CN117704971A - Three-dimensional thermal displacement on-line measuring and predicting system for key parts of power plant boiler - Google Patents
Three-dimensional thermal displacement on-line measuring and predicting system for key parts of power plant boiler Download PDFInfo
- Publication number
- CN117704971A CN117704971A CN202311660373.7A CN202311660373A CN117704971A CN 117704971 A CN117704971 A CN 117704971A CN 202311660373 A CN202311660373 A CN 202311660373A CN 117704971 A CN117704971 A CN 117704971A
- Authority
- CN
- China
- Prior art keywords
- boiler
- data
- monitoring
- unit
- displacement
- 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
- 238000006073 displacement reaction Methods 0.000 title claims abstract description 55
- 238000012544 monitoring process Methods 0.000 claims abstract description 55
- 238000005259 measurement Methods 0.000 claims abstract description 38
- 239000013307 optical fiber Substances 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 27
- 238000004891 communication Methods 0.000 claims abstract description 15
- 238000013135 deep learning Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 238000012806 monitoring device Methods 0.000 claims abstract description 5
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 230000000007 visual effect Effects 0.000 claims description 12
- 238000012800 visualization Methods 0.000 claims description 12
- 238000007726 management method Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000009434 installation Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004321 preservation Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- PNEYBMLMFCGWSK-UHFFFAOYSA-N aluminium oxide Inorganic materials [O-2].[O-2].[O-2].[Al+3].[Al+3] PNEYBMLMFCGWSK-UHFFFAOYSA-N 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 230000009977 dual effect Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000017525 heat dissipation Effects 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 2
- 230000001681 protective effect Effects 0.000 claims description 2
- 230000002265 prevention Effects 0.000 claims 1
- 239000000835 fiber Substances 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000013144 data compression Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000035882 stress Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000008646 thermal stress Effects 0.000 description 1
Abstract
The invention relates to the technical field of power plant boiler pipeline detection equipment, in particular to a system and a method for three-dimensional thermal displacement on-line measurement and prediction of key parts of a power plant boiler; the system comprises: the system comprises a 1# unit, a 2# unit, a 1# access layer switch, a 2# access layer switch, a 1# optical fiber, a 2# optical fiber, a 1# collecting layer switch, a 2# collecting layer switch, a 1# optical fiber, a 2# optical fiber, a core switch, a network server, a recorder, a working condition prediction server, an industrial control domain firewall and an SIS power plant internal interface. The system measures, displays and records the thermal expansion data of the boiler in real time, and sends the field data to a power station control room through optical fiber or wireless communication, so that the system is convenient for monitoring and early warning the operation condition of the boiler in real time and accurately, and the safety and stability of the operation of a unit can be improved; and the boiler thermal expansion displacement monitoring device combining deep learning and double-camera vision realizes high-precision real-time on-line monitoring and prediction functions.
Description
Technical Field
The invention relates to the technical field of power plant boiler pipeline detection equipment, in particular to a three-dimensional thermal displacement on-line measurement and prediction system for key parts of a power plant boiler.
Background
During the continuous cycle of starting-loading-stopping, the power plant boiler is subjected to the test of thermal expansion and heat release shrinkage, the expansion or shrinkage generates corresponding stress, the boiler body and connected thermal components are deformed or displaced, particularly, thermal equipment with very high steam temperature and pressure such as a superheater, a reheater box and the like, if the generated stress exceeds the capability of the corresponding components to bear strain, leakage accidents endangering safe production can be generated, and the thermal equipment such as a steam-water system and the like are in a defective running state. On the other hand, the load fluctuation of the power grid is large, and the thermal power generating unit is required to frequently participate in peak shaving. In the peak shaving process, thermodynamic environment changes of all levels of equipment forming the steam-water system of the boiler are aggravated, so that the thermal physical characteristics and thermal stress distribution inside the equipment are more influenced. Therefore, the frequency of unplanned furnace shutdown caused by safety accidents such as deformation, rupture and the like of the boiler body tends to be increased, and the safe operation of the coal-fired power plant is seriously threatened
At present, a mechanical pointer type expansion indicator is generally arranged in an ultra-supercritical boiler of a power plant and is used for measuring deformation caused by thermal expansion of the boiler, and monitoring of monitoring points of the boiler wall surface is recorded on site by adopting a traditional manual inspection mode when the boiler is started, stopped and normally operated. The manual reading and recording is time-consuming and labor-consuming, the recorded monitoring data is discontinuous, the real-time performance is poor, the automation degree is low, the manual quantity is large, and the efficient and safe production requirements of Chen Jiang Kong companies are difficult to meet. The application number is 202111289798.2, and the application discloses a non-contact three-dimensional thermal displacement detection system and application thereof, wherein the system comprises a binocular camera, a laser range finder, a vision processing cabinet and a computer control system, adopts an artificial intelligence technology, realizes a measurement mode of parameters such as boiler expansion amount, pipeline displacement amount and the like, and can automatically early warn the safe operation of equipment. With the development of neural networks and deep learning technologies, parameters which have significant influence on a boiler main body are identified from a plurality of parameters by using an unsupervised neural network learning method, so that the three-dimensional thermal displacement on-line measurement and prediction of key parts of the power plant boiler are realized, and the method is a trend of future development.
