprint device, an ultrasonic signal sensor is arranged to receive ultrasonic waves generated by partial discharge of primary equipment (such as the inside of a transformer), signals are extracted within an ultrasonic range of 70-150 kHz, magnetic noise of an iron core and mechanical vibration noise of the transformer are avoided, and online monitoring of the size and the position of the partial discharge of the primary equipment is achieved.
And collecting and analyzing data of a sensing layer sensor (generally in three/four areas), wherein the sensing layer sensor data is used as an effective supplement for visible light image identification, and the health state identification of primary equipment is jointly completed.
Oil chromatography, oil sampling of transformer, etc. for oil chromatography, including the following other H 2 、CH 4 、C 2 H 6 、C 2 H 4 、C 2 H 2
CO、CO 2 And compared with relevant standards and trend analyzed.
And (3) carrying out micro-water analysis on oil samples taken by equipment such as a transformer and the like, comparing the oil samples with relevant standards, and carrying out trend analysis.
Partial discharge analysis, namely analyzing the partial discharge of primary equipment based on an ultrahigh frequency partial discharge detection method, a high frequency pulse current method, a chemical detection method, a pulse current method and the like, comparing the partial discharge with relevant standards, and carrying out trend analysis.
And (3) temperature analysis, namely arranging various temperature sensors to measure the temperature of each part or part of the primary equipment, comparing the temperature with related standards and carrying out trend analysis.
And collecting and analyzing monitoring data such as one/two area protection measurement and the like, wherein the monitoring data such as the one/two area protection measurement and the like is used as effective supplement for visible light image identification, and the health state identification of primary equipment is jointly completed.
The grounding current of the iron core is monitored based on the protection measurement and control device, and compared with relevant standards and subjected to trend analysis.
And (3) analyzing electric quantity data, acquiring information such as current, voltage, power, frequency and the like of primary equipment based on a protection or measurement and control device, performing heating analysis through a current load/load, performing fault analysis based on negative sequence and zero sequence current and the like.
And (3) analyzing the non-electric quantity data, analyzing non-electric quantity signals such as heavy gas, light gas, high oil temperature, abnormal oil level, non-energy storage of a spring, abnormal position of a handcart and the like based on a non-electric quantity protection or measurement and control device, comprehensively judging event signal analysis of the protection or measurement and control device, assisting in visible light image recognition, and further improving the accuracy of equipment health state recognition.
The health condition determination of the electrical primary device includes the following cases:
1) Judging the health condition of the primary equipment through health identification results of the collected visible light pictures of the primary equipment in different weather scenes
For general conditions, the health condition of the electrical primary equipment can be judged by the results of the health identification of the visible light images under different weather scenes, specifically:
when the correct recognition rate of the abnormal state of the certain electric primary equipment in a certain scene is higher than 95% (a first threshold) (shown in fig. 2), or the recognition rate of two scenes is simultaneously higher than 90% (a second threshold) (shown in fig. 3), or the recognition rate of three or more scenes is simultaneously higher than 85% (a third threshold) (shown in fig. 4), the abnormal state of the primary equipment is judged, and an abnormal alarm is given.
2) The health condition of the primary equipment is judged through complementation between the visible light picture and other data sources of the non-visible light picture of the primary equipment, and specifically:
when the health recognition is carried out through the visible light picture of the primary equipment, the correct recognition rate of the abnormal health state of the primary equipment in a certain scene is higher than 90%, and meanwhile, the abnormal health state of the primary equipment is recognized through one mode of the infrared image, the voiceprint, the sensor data of the sensing layer and the protection measurement data of the primary equipment, the abnormal health state of the primary equipment is judged, and an abnormal alarm is given (shown in fig. 6).
3) Multi-source data dual-acknowledgement identification
In order to reduce the inspection cost, the situation that the inspection personnel run for one time due to abnormal misjudgment is prevented, when the health condition of a certain piece of equipment is judged to be abnormal (first layer confirmation) through the health recognition result of the collected visible light pictures of the primary equipment under different weather scenes, namely, the correct recognition rate of the abnormal health condition of the certain electrical primary equipment in the visible light picture recognition certain scene in the step 2 is higher than 95%, or the recognition rate of two kinds of scenes is simultaneously higher than 90%, or the recognition rate of three kinds of scenes and more than 85% (fig. 5), and when the health condition of the primary equipment is judged to be abnormal (second layer confirmation) through one mode of the infrared image, voiceprint, the sensor data of the sensing layer and the protection measurement data of the primary equipment, the health condition of the primary equipment is judged to be abnormal (shown in fig. 7).
