CN112162543B - Blade rotor test bed predictive maintenance method and system based on digital twinning - Google Patents
Blade rotor test bed predictive maintenance method and system based on digital twinning Download PDFInfo
- Publication number
- CN112162543B CN112162543B CN202011006102.6A CN202011006102A CN112162543B CN 112162543 B CN112162543 B CN 112162543B CN 202011006102 A CN202011006102 A CN 202011006102A CN 112162543 B CN112162543 B CN 112162543B
- Authority
- CN
- China
- Prior art keywords
- data
- module
- fault
- twin
- test bed
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a predictive maintenance method and system for a blade rotor test bed based on digital twinning. The system comprises three modules, a physical space module, an edge gateway module and a twin space module, wherein the physical space module mainly comprises static data and dynamic data of the blade rotor test bed, the edge gateway module mainly comprises sub modules such as an information receiving module, a data processing module, an abnormity detection module and a fault prediction module, the twin space module comprises sub modules such as a twin equipment module, a virtual scene module, a mathematical model module, a data storage module, a real-time monitoring module and a fault report module, and the predictive maintenance method of the blade rotor test bed is realized based on the system. On one hand, the invention realizes the predictive maintenance of the blade rotor test bed, and more importantly, the invention ensures the healthy operation of the blade rotor test bed, reduces the downtime and improves the test accuracy.
Description
Technical Field
The invention relates to the field of digital twinning, in particular to a blade rotor test bed predictive maintenance method and system based on digital twinning
Background
The digital twins are a simulation process integrating multiple disciplines, multiple probabilities and multiple scales, the digital twins technology is an important technology for realizing interactive fusion of a virtual space corresponding to a physical space established in a digital mode, the virtual twins and the physical entity operate synchronously on the premise of information interaction, and the physical space is fed back by analyzing behaviors and states of the virtual space.
Generally, during the experimental study of a test bed for a blade rotor, the equipment is repaired and maintained when it fails. However, in the process, the downtime is increased, so that the progress of the scientific research is greatly influenced. Based on this, how to realize the health management of the blade rotor test bed is particularly important.
Therefore, in the invention, a digital twin technology is utilized to create a test bed virtual twin body corresponding to the physical entity in a virtual space and monitor the test bed virtual twin body in real time, and maintenance decision is made through real-time state evaluation of the equipment, so that predictive maintenance of the equipment is realized. And judging the type and time of equipment failure according to data comparison and analysis, so as to formulate a reasonable failure solution and ensure the healthy operation of the blade rotor test bed.
Disclosure of Invention
The invention provides a predictive maintenance method and a predictive maintenance system for a blade rotor test bed based on digital twins, wherein a twins space corresponding to a physical space is established according to the physical space, an edge gateway is established to realize the connection between the physical space and the twins space, signals of vibration, temperature, pressure and the like of various parts such as a test bed driving system, a lubricating system, a rotor, a wheel disc, a bearing, a base and the like are collected in real time through establishing a bottom layer mixed sensing network, and the accurate fault location of the blade rotor test bed is realized through establishing a method of a data abnormity detection model to carry out abnormity detection on running data and establishing a network prediction model, so that a fault solution is formulated. The method firstly acts on the twin space, and verifies the reasonability and feasibility of the scheme by means of dismounting process simulation, control algorithm, man-machine work efficiency and the like, thereby achieving the aim of research by virtual control and reality. On one hand, the invention realizes the predictive maintenance of the blade rotor test bed, and more importantly, the invention ensures the healthy operation of the blade rotor test bed, reduces the downtime and improves the test accuracy.
A predictive maintenance system of a blade rotor test bed based on digital twinning comprises a physical space module, a twinning space module and an edge gateway module, wherein the edge gateway module is an important medium for realizing the effect of the twinning space on the physical space.
The physical space module is a set formed by people, machines, objects and environments thereof, and on one hand, provides static bottom layer data of the equipment, including equipment information, working parameters and the like; on the other hand, dynamic data are collected through various sensors when the blade rotor test bed operates, and the dynamic data comprise multi-source heterogeneous data such as rotor signals (rotating speed and three-dimensional vibration), blade signals (rotating speed and three-dimensional vibration), lubricating signals (lubricating oil pressure, oil temperature and liquid level), motor signals (rotating speed and torque), bearing signals (oil film pressure and temperature) and temperature and humidity in the surrounding environment.
And the data collected in the physical space module is sent to the edge gateway module in real time.
The twin space module comprises a twin equipment module, a virtual scene module, a mathematical model module, a data storage module, a real-time monitoring module, a fault reporting module and other sub-modules.
The twin equipment module is a blade rotor twin test bed constructed by three-dimensional modeling software and related experimental equipment and equipment.
