Nuclear power intelligent inspection system based on industrial internet technology

Through the intelligent inspection system using industrial Internet technology in nuclear power plants, and the use of intelligent hardware and image recognition algorithms to realize automated data acquisition and analysis, the intelligent needs of nuclear power plant inspection business are solved, and the inspection efficiency and management quality are improved. Solve the shortcomings of the existing inspection methods.

CN119094556APending Publication Date: 2024-12-06CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411168245.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Nuclear power plant inspection business lacks intelligent means. The existing inspection methods include incomplete inspection scope, personnel mistakes, large manpower investment, difficulty in management and control, unauthorized evidence analysis, serious information islands, low degree of informationization, and narrow application scope The problem.

Method used

Adopt the intelligent nuclear power inspection system based on industrial Internet technology, and automatically collects data by installing intelligent hardware devices on digital site, transmits data to the industrial Internet platform using image recognition algorithms and transmission networks, and analyzes and compares them to realize intelligent inspection applications. Automatic inspection of the system.

Benefits of technology

Effectively reduce the workload of inspection personnel, improve the quality and efficiency of inspection management, realize comprehensive automated inspection of nuclear power plant scenarios, reduce human error, and improve the accuracy and efficiency of data collection and analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119094556A_ABST
    Figure CN119094556A_ABST
Patent Text Reader

Abstract

The invention discloses a nuclear power intelligent inspection system based on an industrial internet technology. The nuclear power intelligent inspection system comprises a digital field, a transmission network, an industrial internet platform, intelligent inspection application and the like. Wherein the field hardware comprises a track robot, a wheeled robot, an intelligent meter cover, a bolt camera and a macro camera which are adaptive to a switching station plant, an electrical plant, a steam turbine plant, a nuclear auxiliary plant and a peripheral plant, and a pan-tilt camera and sensors which are adaptive to security monitoring and equipment diagnosis scenes; the transmission network comprises Ethernet, WiFi, Lora and the like; the industrial internet platform comprises identification algorithms such as meter reading, state monitoring, appearance defect, environment monitoring and the like, and application functions such as algorithm configuration, service, application and the like; the inspection application has the functions of inspection configuration, task triggering, process execution, result analysis, report management and the like. The system and the method have the beneficial effects that the inspection work of each plant of the nuclear power plant is completed by integrating the intelligent equipment adaptive to the inspection of each scene.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of nuclear power intelligent management and control, and relates to a nuclear power intelligent inspection system based on industrial Internet technology. Background Art

[0002] In the power generation industry, thermal power plants have used photoresistor sensors to complete the collection and inspection of indicator light status data, and used slide-rail inspection robots and cameras to complete the inspection of coal handling trestles, GIS rooms, booster stations and other scenes; hydropower plants have used vibration sensors, equipment heating sensors, oil chromatographs and other terminal online equipment to complete the status monitoring of generator equipment; photovoltaic power plants have used logical reasoning platforms to complete the collection, identification and analysis of power plant personnel wear and personnel behavior data, and used UWB high-precision personnel positioning combined with personnel vital signs monitoring sensors to complete the management of personnel movement trajectories and electronic fences, as well as the monitoring of personnel health conditions such as heart rate and blood pressure; wind farms have used wheeled inspection robots to complete the inspection of the status of cabinets in distribution rooms and relay protection rooms.

[0003] For nuclear power plant inspection business, all nuclear power bases currently use the manual "paper record + handheld inspection instrument" method. The data perception method mainly relies on seeing, listening, and touching. The data entry method mainly relies on manual entry. The operation and maintenance personnel perform inspection tasks regularly according to the arrangement. At present, there is no intelligent means to meet the special inspection needs of complex production occasions such as nuclear power plants. In addition, the existing conventional inspection methods and information systems have problems such as incomplete inspection scope, personnel errors, large manpower investment, difficult control, inability to obtain evidence and analyze abnormal situations, serious information islands, low degree of informatization, and narrow application scope.