Disclosure of Invention
The purpose of the invention is that: in order to solve the defects of the existing power plant boiler expansion amount measurement and method, the invention provides an online measurement and prediction system for three-dimensional thermal displacement of a key part of a power plant boiler, and adopts a machine learning method to deeply excavate measured real-time data and accumulated historical data so as to realize real-time monitoring of physical parameters related to thermal expansion, deformation and displacement of the key part of the power plant boiler.
The technical scheme is as follows: in order to solve the problems, the invention provides a three-dimensional thermal displacement on-line measuring and predicting system for key parts of a power plant boiler, which has the following innovation and novel bright points:
the invention provides a three-dimensional thermal displacement on-line measurement and prediction system for key parts of a power plant boiler, which comprises the following components: the system comprises a 1# unit, a 2# unit, a 1# access layer switch, a 2# access layer switch, a 1# optical fiber, a 2# optical fiber, a 1# collecting layer switch, a 2# collecting layer switch, a 1# optical fiber, a 2# optical fiber, a core switch, a network server, a recorder, a working condition prediction server, an industrial control domain firewall and an SIS power plant internal interface; the system comprises a 1# unit, a 1# access layer switch, a 1# collecting layer switch, a 1# splitting fiber, a core switch and a 1# optical fiber, wherein various information data on the 1# unit is connected with the 1# access layer switch through the 1# optical fiber and then collected on the 1# collecting layer switch, and then collected on the core switch through the 1# splitting fiber; similarly, various information data on the 2# unit are connected with a 2# access layer switch through a 2# optical fiber, and then are collected on a 2# collection layer switch, and then are collected on a core switch through a 2# optical fiber; the core exchanger transmits various information data to the network server, the recorder and the working condition prediction server respectively, and the industrial control domain firewall is connected with the internal interface of the SIS power plant and the network server.
Specifically, the 1# access layer switch and the 2# access layer switch adopt gigabit ethernet switches for network construction.
Specifically, the 1# collection layer switch and the 2# collection layer switch are industrial products, and are provided with 24 kilomega lights, 8 kilomega Ethernet, 4 megamega lights, enterprise network WEB network management weak three-layer switches, lightning protection common mode earth l OkV, and the exchange capacity is more than or equal to 336Gbps.
Specifically, the working condition prediction server is used for being connected with a network switch through a gigabit ethernet card so as to improve the data access speed, has a larger storage capacity and an advanced data compression mode, is used for storing real-time data, historical data and calculation and analysis results of the running states of all devices, and provides a data backup means; the disk capacity at least can meet the requirement of storing the operation data of the main auxiliary equipment for 10 years; the server must adopt mature and reliable fault-tolerant technology, and by configuring necessary software and hardware, the availability of the server system is ensured to be 99.9%, no single point fault exists, the system is not interrupted, and continuous monitoring and analysis of the unit are ensured.
Specifically, the SIS power plant internal interface is connected with a power plant internal system and is mainly used for collecting and outputting information.
Specifically, the outer protection level of the server cabinet is NEMA12, the cabinet door is provided with a conductive door gasket strip, an exhaust fan and an internal circulation fan are provided for a power supply device needing heat dissipation and are easy to replace, the server cabinet is internally illuminated, the communication cable is connected by adopting an optical fiber or twisted pair plug, and other cables are connected by adopting a terminal strip; each terminal strip or communication interface in the cabinet should be provided with a clear mark, and is in accordance with common paper and a wiring meter, and cabinet color codes adopt RAL9005.