The virtual scene module is built by a real scene, and immersive experience can be brought.
The mathematical model module is used for constructing a data anomaly detection model and a network prediction model, effectively storing the data anomaly detection model and the network prediction model and sending the data anomaly detection model and the network prediction model to the edge gateway module.
The established mathematical model is not limited to realizing fault location on the edge gateway, and based on multidimensional external factors, machine faults are not only caused by one reason, and fault prediction of partial multidimensional factors can be realized in a twin space.
The data storage module is used for storing the information transmitted by the edge gateway module.
The real-time monitoring module is used for displaying the data in a two-dimensional or three-dimensional form so as to realize real-time monitoring on data change.
The fault reporting module transmits fault information to the outside and makes a solution in time.
The edge gateway module is composed of sub-modules such as an information receiving module, a data processing module, an abnormality detection module and a fault prediction module.
The information receiving module is used for receiving current operation data when the physical space equipment works and a mathematical model constructed in the twin space.
The data processing module is used for optimizing and preprocessing the operation data.
The anomaly detection module is used for detecting the anomaly of the data based on an Isolation Forest (Isolation Forest) algorithm.
And the failure prediction module is used for predicting potential failures and potential safety hazards of the blade rotor test bed according to the established Bayesian network prediction model.
The physical space module, the twin space module and the edge gateway module are connected by Ethernet, and the physical space module and the twin space module are connected by a communication protocol.
The invention provides a predictive maintenance method of a blade rotor test bed based on digital twin, which is realized based on the system.
The technical scheme adopted by the invention is as follows:
a predictive maintenance method for a blade rotor test bed based on digital twinning comprises the following steps:
s1: creating a twin space corresponding to a physical entity in a physical space;
s2: collecting operating data of the blade rotor test bed during working, and forming operating data with unified data structure and data type after data optimization and data fusion;
s3: establishing a data anomaly detection model, inputting operation data with unified data structure and data type into the model for data anomaly detection, and removing the abnormal data if the abnormal data occurs;
s4: establishing a network prediction model, predicting potential faults of the blade rotor test bed, and formulating a fault solution according to fault types;
s5: applying a fault solution to the twin space for detecting the rationality and feasibility of the solution;
s6: judging whether the fault is solved in the twin space by means of dismounting process simulation, control algorithm, man-machine work efficiency and the like;
s7: if the fault is solved, maintaining the physical entity in the physical space, otherwise, re-formulating the scheme;
in step S1, a twin space of the blade rotor test stand is established by three-dimensional modeling software and simulation software.
In step S2, the upper computer sends a collection instruction to the data collection device, and the data is collected by the sensor.
In the data anomaly detection described in step S3, an isolated Forest (Isolation Forest) algorithm is used, an isolated Tree (Isolation Tree) is randomly generated by giving a string of data sets of continuous variables, each node on the Tree is either strung with two child nodes or forming a single leaf node, then data detection is performed according to the formed isolated Tree, the data is walked on the isolated Tree once, and the leaf node where the data falls is recorded, generally, the anomaly data is rare and is quickly distributed to the leaf nodes, so whether the data is the anomaly data is determined according to the height from the leaf node to the root node of the Tree.
In step S4, a bayesian network prediction model is built to predict a potential fault of the blade rotor test bed, a bayesian network prediction model is built by obtaining historical operating data of the device to predict a potential fault and a potential safety hazard of the test device, current operating data after data processing and data anomaly detection is obtained, overall reliability prediction and rapid fault positioning of the blade rotor test bed are realized according to the current operating data and the bayesian network prediction model, and a fault type and fault occurrence time are determined, so that a fault solution is formulated.
The types of failure described in step S4 are as follows: bearing abrasion, excessive oil film temperature or unstable rotor vibration and other fault types.
The bearing abrasion fault can determine the sequence of disassembling test equipment according to the disassembling process simulation, so as to determine the bearing abrasion part.
The oil temperature can be reduced by replacing lubricating oil or reducing the rotating speed of the bearing when the temperature of the oil film is too high.
The unstable vibration of the rotor can be caused by external micro fluctuation or unstable rotating speed of the rotor and other factors, and the vibration of the rotor can be stabilized by adjusting the rotating speed of the rotor and stabilizing the frequency.
The method has the advantages that the method is based on the idea of digital twins on the whole, the physical real world is acted through the twin world, the core idea of the digital twins is interpreted perfectly, the predictive maintenance of the blade rotor test bed is realized according to the analysis and processing of the running data of the test equipment and the idea of constructing a mathematical model, the unplanned downtime is avoided, the working benefit of the test equipment is improved, and the accuracy of test research is improved.