[0004] In view of the above shortcomings, the intelligent digital management of nuclear power plants is realized by establishing "digital site, software support platform, and application system", which effectively reduces the workload of power plant inspection personnel, improves work efficiency, and achieves a dual improvement in the quality of power plant inspection management and work efficiency. Summary of the invention

[0005] Based on the above problems, the patent of this invention provides a nuclear power intelligent inspection system based on industrial Internet technology, which features include: digital field hardware, transmission network, industrial Internet platform, intelligent inspection application and other modules. The original data is automatically collected by various intelligent inspection hardware devices installed in the digital field, transmitted to the industrial Internet platform through the network, and analyzed by image recognition algorithm to form recognized text data, which is then presented through the intelligent inspection application system after comparison with the inspection standard, replacing manual intelligent inspection work.

[0006] Furthermore, the digital field hardware includes a rail robot adapted to the switch station plant inspection business scenario, a wheeled robot adapted to the electrical plant inspection business scenario, a smart meter cover adapted to the turbine plant inspection business scenario, a gun camera adapted to the nuclear auxiliary plant inspection business scenario, a macro camera adapted to the peripheral plant inspection business scenario, as well as a pan-tilt camera adapted to the security monitoring business scenario and various sensors adapted to the main equipment diagnosis business scenario.

[0007] Preferably, the track robot includes a robot track, a robot body, a robot background software, a robot data acquisition device, and a robot power supply and network supply module, which can replace manual labor to complete the collection of inspection image data for scenes such as 220kv switch station buildings, 500kv switch station buildings, and 10kv switch station buildings.

[0008] Preferably, the wheeled robot includes a robot body, robot background software, robot data acquisition equipment, and robot power supply and networking modules, which can replace manual work to complete the collection of inspection image data for scenes such as instrumentation and control equipment areas, power equipment areas, and battery areas.

[0009] Preferably, the smart meter cover is image data acquisition hardware fixedly installed on the meter, and has built-in fill light function and fuzzy image sharpening processing function, which can replace manual completion of the collection of scene inspection image data of oil-filled pressure gauges, non-oil-filled pressure gauges, temperature meters, humidity meters, ammeters, voltmeters and other scenes in various scenes of the turbine plant.

[0010] Preferably, the gun camera is a special camera with explosion-proof, dust-proof and radiation-proof functions, and has a built-in fill light function, which can replace manual work to complete the collection of inspection image data of radioactive pipeline corridors, RRI water tanks, upper allowance pumps, sealed water-electricity exchangers and other scenes in nuclear auxiliary plants.

[0011] Preferably, the macro camera has a fill light function, which can replace manual work to complete the collection of inspection image data of rotor flow meters, oil level meters, liquid level meters, SEC pump pit meters and other scenes in peripheral factory buildings.

[0012] Preferably, the pan-tilt camera has preset position rotation, focusing, zooming, fill light and other functions, which can replace manual completion of the collection of image data of smoke, flames and abandoned items in scenes such as GA corridors, GB corridors, SEC pump pits and claustrophobic equipment areas in the security field.

[0013] Preferably, the sensors include main equipment such as main pumps, main transformers, steam turbine generators, temperature monitoring sensors for heating scenarios, vibration monitoring sensors for vibration scenarios, and voiceprint recognition sensors for abnormal sound scenarios.

[0014] Furthermore, the transmission network includes Ethernet, WiFi, Lora, Bluetooth, etc., which can realize the aggregation of data from various data collection terminals on the one hand, and the transmission of the aggregated data to the intelligent inspection application through the industrial Internet platform on the other hand.

[0015] Furthermore, the industrial Internet platform includes: meter reading image recognition algorithms, status monitoring image recognition algorithms, appearance defect image recognition algorithms, environment monitoring image recognition algorithms, etc., as well as algorithm application functions such as algorithm configuration, algorithm service, and algorithm application.

[0016] Preferably, the meter reading image recognition algorithm specifically includes: oil-filled pressure gauge recognition algorithm, non-oil-filled pressure gauge recognition algorithm, temperature meter recognition algorithm, humidity meter recognition algorithm, oil temperature meter recognition algorithm, oil window type oil level gauge recognition algorithm, oil level type oil level gauge recognition algorithm, liquid level height recognition algorithm, flow meter recognition algorithm, etc.

[0017] Preferably, the status monitoring image recognition algorithm specifically includes: a switch status recognition algorithm, an indicator light status recognition algorithm, a pressure plate status recognition algorithm, a cabinet door opening and closing status recognition algorithm, a circuit breaker status recognition algorithm, an isolating switch status recognition algorithm, and a grounding switch status recognition algorithm.