Further, the 1# unit and the 2# unit comprise a boiler, a monitoring grid target, a double-camera measuring device, a three-dimensional coordinate detector, a server cabinet and a central control unit;
specifically, the monitoring grid target adopts a high-precision alumina checkerboard, the precision is +/-0.001 mm, the external dimension is 1000mm x 800mm, the grid dimension is 10mm x 10mm, and the number of effective corner points is 95 x 64; the large target is selected to obtain enough data once in an XY plane, and the monitoring grid target only needs to move along the Z direction; in practical application, the angles of the two cameras are required to be adjusted, so that each camera can shoot in a calibration range to obtain a complete target pattern.
Specifically, the three-dimensional coordinate detector adopts a visual measurement mode to perform three-dimensional measurement on a target point, the period interval time of each group of data is not more than 1 second, the period interval is adjustable, and the measurement error is less than 1%; the data network disconnection preservation is required to be supported, the preservation time of the historical data is not less than 2 months, and the historical data can be circularly covered; the monitoring function is needed, and video monitoring can be performed near the detection area; a data interface is provided for data communication with other systems.
Specifically, the central control unit is used for obtaining an image shot by the dual-camera measuring device, processing and analyzing the image to obtain pixel coordinates of the monitoring grid target corner point on the dual-camera measuring device 23, calculating to obtain space coordinates of corresponding points based on the BP neural network, and completing measurement of thermal expansion displacement of the boiler;
further, the dual-camera measuring device comprises a left camera, a right camera and a camera fixing bracket for fixing the left camera and the right camera; according to the binocular vision measurement principle, when the parallax of the two cameras is large, the binocular vision measurement principle is sensitive to the measurement of the depth direction, but the actual field installation is convenient, and the distance between the two cameras is designed to be 1000mm; the camera used for the experiment supported 300 ten thousand pixels (2048×1536), 1/2.8"cmos; before use, the calibration is carried out, the calibration range is in the range of 1500mm to 2000mm in the positive direction of the Z axis (the distance range can be adjusted according to actual conditions), and a lens with the focal length of 8mm is used for imaging; the specific calibration method is as follows:
the invention also provides a method for using the three-dimensional thermal displacement on-line measuring and predicting system for the key parts of the power plant boiler, which is characterized by comprising the following steps:
s1: firstly, establishing a three-dimensional model of a main body structure of a boiler; building a full-size 1:1 ratio three-dimensional visualization model of the boiler main body;
s2: on the basis of the three-dimensional visualization model of the boiler main body, a three-dimensional visualization information management platform is established, wherein the three-dimensional visualization information management platform comprises a boiler wall monitoring point information monitoring real-time database, a history database and a statistics database, and can be used for performing function expansion;
s3: on the three-dimensional visual information management platform, a prediction model is established under different working condition modes, and effective identification and main influence parameter identification are carried out on the thermal expansion operation working condition of the boiler; introducing an unsupervised neural network learning method, and identifying parameters which have significant influence on the boiler main body from a plurality of parameters;
s4: then, an LSTM neural network is used for establishing an intelligent prediction model of the thermal expansion operation condition of the boiler, l displacement data of each measuring point on the wall surface of the boiler within a few minutes in the future are predicted, meanwhile, the predicted data and the real data are compared and analyzed, and whether the boiler operates normally or not is detected through the degree of deviation of the data from a threshold value;
s5: finally, an AI visual intelligent monitoring, preventing and feedback platform is established based on a prediction model through the evaluation of the thermal expansion running state of the boiler and the abnormal early warning; the method comprises the steps of comparing and analyzing the information of a database and a set threshold value, and carrying out real-time monitoring and abnormal early warning on the installation state of the thermal expansion of the boiler; if the alarm is failed, the alarm trigger sound and the three-dimensional scene follow-up are carried out, the computer host generates the alarm sound when the alarm is carried out, meanwhile, the icon flashes, the real-time monitoring platform automatically zooms and translates to the alarm position, the detailed information of the alarm and the associated camera video are popped up on the screen, and the boiler thermal expansion displacement monitoring device combining deep learning and binocular vision is used for realizing the high-precision and real-time online monitoring function.