The invention provides an online inspection recognition system based on multi-source data, which is used for loading the online inspection recognition method based on the multi-source data, and comprises the following steps:
health identification module: the method comprises the steps of carrying out visible light health recognition on a collected visible light picture of primary equipment aiming at the primary equipment to be detected through a trained recognition model;
the method comprises the steps of performing infrared image health identification on a primary device through an infrared image of the primary device;
performing voiceprint health identification on the primary equipment through the voiceprint of the primary equipment;
performing sensing sensor data health identification on the primary equipment through sensing sensor data of the primary equipment;
performing protection measurement data health identification on the primary equipment through the protection measurement data of the primary equipment;
the visible light health condition determining module is used for judging the health condition of the primary equipment according to the visible light health identification results of the primary equipment in different weather scenes;
the multi-data source health condition determining module is used for judging the health condition of the primary equipment according to the primary equipment visible light health recognition result and the infrared image health recognition, voiceprint health recognition, perception sensor data health recognition and protection measurement data health recognition result thereof.
The multi-data source health condition double-determination module is used for judging that the health condition of one piece of equipment is abnormal when the health condition of the one piece of equipment is judged to be abnormal through health identification results of collected visible light pictures of the one piece of equipment under different weather scenes and judging that the health condition of the one piece of equipment is abnormal through one mode of infrared images, voiceprints, sensor data of a sensing layer and protection measurement data of the one piece of equipment.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (2)
1. The online inspection identification method based on the multi-source data is characterized by comprising the following steps of:
aiming at primary equipment to be detected, carrying out visible light health recognition on visible light pictures of the primary equipment through a trained recognition model to obtain a visible light health recognition result; and
respectively obtaining an infrared image health recognition result, a voiceprint health recognition result, a perception sensor data health recognition result and a protection measurement data health recognition result of the primary equipment;
judging the health condition of primary equipment according to the visible light health identification result;
or judging the health condition of the primary equipment according to the visible light health recognition result, the infrared image health recognition result, the voiceprint health recognition result, the perception sensor data health recognition result and the protection measurement data health recognition result, including: when the visible light health recognition is carried out on the primary equipment to be detected, the correct recognition rate of the abnormal state of the primary equipment in a certain weather scene is higher than a second threshold value, and meanwhile, the abnormal state of the primary equipment is recognized in one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition, the abnormal state of the primary equipment is judged, and an abnormal alarm is given; further comprises: a multi-data source health dual-acknowledgement determination, comprising: judging the health condition of the primary equipment as abnormal according to the visible light health recognition results of the primary equipment in different weather scenes, and judging that the health condition of the primary equipment is abnormal by one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition junction;
the judging of the health condition of the primary equipment according to the visible light health identification result comprises one of the following steps:
the correct recognition rate of the abnormal health state occurrence in at least two weather scenes is higher than a second threshold value, and the abnormal health state occurrence of the primary equipment is determined;
and if the correct recognition rates of the abnormal health states in at least three weather scenes are higher than a third threshold value, determining that the abnormal health states of the primary equipment occur.
2. An online inspection and recognition system based on multi-source data is characterized by comprising:
health identification module: the method comprises the steps of carrying out visible light health recognition on visible light pictures of primary equipment to be detected through a trained recognition model aiming at the primary equipment to be detected, and obtaining a visible light health recognition result; and
respectively obtaining an infrared image health recognition result, a voiceprint health recognition result, a perception sensor data health recognition result and a protection measurement data health recognition result of the primary equipment;
one of a visible light health status determination module or a multiple data source health status determination module;
the visible light health condition determining module is used for judging the health condition of the primary equipment according to the visible light health identification result, and the operation executed by the visible light health condition determining module comprises one of the following steps: the correct recognition rate of the abnormal health state occurrence in at least two weather scenes is higher than a second threshold value, and the abnormal health state occurrence of the primary equipment is determined;
the correct recognition rate of the abnormal health state occurrence in at least three weather scenes is higher than a third threshold value, and the abnormal health state occurrence of the primary equipment is determined;
the multi-data source health condition determining module is used for judging the health condition of the primary equipment according to the visible light health recognition result, the infrared image health recognition result, the voiceprint health recognition result, the perception sensor data health recognition result and the protection measurement data health recognition result;