[0018] Preferably, the appearance defect image recognition algorithm specifically includes: oil stain recognition algorithm, rust recognition algorithm, paint peeling recognition algorithm, crack recognition algorithm, fissure recognition algorithm, burn mark recognition algorithm, dent recognition algorithm, bulge recognition algorithm, water accumulation recognition algorithm, etc.

[0019] Preferably, the environment monitoring image recognition algorithm specifically includes: a smoke recognition algorithm, a flame recognition algorithm, a liquid leakage recognition algorithm, a gas emission recognition algorithm, an object left behind recognition algorithm, an animal intrusion recognition algorithm, etc.

[0020] Preferably, the algorithm configuration functions include: newly added algorithm deployment configuration, algorithm combination scheduling configuration, algorithm call configuration, algorithm license authorization configuration and other functions.

[0021] Preferably, the algorithm service functions include: monitoring of hardware resources occupied by the algorithm, monitoring of the running status of the algorithm service, monitoring of the number of algorithm calls, as well as algorithm service activation function and algorithm service deactivation function, etc.

[0022] Preferably, the algorithm application functions include: hardware intelligent device registration, real-time image data access, offline image data import, as well as pre-recognition data reception, raw data recognition and analysis, and post-recognition data feedback.

[0023] Furthermore, the intelligent inspection application includes: basic inspection configuration, inspection task triggering, inspection execution, inspection result analysis, inspection report management, inspection statistical analysis and other functions.

[0024] Preferably, the basic inspection configuration includes functions such as inspection object configuration, inspection point configuration, point inspection standard configuration, inspection route configuration, inspection task configuration, and functions of writing, checking, reviewing and approving each configuration.

[0025] Preferably, the inspection task triggering includes: the inspection task system automatically issues the inspection task execution instruction and the business personnel manually issues the inspection task execution instruction operation.

[0026] Preferably, the inspection task execution includes: starting each intelligent inspection device in each scene, each intelligent inspection device in each scene collecting original data in turn, transmitting the original data to the industrial Internet platform, identifying the original data into text data, comparing and analyzing the text data with the inspection standards, generating inspection results and other functions.

[0027] Preferably, the inspection result analysis includes: over-limit and cross-boundary analysis of text data after original data recognition, trend curve analysis, dead point analysis, shaking analysis, etc.

[0028] Preferably, the inspection report management includes: inspection report generation, inspection result value correction, inspection report approval, inspection report archiving and other functions.

[0029] Preferably, the inspection statistical analysis includes: online status analysis of intelligent hardware devices, inspection task completion rate statistical analysis, alarm type statistical analysis, alarm level statistical analysis, etc.

[0030] The drawings constituting a part of this application are used to provide a further understanding of this application, so that other features, purposes and advantages of this application become more obvious. The schematic implementation drawings and descriptions of this application are used to explain this application and do not constitute an improper limitation on this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a structural schematic diagram of the patent of this invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only embodiments of a part of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0033] like Figure 1 As shown, a nuclear power intelligent inspection system based on industrial Internet technology includes digital field hardware, transmission network, industrial Internet platform, and intelligent inspection application. (1) The digital field hardware includes: 1-1 track robot, 1-2 wheeled robot, 1-3 smart cover, 1-4 gun camera, 1-5 macro camera, 1-6 pan-tilt camera, and 1-7 sensor. (2) The transmission networks include: 2-1 Ethernet, 2-2 WiFi, 2-3 Lora, and 2-4 Bluetooth. (3) The Industrial Internet Platform includes: 3-1 meter reading recognition algorithm, 3-2 status monitoring recognition algorithm, 3-3 appearance defect recognition algorithm, 3-4 environment monitoring recognition algorithm, as well as 3-5 algorithm application function, 3-5 algorithm service function, 3-6 algorithm configuration function and other algorithm service application functions. (4) The intelligent inspection application includes: inspection basic configuration function, inspection task triggering function, inspection task execution function, inspection result analysis function, inspection report management function, and inspection statistical analysis function.