Furthermore, the present invention also provides an AI visual monitoring, preventing and feedback platform device, which is characterized by comprising:
a camera configuration module: parameters used for configuring cameras, such as a scene head distortion parameter, an internal azimuth element and the like, are used for assisting in the installation of an observation target, and the proportionality coefficients of the object space distance and the image space distance are calculated;
and an image acquisition module: the system is used for directly reading the acquired data through a specific program, and comprises triaxial displacement and triaxial coordinates;
and an image processing and displacement calculating module: the method is used for obtaining a calibrated angular point extraction diagram by adopting an image correlation algorithm and a binocular vision algorithm, calculating displacement in three directions of XYZ, and carrying out early warning operation and alarm operation;
and the deep learning module is used for: the method is used for excavating displacement variation trend through deep learning and big data analysis, and finding measures for preventing the gradual expansion of the thermal expansion displacement of the boiler;
and a data communication module: the method is used for automatically realizing the processes of positive plate acquisition, angular point extraction, neural network calculation and displacement calculation by software according to the set time interval, and displaying the measurement result in an interface to realize full-automatic monitoring of displacement.
By implementing the invention, the following beneficial effects can be obtained: the system measures, displays and records the thermal expansion data of the boiler in real time, and sends the field data to a power station control room through optical fiber or wireless communication, so that the system is convenient for monitoring and early warning the operation condition of the boiler in real time and accurately, and the safety and stability of the operation of a unit can be improved; and the boiler thermal expansion displacement monitoring device combining deep learning and double-camera vision realizes high-precision real-time on-line monitoring and prediction functions.
Drawings
FIG. 1 is a schematic diagram of a three-dimensional thermal displacement on-line measurement and prediction system for key parts of a power plant boiler according to an embodiment of the invention.
FIG. 2 is a schematic diagram of the on-line measurement system for the thermal expansion of the boiler according to the embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an AI visual monitoring, preventing and feedback platform system according to an embodiment of the present invention.
In the above figures, 1.1# machine set, 2.2# machine set, 3.1# access layer switch, 4.2# access layer switch, 5.1# optical fiber, 6.2# optical fiber, 7.1# convergence layer switch, 8.2# convergence layer switch, 9.1# optical fiber, 10.2# optical fiber, 13. Core switch, 14. Network server, 15. Recorder, 16. Working condition prediction server, 17. Industrial control domain firewall, 18.sis power plant internal interface, 21. Boiler, 22. Monitoring grid target, 23. Dual camera measuring device, 24. Three-dimensional coordinate detector, 25. Server cabinet, 26. Central control unit, 31. Camera configuration module, 32. Image acquisition module, 33. Image processing and displacement calculation module, 34. Deep learning module and 35. Data communication module.
Detailed Description
Example 1
The following describes the embodiments of the present invention in detail with reference to the drawings.
The invention provides a three-dimensional thermal displacement on-line measuring and predicting system for key parts of a power plant boiler, as shown in fig. 1, comprising: 1# machine set 1, 2# machine set 2, 1# access layer switch 3, 2# access layer switch 4, 1# optical fiber 5, 2# optical fiber 6, 1# collecting layer switch 7, 2# collecting layer switch 8, 1# optical fiber 9, 2# optical fiber 10, core switch 13, network server 14, recorder 15, working condition prediction server 16, industrial control domain firewall 17, SIS power plant internal interface 18; various information data on the 1# unit 1 are connected with the 1# access layer switch 3 through the 1# optical fiber 5, are collected to the 1# collection layer switch 7, and are collected to the core switch 13 through the 1# optical fiber 9; similarly, various information data on the 2# machine set 2 are connected with the 2# access layer switch 4 through the 2# optical fiber 6, are collected on the 2# collection layer switch 8, and are collected on the core switch 13 through the 2# light splitting fiber 10; the core switch 13 transmits various information data to the network server 14, the recorder 15 and the working condition prediction server 16 respectively, and the industrial control domain firewall 17 is connected with the internal interface 18 of the SIS power plant and also connected with the network server 14.
Specifically, the 1# access layer switch 3 and the 2# access layer switch 4 belong to gigabit ethernet switches and are used for network construction.
Specifically, the 1# collection layer switch 7 and the 2# collection layer switch 8 should be industrial products, and should have 24 gigabit optical, 8 gigabit ethernet, 4 gigabit optical, enterprise network WEB network management weak three-layer switches, lightning protection common mode earth l OkV, and the switching capacity is more than or equal to 336Gbps.
Specifically, the working condition prediction server 16 is configured to be connected to a network switch through a gigabit ethernet card, so as to increase the data access speed, have a larger storage capacity and an advanced data compression mode, and be configured to store real-time data, historical data and calculation and analysis results of all operating states of the device, and provide a data backup means; the disk capacity at least can meet the requirement of storing the operation data of the main auxiliary equipment for 10 years; the server must adopt mature and reliable fault-tolerant technology, and by configuring necessary software and hardware, the availability of the server system is ensured to be 99.9%, no single point fault exists, the system is not interrupted, and continuous monitoring and analysis of the unit are ensured.