Drawings
FIG. 1 is a schematic diagram of a digital twinning-based predictive maintenance system for a test bed of a bladed rotor
FIG. 2 is a flow chart of a method for predictive maintenance of a bucket rotor test stand based on digital twinning
Detailed Description
The digital twin-based predictive maintenance method and system for a test bed of a bladed rotor according to the present invention will be described in detail by way of example with reference to the accompanying drawings. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto.
The first embodiment is as follows:
the embodiment provides a predictive maintenance method for a blade rotor test bed based on digital twinning, which comprises the following steps:
s1: creating a twin space corresponding to a physical entity in a physical space;
s2: collecting operating data of the blade rotor test bed during working, and forming operating data with unified data structure and data type after data optimization and data fusion;
s3: establishing a data anomaly detection model, inputting operation data with unified data structure and data type into the model for data anomaly detection, and removing the abnormal data if the abnormal data occurs;
s4: establishing a network prediction model, predicting potential faults of the blade rotor test bed, and formulating a fault solution according to fault types;
s5: applying a fault solution to the twin space for detecting the rationality and feasibility of the solution;
s6: judging whether the fault is solved in the twin space by means of dismounting process simulation, control algorithm, man-machine work efficiency and the like;
s7: if the fault is solved, maintaining the physical entity in the physical space, otherwise, re-formulating the scheme;
in step S1, a twin space of the blade rotor test stand is established by three-dimensional modeling software and simulation software.
In the embodiment, the physical real world is acted through the twin world based on the idea of the digital twin, the core idea of the digital twin is interpreted perfectly, the predictive maintenance of the blade rotor test bed is realized according to the analysis and processing of the operation data of the test equipment and the idea of constructing a mathematical model, the unplanned downtime is avoided, the working benefit of the test equipment is improved, and the accuracy of the test research is improved.
Example two:
the embodiment provides a predictive maintenance method for a blade rotor test bed based on digital twinning, which comprises the following steps:
s1: creating a twin space corresponding to a physical entity in a physical space;
s2: collecting operating data of the blade rotor test bed during working, and forming operating data with unified data structure and data type after data optimization and data fusion;
s3: establishing a data anomaly detection model, inputting operation data with unified data structure and data type into the model for data anomaly detection, and removing the abnormal data if the abnormal data occurs;
s4: establishing a network prediction model, predicting potential faults of the blade rotor test bed, and formulating a fault solution according to fault types;
s5: applying a fault solution to the twin space for detecting the rationality and feasibility of the solution;
s6: judging whether the fault is solved in the twin space by means of dismounting process simulation, control algorithm, man-machine work efficiency and the like;
s7: if the fault is solved, maintaining the physical entity in the physical space, otherwise, re-formulating the scheme;
in step 2, an upper computer sends a collection instruction to a data collection instrument, data collection is carried out through a rotating speed sensor, an acceleration sensor, an eddy current sensor and the like, and collected operation data comprise rotor signals (rotating speed and three-dimensional vibration), blade signals (rotating speed and three-dimensional vibration), lubricating signals (lubricating oil pressure, oil temperature and liquid level), motor signals (rotating speed and torque), bearing signals (oil film pressure and temperature) and multi-source heterogeneous data such as temperature and humidity in the surrounding environment.
Example three:
the embodiment provides a predictive maintenance method for a blade rotor test bed based on digital twinning, which comprises the following steps:
s1: creating a twin space corresponding to a physical entity in a physical space;
s2: collecting operating data of the blade rotor test bed during working, and forming operating data with unified data structure and data type after data optimization and data fusion;
s3: establishing a data anomaly detection model, inputting operation data with unified data structure and data type into the model for data anomaly detection, and removing the abnormal data if the abnormal data occurs;
s4: establishing a network prediction model, predicting potential faults of the blade rotor test bed, and formulating a fault solution according to fault types;
s5: applying a fault solution to the twin space for detecting the rationality and feasibility of the solution;
s6: judging whether the fault is solved in the twin space by means of dismounting process simulation, control algorithm, man-machine work efficiency and the like;
s7: if the fault is solved, maintaining the physical entity in the physical space, otherwise, re-formulating the scheme;
in the data anomaly detection described in step S3, an isolated Forest (Isolation Forest) algorithm is used, an isolated Tree (Isolation Tree) is randomly generated by giving a string of data sets of continuous variables, each node on the Tree is either strung with two child nodes or forming a single leaf node, then data detection is performed according to the formed isolated Tree, the data is walked on the isolated Tree once, and the leaf node where the data falls is recorded, generally, the anomaly data is rare and is quickly distributed to the leaf nodes, so the anomaly score condition of the point is determined according to the height from the leaf node to the root node of the Tree.