[0034] Through the rail robot installed in the digital site, the inspection data of the 220kv switch station, 500kv switch station and 10kv switch station in the switch station plant can be collected; through the wheeled robot installed in the digital site, the inspection data of the instrumentation and control equipment area, power supply equipment area and battery area in the electrical plant can be collected; through the smart meter covers installed in the turbine plant, the inspection data of the oil-filled pressure gauge, non-oil-filled pressure gauge, temperature meter, humidity meter, voltmeter and ammeter can be collected; through the gun camera installed in the nuclear auxiliary plant, the inspection data of the oil-filled pressure gauge, non-oil-filled pressure gauge, temperature meter, humidity meter, voltmeter and ammeter can be collected; The inspection data of radioactive pipe gallery, RRI water tank, upper allowance pump, sealed water heat exchanger and other scenes are collected; through the macro cameras installed in the peripheral plant, the rotor flow meter, oil window type oil level gauge, oil level type oil level gauge, liquid level height, SEC pump pit packing leakage data are collected; through the installed pan-tilt camera, the smoke, flame, and leftover items in the GA corridor, GB corridor, SEC pump pit, and claustrophobic equipment area are collected; through the temperature sensor, vibration sensor, and sound sensor installed on the main pump, main transformer, and steam turbine generator, the main equipment status data is collected. The equipment collected by the intelligent terminals such as rail robots, wheeled robots, smart meter covers, gun cameras, macro cameras, pan-tilt cameras, sensors, etc. is transmitted through the Ethernet, WiFi, Lora, Bluetooth and other networks of the transmission network layer, and is gathered and transmitted to the industrial Internet algorithm platform. After identification and analysis, text data is generated, and compared with the inspection standards of each point, it is presented in the intelligent inspection application system.

Claims

1. A nuclear power intelligent inspection system based on industrial Internet technology, the characteristics of which include: Modules such as digital field hardware, transmission network, industrial Internet platform, and intelligent inspection application. The original data is automatically collected through various intelligent inspection hardware devices installed in the digital field, transmitted to the industrial Internet platform through the network, and analyzed through image recognition algorithms to form recognized text data. After being compared with the inspection standards, it is presented through the intelligent inspection application system, replacing manual intelligent inspection work.

2. According to claim 1, a nuclear power intelligent inspection system based on industrial Internet technology is characterized in that: The digital field hardware includes rail robots adapted to switch station plant inspection business scenarios, wheeled robots adapted to electrical plant inspection business scenarios, smart meter covers adapted to turbine plant inspection business scenarios, gun cameras adapted to nuclear auxiliary plant inspection business scenarios, macro cameras adapted to peripheral plant inspection business scenarios, pan-tilt cameras adapted to security monitoring business scenarios, and various sensors adapted to main equipment diagnosis business scenarios. S0: The track robot includes a robot track, a robot body, a robot background software, a robot data acquisition device, and a robot power supply and network supply module, which can replace manual work to complete the collection of inspection image data for scenes such as 220kv switch station plants, 500kv switch station plants, and 10kv switch station plants. S1: The wheeled robot includes a robot body, robot background software, robot data acquisition equipment, and robot power supply and networking modules, which can replace manual work to complete the collection of inspection image data for scenes such as instrumentation and control equipment areas, power equipment areas, and battery areas. S2: The smart meter cover is an image data acquisition hardware fixedly installed on the meter, and has built-in fill light function and fuzzy image sharpening processing function. It can replace manual completion of the collection of inspection image data of oil-filled pressure gauges, non-oil-filled pressure gauges, temperature meters, humidity meters, ammeters, voltmeters and other scenes in various scenes of the turbine plant. S3: The gun camera is a special camera with explosion-proof, dust-proof and radiation-proof functions. It has a built-in fill light function and can replace manual work to complete the collection of inspection image data for radioactive pipe corridors, RRI water tanks, upper allowance pumps, sealed water-electricity exchangers and other scenes in nuclear auxiliary plants. S4: The macro camera has a fill light function, which can replace manual work to complete the collection of inspection image data of rotor flow meters, oil level meters, liquid level meters, SEC pump pit meters and other scenes in the peripheral factory buildings. S5: The PTZ camera has preset position rotation, focus, zoom, fill light and other functions, which can replace manual work to complete the collection of image data of smoke, flame and abandoned items in scenes such as GA corridor, GB corridor, SEC pump pit and claustrophobic equipment area in the security field. S6: The sensors include those adapted to main equipment such as main pumps, main transformers, steam turbine generators, temperature monitoring sensors for heating scenarios, vibration monitoring sensors for vibration scenarios, and voiceprint recognition sensors for abnormal sound scenarios.