Specifically, the SIS plant internal interface 18 is connected to the plant internal system and is primarily used to collect and output information.
Further, the 1# unit 1 and the 2# unit 2 further comprise a boiler 21, a monitoring grid target 22, a dual-camera measuring device 23, a three-dimensional coordinate detector 24, a server cabinet 25 and a central control unit 26, as shown in fig. 2;
specifically, the server cabinet 25 has a protective level of NEMA12, a cabinet door has a conductive door gasket strip, an exhaust fan and an internal circulation fan should be provided for a power supply device needing heat dissipation, and the power supply device is easy to replace, has illumination in the cabinet, and is connected by an optical fiber or twisted pair plug, and other cables are connected by terminal bars; each terminal strip or communication interface in the cabinet should be provided with a clear mark, and is in accordance with common paper and a wiring meter, and cabinet color codes adopt RAL9005.
Specifically, the monitoring grid target 22 adopts a high-precision alumina checkerboard, the precision is +/-0.001 mm, the external dimension is 1000mm x 800mm, the grid dimension is 10mm x 10mm, and the number of effective corner points is 95 x 64; the large target is selected to obtain enough data once in the XY plane, and the monitoring grid target 22 only needs to move along the Z direction; in practical application, the angles of the two cameras are required to be adjusted, so that each camera can shoot in a calibration range to obtain a complete target pattern.
Specifically, the three-dimensional coordinate detector 24 adopts a visual measurement mode to perform three-dimensional measurement on the target point, the period interval time of each group of data is not more than 1 second, the period interval is adjustable, and the measurement error is less than 1%; the data network disconnection preservation is required to be supported, the preservation time of the historical data is not less than 2 months, and the historical data can be circularly covered; the monitoring function is needed, and video monitoring can be performed near the detection area; a data interface is provided for data communication with other systems.
Specifically, the central control unit 26 is configured to obtain an image captured by the dual-camera measurement device 23, process and analyze the image to obtain pixel coordinates of the corner of the monitoring grid target 22 on the dual-camera measurement device 23, calculate, based on the BP neural network, spatial coordinates of corresponding points, and complete measurement of thermal expansion displacement of the boiler 21;
further, the dual camera measuring device 23 includes a left camera and a right camera, and a camera fixing bracket for fixing the left camera and the right camera; according to the binocular vision measurement principle, when the parallax of the two cameras is large, the binocular vision measurement principle is sensitive to the measurement of the depth direction, but the actual field installation is convenient, and the distance between the two cameras is designed to be 1000mm; the camera used for the experiment supported 300 ten thousand pixels (2048×1536), 1/2.8"cmos; before use, the calibration is carried out, the calibration range is in the range of 1500mm to 2000mm in the positive direction of the Z axis (the distance range can be adjusted according to actual conditions), and a lens with the focal length of 8mm is used for imaging; the specific calibration method is as follows:
the invention also provides a three-dimensional thermal displacement on-line measuring and predicting system for the key parts of the power plant boiler, which is characterized by comprising the following steps:
s1: firstly, establishing a three-dimensional model of a main body structure of a boiler; building a full-size 1:1 ratio three-dimensional visualization model of the boiler main body;
s2: on the basis of the three-dimensional visualization model of the boiler main body, a three-dimensional visualization information management platform is established, wherein the three-dimensional visualization information management platform comprises a boiler wall monitoring point information monitoring real-time database, a history database and a statistics database, and can be used for performing function expansion;
s3: on the three-dimensional visual information management platform, a prediction model is established under different working condition modes, and effective identification and main influence parameter identification are carried out on the thermal expansion operation working condition of the boiler; introducing an unsupervised neural network learning method, and identifying parameters which have significant influence on the boiler main body from a plurality of parameters;
s4: then, an LSTM neural network is used for establishing an intelligent prediction model of the thermal expansion operation condition of the boiler, l displacement data of each measuring point on the wall surface of the boiler within a few minutes in the future are predicted, meanwhile, the predicted data and the real data are compared and analyzed, and whether the boiler operates normally or not is detected through the degree of deviation of the data from a threshold value;
s5: finally, an AI visual intelligent monitoring, preventing and feedback platform is established based on a prediction model through the evaluation of the thermal expansion running state of the boiler and the abnormal early warning; the method comprises the steps of comparing and analyzing the information of a database and a set threshold value, and carrying out real-time monitoring and abnormal early warning on the installation state of the thermal expansion of the boiler; if the alarm is failed, the alarm trigger sound and the three-dimensional scene follow-up are carried out, the computer host generates the alarm sound when the alarm is carried out, meanwhile, the icon flashes, the real-time monitoring platform automatically zooms and translates to the alarm position, the detailed information of the alarm and the associated camera video are popped up on the screen, and the boiler thermal expansion displacement monitoring device combining deep learning and binocular vision is used for realizing the high-precision and real-time online monitoring function.