the operations performed by the multiple data source health determination module include: when the visible light health recognition is carried out on the primary equipment to be detected, the correct recognition rate of the abnormal state of the primary equipment in a certain weather scene is higher than a second threshold value, and meanwhile, the abnormal state of the primary equipment is recognized in one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition, the abnormal state of the primary equipment is judged, and an abnormal alarm is given;
the multi-data source health condition determining module further comprises a multi-data source health condition double determining module, the health condition of the primary equipment is judged to be abnormal according to the visible light health recognition results of the primary equipment under different weather scenes, and when the health condition of the primary equipment is judged to be abnormal through one mode of infrared image health recognition, voiceprint health recognition, perception sensor data health recognition or protection measurement data health recognition, the health condition of the primary equipment is judged to be abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111358855.8A CN114283330B (en) | 2021-11-17 | 2021-11-17 | Online inspection identification method and system based on multi-source data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111358855.8A CN114283330B (en) | 2021-11-17 | 2021-11-17 | Online inspection identification method and system based on multi-source data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114283330A CN114283330A (en) | 2022-04-05 |
| CN114283330B true CN114283330B (en) | 2024-03-26 |
Family
ID=80869509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111358855.8A Active CN114283330B (en) | 2021-11-17 | 2021-11-17 | Online inspection identification method and system based on multi-source data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114283330B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117037851B (en) * | 2023-08-31 | 2024-03-19 | 江苏南工科技集团有限公司 | Fire sensing and detecting method based on voiceprint analysis technology |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104253482A (en) * | 2014-08-08 | 2014-12-31 | 济南大学 | Image data base and inspection robot-based equipment trouble detection method |
| CN112381784A (en) * | 2020-11-12 | 2021-02-19 | 国网浙江省电力有限公司信息通信分公司 | Equipment detecting system based on multispectral image |
| CN112466000A (en) * | 2020-09-23 | 2021-03-09 | 国网上海市电力公司 | Inspection system based on power inspection robot and inspection control method |
| CN113344026A (en) * | 2021-04-29 | 2021-09-03 | 国网浙江省电力有限公司嘉兴供电公司 | Transformer substation equipment abnormity identification and positioning method based on multivariate fusion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10542211B2 (en) * | 2017-10-05 | 2020-01-21 | GM Global Technology Operations LLC | Camera subsystem evaluation using sensor report integration |
-
2021
- 2021-11-17 CN CN202111358855.8A patent/CN114283330B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104253482A (en) * | 2014-08-08 | 2014-12-31 | 济南大学 | Image data base and inspection robot-based equipment trouble detection method |
| CN112466000A (en) * | 2020-09-23 | 2021-03-09 | 国网上海市电力公司 | Inspection system based on power inspection robot and inspection control method |
| CN112381784A (en) * | 2020-11-12 | 2021-02-19 | 国网浙江省电力有限公司信息通信分公司 | Equipment detecting system based on multispectral image |
| CN113344026A (en) * | 2021-04-29 | 2021-09-03 | 国网浙江省电力有限公司嘉兴供电公司 | Transformer substation equipment abnormity identification and positioning method based on multivariate fusion |
Non-Patent Citations (1)
| Title |
|---|
| 电气设备红外成像图库与远程辅助诊断***研究;陈锋;周业如;;华东电力(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114283330A (en) | 2022-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110018389B (en) | Online fault monitoring method and system for power transmission line | |
| CN103499774B (en) | Transmission of electricity insulator arc-over in-service monitoring safety early warning device | |
| CN103926510B (en) | A kind of cable sheath electric current and current-carrying capacity on-line monitoring and fault diagonosing localization method | |
| US8400504B2 (en) | Contamination monitoring of high voltage insulators | |
| CN104977087B (en) | A kind of method of the electrical equipment fault automatic early-warning based on infrared imaging temperature measuring | |
| CN107767374B (en) | Intelligent diagnosis method for local overheating of inner conductor of GIS basin-type insulator | |
| CN105447471A (en) | Infrared detection based device gas leakage identification method and apparatus | |
| CN102520286B (en) | Hyperspectrum-based composite insulator operation state classification method | |
| CN103954897A (en) | Intelligent power grid high-voltage insulation damage monitoring system and method based on ultraviolet imaging | |
| CN104990629A (en) | Infrared-imaging-and-temperature-measurement-based automatic early warning system for power equipment fault | |
| CN114460411A (en) | Fault positioning method, device and system for power transmission network, processor and electronic equipment | |
| CN114283330B (en) | Online inspection identification method and system based on multi-source data | |
| CN109342882A (en) | Cable fault on-line monitoring method based on multi-source Eigenvalues analysis | |
| CN110672980A (en) | Electric power inspection on-line monitoring system based on ultraviolet, infrared and visible imaging | |
| CN114088212A (en) | Diagnosis method and diagnosis device based on temperature vision | |
| CN117239926A (en) | Power distribution area monitoring system and method based on multi-parameter sensing | |
| Zhang et al. | Intelligent detection technology of infrared image of substation equipment based on deep learning algorithm | |
| CN117612345A (en) | Power equipment state monitoring and alarming system and method | |
| CN116169778A (en) | Processing method and system based on power distribution network anomaly analysis | |
| CN106355187A (en) | Application of visual information to electrical equipment monitoring | |
| CN108828398A (en) | A kind of single-core cable partial discharge location method based on structural similarity algorithm | |
| EP3706270B1 (en) | Artificial intelligence monitoring system using infrared images to identify hotspots in a switchgear | |
| CN115731455A (en) | Looped network metering cabinet | |
| CN104849654A (en) | Method for online monitoring breaker | |
| CN115459449A (en) | Transformer substation monitoring method and system capable of analyzing operation performance of transformer in real time |
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 |