3. According to claim 1, a nuclear power intelligent inspection system based on industrial Internet technology is characterized in that: The transmission network includes Ethernet, WiFi, Lora, Bluetooth, etc., which can realize the aggregation of data from various data collection terminals on the one hand, and the transmission of the aggregated data to the intelligent inspection application through the industrial Internet platform on the other hand.

4. According to claim 1, a nuclear power intelligent inspection system based on industrial Internet technology is characterized in that: The industrial Internet platform includes: meter reading image recognition algorithms, status monitoring image recognition algorithms, appearance defect image recognition algorithms, environment monitoring image recognition algorithms, etc., as well as algorithm application functions such as algorithm configuration, algorithm service, and algorithm application. S0: The meter reading image recognition algorithm specifically includes: oil-filled pressure gauge recognition algorithm, non-oil-filled pressure gauge recognition algorithm, temperature meter recognition algorithm, humidity meter recognition algorithm, oil temperature meter recognition algorithm, oil window type oil level gauge recognition algorithm, oil level type oil level gauge recognition algorithm, liquid level height recognition algorithm, flow meter recognition algorithm, etc. S1: The state monitoring image recognition algorithm specifically includes: switch state recognition algorithm, indicator light state recognition algorithm, pressure plate state recognition algorithm, cabinet door open and closed state recognition algorithm, circuit breaker state recognition algorithm, isolating switch state recognition algorithm, and grounding switch state recognition algorithm. S2: The appearance defect image recognition algorithm specifically includes: oil stain recognition algorithm, rust recognition algorithm, paint peeling recognition algorithm, crack recognition algorithm, fissure recognition algorithm, burn mark recognition algorithm, dent recognition algorithm, bulge recognition algorithm, water accumulation recognition algorithm, etc. S3: The environmental monitoring image recognition algorithm specifically includes: smoke recognition algorithm, flame recognition algorithm, liquid leakage recognition algorithm, gas emission recognition algorithm, object left behind recognition algorithm, animal intrusion recognition algorithm, etc. S4: The algorithm configuration functions include: newly added algorithm deployment configuration, algorithm combination scheduling configuration, algorithm call configuration, algorithm license authorization configuration and other functions. S5: The algorithm service functions include: monitoring of hardware resources occupied by the algorithm, monitoring of the running status of the algorithm service, monitoring of the number of algorithm calls, as well as the algorithm service activation function and the algorithm service deactivation function, etc. S6: Algorithm application functions include: hardware intelligent device registration, real-time image data access, offline image data import, as well as pre-recognition data reception, raw data recognition and analysis, and post-recognition data feedback.

5. According to claim 1, a nuclear power intelligent inspection system based on industrial Internet technology is characterized in that: The intelligent inspection application includes: basic inspection configuration, inspection task triggering, inspection execution, inspection result analysis, inspection report management, inspection statistical analysis and other functions. S0: The basic inspection configuration includes: inspection object configuration, inspection point configuration, point inspection standard configuration, inspection route configuration, inspection task configuration and other functions, as well as the writing, checking, reviewing and approving functions of each configuration. S1: The inspection task triggering includes: the inspection task system automatically issues the inspection task execution instruction and the business personnel manually issues the inspection task execution instruction operation. S2: The inspection task execution includes: starting each intelligent inspection device in each scene, each intelligent inspection device in each scene collecting original data in turn, transmitting the original data to the industrial Internet platform, identifying the original data into text data, comparing and analyzing the text data with the inspection standard, generating the inspection results and other functions. S3: The inspection result analysis includes: over-limit and over-bounds analysis of text data after original data recognition, trend curve analysis, dead point analysis, shaking analysis, etc. S4: The inspection report management includes: inspection report generation, inspection result value correction, inspection report approval, inspection report archiving and other functions. S5: The inspection statistical analysis includes: online status analysis of intelligent hardware devices, inspection task completion rate statistical analysis, alarm type statistical analysis, alarm level statistical analysis, etc.