The invention also provides an AI visual monitoring, preventing and feedback platform device, as shown in figure 3, which is characterized by comprising:
a camera configuration module: parameters used for configuring cameras, such as a scene head distortion parameter, an internal azimuth element and the like, are used for assisting in the installation of an observation target, and the proportionality coefficients of the object space distance and the image space distance are calculated;
and an image acquisition module: the system is used for directly reading the acquired data through a specific program, and comprises triaxial displacement and triaxial coordinates;
and an image processing and displacement calculating module: the method is used for obtaining a calibrated angular point extraction diagram by adopting an image correlation algorithm and a binocular vision algorithm, calculating displacement in three directions of XYZ, and carrying out early warning operation and alarm operation;
and the deep learning module is used for: the method is used for excavating displacement variation trend through deep learning and big data analysis, and finding measures for preventing the gradual expansion of the thermal expansion displacement of the boiler;
and a data communication module: the method is used for automatically realizing the processes of positive plate acquisition, angular point extraction, neural network calculation and displacement calculation by software according to the set time interval, and displaying the measurement result in an interface to realize full-automatic monitoring of displacement.
The invention provides a three-dimensional thermal displacement on-line measuring and predicting system of a boiler key part based on the combination of a double-camera measuring device 23 and a neural network, which can measure, display and record the thermal expansion data of the boiler in real time, and send the field data to a power station control room through optical fiber or wireless communication, thereby being convenient for monitoring the operation condition of the boiler in real time and accurately, and calibrating and carrying out field measurement test on the measuring system; the results show that: the standard deviation of the system for multiple measurement is less than 0.05mm, the repeatability is good, the errors in different directions and under different displacement distances are all below 0.6%, the zero drift is less than 0.1mm, and the real-time measurement requirement of the three-dimensional thermal displacement of the key part of the boiler can be met.
The invention is not limited to the specific embodiments described above, which are intended to be illustrative only and not limiting; those skilled in the art, having the benefit of this disclosure, may make numerous forms without departing from the spirit of the invention and the scope of the claims which follow.
Claims (10)
1. The utility model provides a three-dimensional heat displacement on-line measuring of power plant boiler key position and prediction system which characterized in that: the system comprises a 1# unit (1), a 2# unit (2), a 1# access layer switch (3), a 2# access layer switch (4), a 1# optical fiber (5), a 2# optical fiber (6), a 1# collecting layer switch (7), a 2# collecting layer switch (8), a 1# optical fiber (9), a 2# optical fiber (10), a core switch (13), a network server (14), a recorder (15), a working condition prediction server (16), an industrial control domain firewall (17) and an SIS power plant internal interface (18); various information data on the 1# unit (1) are connected with the 1# access layer switch (3) through a 1# optical fiber (5), and then are collected on the 1# collection layer switch (7), and then are collected on the core switch (13) through a 1# optical fiber (9); similarly, various information data on the 2# unit (2) are connected with the 2# access layer switch (4) through the 2# optical fiber (6), and then are collected on the 2# collection layer switch (8), and then are collected on the core switch (13) through the 2# optical fiber (10); the core switch (13) transmits various information data to the network server (14), the recorder (15) and the working condition prediction server (16) respectively, and the industrial control domain firewall (17) is connected with the SIS power plant internal interface (18) and is also connected with the network server (14).
2. The system for three-dimensional thermal displacement on-line measurement and prediction of key parts of a power plant boiler according to claim 1, wherein the working condition prediction server (16) is used for storing real-time data, historical data and calculation and analysis results of all equipment operation states and providing a data backup means by being connected with a network switch through a gigabit ethernet card; the disk capacity at least can meet the requirement of storing the operation data of the main auxiliary equipment for 10 years; the server must adopt mature and reliable fault-tolerant technology, and by configuring necessary software and hardware, the availability of the server system is ensured to be 99.9%, no single point fault exists, the system is not interrupted, and continuous monitoring and analysis of the unit are ensured.
3. The system for three-dimensional thermal displacement on-line measurement and prediction of key parts of a power plant boiler according to claim 1, wherein the No. 1 unit (1) and the No. 2 unit (2) further comprise a boiler (21), a monitoring grid target (22), a double-camera measuring device (23), a three-dimensional coordinate detector (24), a server cabinet (25) and a central control unit (26).
4. A 1# unit (1) and a 2# unit (2) according to claim 3, wherein the server cabinet (25) has a protective level of NEMA12, a cabinet door is provided with a conductive door gasket strip, an exhaust fan and an internal circulation fan are provided for a power supply device needing heat dissipation and are easy to replace, the power supply device has illumination in the cabinet, the connection of communication cables adopts optical fibers or twisted pair plugs, and other cables adopt terminal row connection; each terminal strip or communication interface in the cabinet should be provided with a clear mark, and is in accordance with common paper and a wiring meter, and cabinet color codes adopt RAL9005.
5. A 1# unit (1) and a 2# unit (2) according to claim 3, wherein the monitoring grid targets (22) are high-precision alumina grids with a precision of ±0.001mm, an external dimension of 1000mm x 800mm, a grid dimension of 10mm x 10mm, and an effective number of angles of 95 x 64.
6. A 1# unit (1) and a 2# unit (2) according to claim 3, wherein the three-dimensional coordinate detector (24) adopts a visual measurement mode to perform three-dimensional measurement on a target point, the period interval time of each group of data is no more than 1 second, the period interval is adjustable, and the measurement error is less than 1%; the data network disconnection preservation is required to be supported, the preservation time of the historical data is not less than 2 months, and the historical data can be circularly covered; the monitoring function is needed, and video monitoring can be performed near the detection area; a data interface is provided for data communication with other systems.
7. A 1# unit (1) and a 2# unit (2) according to claim 3, wherein the central control unit (26) is used for obtaining an image shot by the dual-camera measuring device, processing and analyzing the image to obtain pixel coordinates of the corner points of the monitoring grid target (22) on the dual-camera measuring device (23), calculating to obtain space coordinates of corresponding point positions based on the BP neural network, and completing the measurement of the thermal expansion displacement of the boiler.
8. A 1# unit (1) and a 2# unit (2) according to claim 3, wherein the dual camera measuring device (23) comprises left and right cameras and a camera fixing bracket for fixing the left and right cameras; the distance between the two cameras is 1000mm; the camera used supports 300 ten thousand pixels (2048 x 1536), 1/2.8"cmos; before use, calibration is carried out, the calibration range is in the range of 1500mm to 2000mm in the positive direction of the Z axis (the distance range can be adjusted according to practical conditions), and a lens with the focal length of 8mm is used for imaging.
9. A system for three-dimensional thermal displacement on-line measurement and prediction of critical parts of a power plant boiler according to claim 1, comprising the following steps:
s1: firstly, establishing a three-dimensional model of a main body structure of a boiler; building a full-size 1:1 ratio three-dimensional visualization model of the boiler main body;
s2: on the basis of the three-dimensional visualization model of the boiler main body, a three-dimensional visualization information management platform is established, wherein the three-dimensional visualization information management platform comprises a boiler wall monitoring point information monitoring real-time database, a history database and a statistics database, and can be used for performing function expansion;
s3: on the three-dimensional visual information management platform, a prediction model is established under different working condition modes, and effective identification and main influence parameter identification are carried out on the thermal expansion operation working condition of the boiler; introducing an unsupervised neural network learning method, and identifying parameters which have significant influence on the boiler main body from a plurality of parameters;
s4: then, an LSTM neural network is used for establishing an intelligent prediction model of the thermal expansion operation condition of the boiler, l displacement data of each measuring point on the wall surface of the boiler within a few minutes in the future are predicted, meanwhile, the predicted data and the real data are compared and analyzed, and whether the boiler operates normally or not is detected through the degree of deviation of the data from a threshold value;
s5: finally, an AI visual intelligent monitoring, preventing and feedback platform is established based on a prediction model through the evaluation of the thermal expansion running state of the boiler and the abnormal early warning; the method comprises the steps of comparing and analyzing the information of a database and a set threshold value, and carrying out real-time monitoring and abnormal early warning on the installation state of the thermal expansion of the boiler; if the alarm is failed, the alarm trigger sound and the three-dimensional scene follow-up are carried out, the computer host generates the alarm sound when the alarm is carried out, meanwhile, the icon flashes, the real-time monitoring platform automatically zooms and translates to the alarm position, the detailed information of the alarm and the associated camera video are popped up on the screen, and the boiler thermal expansion displacement monitoring device combining deep learning and binocular vision is used for realizing the high-precision and real-time online monitoring function.
10. An AI vision monitoring, prevention and feedback platform, comprising:
a camera configuration module: parameters used for configuring cameras, such as a scene head distortion parameter, an internal azimuth element and the like, are used for assisting in the installation of an observation target, and the proportionality coefficients of the object space distance and the image space distance are calculated;
and an image acquisition module: the system is used for directly reading the acquired data through a specific program, and comprises triaxial displacement and triaxial coordinates;
and an image processing and displacement calculating module: the method is used for obtaining a calibrated angular point extraction diagram by adopting an image correlation algorithm and a binocular vision algorithm, calculating displacement in three directions of XYZ, and carrying out early warning operation and alarm operation;
and the deep learning module is used for: the method is used for excavating displacement variation trend through deep learning and big data analysis, and finding measures for preventing the gradual expansion of the thermal expansion displacement of the boiler;
and a data communication module: the method is used for automatically realizing the processes of positive plate acquisition, angular point extraction, neural network calculation and displacement calculation by software according to the set time interval, and displaying the measurement result in an interface to realize full-automatic monitoring of displacement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311660373.7A CN117704971A (en) | 2023-12-06 | 2023-12-06 | Three-dimensional thermal displacement on-line measuring and predicting system for key parts of power plant boiler |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311660373.7A CN117704971A (en) | 2023-12-06 | 2023-12-06 | Three-dimensional thermal displacement on-line measuring and predicting system for key parts of power plant boiler |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117704971A true CN117704971A (en) | 2024-03-15 |
Family
ID=90143463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311660373.7A Pending CN117704971A (en) | 2023-12-06 | 2023-12-06 | Three-dimensional thermal displacement on-line measuring and predicting system for key parts of power plant boiler |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117704971A (en) |
-
2023
- 2023-12-06 CN CN202311660373.7A patent/CN117704971A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109870108B (en) | Method and device for detecting icing of power transmission line | |
| CN109915736B (en) | Thermal pipe network system and leakage detection method thereof | |
| CN107167497A (en) | A kind of equipment fault detection method and system | |
| CN112083000B (en) | Intelligent identification method and system for appearance defects of substation equipment | |
| CN108627794B (en) | Intelligent instrument detection method based on deep learning | |
| CN109945977B (en) | Thermal fault diagnosis method and system for dry-type transformer | |
| CN113299035A (en) | Fire identification method and system based on artificial intelligence and binocular vision | |
| CN114417669A (en) | Power transformation equipment fault monitoring and early warning method and device based on digital twinning | |
| CN108680808A (en) | Fault diagnosis method and device | |
| CN110796754A (en) | Machine room inspection method based on image processing technology | |
| KR20200030317A (en) | Augmented Reality Platform for On-site inspection of electric power facilities using thermal imaging camera and IOT sensor | |
| CN106706159A (en) | Fiber grating transformer multi-parameter intelligent monitoring system based on multiagent technology | |
| CN109323765A (en) | A method of using curved profile method to the accurate thermometric of power equipment | |
| CN117704971A (en) | Three-dimensional thermal displacement on-line measuring and predicting system for key parts of power plant boiler | |
| CN117335556A (en) | Power distribution early warning method and system of power distribution cabinet | |
| EP1314669B1 (en) | A method and a system for monitoring a paper web or the like | |
| CN114942075A (en) | Power equipment fault detection and analysis method | |
| CN112364684B (en) | Machine room server state acquisition and three-dimensional management and control system and method thereof | |
| CN114821025A (en) | Meter identification method and system based on deep learning | |
| CN111289139A (en) | Optical fiber temperature measurement abnormal data positioning method based on two-dimensional space | |
| CN110031152B (en) | Thermal heating system and method for detecting temperature deviation accumulation leakage of adjacent days | |
| CN110131784B (en) | Thermal heating system and all-day temperature deviation accumulation leakage detection method thereof | |
| CN114237190A (en) | 3D digital factory equipment management method and system | |
| CN112611331A (en) | Displacement monitoring device and measuring method | |
| CN117387775B (en) | Infrared temperature measurement and wireless temperature measurement monitoring system for electrical equipment |
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 |