The height from the leaf node to the tree root node is used for judging the abnormal score condition, and the lower the height from the tree root node of the point is, the higher the abnormal score is, and the possibility of abnormal data is higher.
The anomaly score formula for one data point x is as follows:
s(x)＝2-E(h(x))
where E (h (x)) represents the average height in all the isolated trees (Isolation trees), the lower the height, the higher the anomaly score.
Often, when there are many data points, the height of the whole tree is relatively high, and abnormal data can be distinguished only by multiple segmentations, so that the average height is expressed by c (n) for normalization, and if there are n data, the average height formula is:
c(n)＝2H(n-1)-(2(n-1)/n)
where H (i) is the harmonic number, which can be approximated as:
H(i)≈ln(i)+0.5772156649
since c (n) represents the average height at n data points, we use to normalize h (x), then the normalized anomaly score is formulated as:
according to the above formula, the following conclusions can be drawn:
if E (h (x)) → 0, then s → 1, then an outlier data point is determined;
if E (h (x)) → n-1, then s → 0, then the data point is determined to be normal;
if E (h (x)) → c (n), then s → 0.5, it is not possible to distinguish between normal and abnormal.
Example four:
the embodiment provides a predictive maintenance method for a blade rotor test bed based on digital twinning, which comprises the following steps:
s1: creating a twin space corresponding to a physical entity in a physical space;
s2: collecting operating data of the blade rotor test bed during working, and forming operating data with unified data structure and data type after data optimization and data fusion;
s3: establishing a data anomaly detection model, inputting operation data with unified data structure and data type into the model for data anomaly detection, and removing the abnormal data if the abnormal data occurs;
s4: establishing a network prediction model, predicting potential faults of the blade rotor test bed, and formulating a fault solution according to fault types;
in step S4, a method of constructing a bayesian network prediction model is used to predict a fault of the blade rotor test bed, the bayesian network prediction model is established by obtaining historical operating data of the device, a potential fault and a potential safety hazard of the experimental device are predicted, current operating data after data processing and data anomaly detection are obtained, overall reliability prediction and rapid fault positioning of the blade rotor test bed are realized according to the current operating data and the bayesian network prediction model, a fault type and fault occurrence time are determined, and a fault solution is formulated.
S5: applying a fault solution to the twin space for detecting the rationality and feasibility of the solution;
s6: judging whether the fault is solved in the twin space by means of dismounting process simulation, control algorithm, man-machine work efficiency and the like;
s7: if the fault is solved, maintaining the physical entity in the physical space, otherwise, re-formulating the scheme;
example five:
the predictive maintenance system for the blade rotor test bed for realizing the method comprises a physical space module, a twin space module and an edge gateway module, wherein the edge gateway module is an important medium for realizing the twin space acting on the physical space.
The physical space module is a set formed by people, machines, objects and environments thereof, and on one hand, provides static bottom layer data of the equipment, including equipment information, working parameters and the like; on the other hand, dynamic data are collected through various sensors when the blade rotor test bed operates, and the dynamic data comprise multi-source heterogeneous data such as rotor signals (rotating speed and three-dimensional vibration), blade signals (rotating speed and three-dimensional vibration), lubricating signals (lubricating oil pressure, oil temperature and liquid level), motor signals (rotating speed and torque), bearing signals (oil film pressure and temperature) and temperature and humidity in the surrounding environment.
The data collected in the physical space module will be sent to the edge gateway module in real time.
The twin space module comprises a twin equipment module, a virtual scene module, a mathematical model module, a data storage module, a real-time monitoring module, a fault reporting module and other sub-modules.
The twin equipment module is a blade rotor twin test bed constructed by three-dimensional modeling software and related experimental equipment and equipment.
The virtual scene module is built by a real scene, and immersive experience can be brought.
The mathematical model module is used for constructing a data anomaly detection model and a network prediction model, effectively storing the data anomaly detection model and the network prediction model and sending the data anomaly detection model and the network prediction model to the edge gateway module.
The established mathematical model is not limited to realizing fault location on the edge gateway, and based on multidimensional external factors, machine faults are not only caused by one reason, and fault prediction of partial multidimensional factors can be realized in a twin space.
The data storage module is used for storing the information transmitted by the edge gateway module.
The real-time monitoring module is used for displaying the data in a two-dimensional or three-dimensional form so as to realize real-time monitoring on data change.
The fault reporting module is used for transmitting fault information to the outside and making a solution in time.
The edge gateway module is composed of sub-modules such as an information receiving module, a data processing module, an abnormality detection module and a fault prediction module.
The information receiving module is used for receiving current operation data when the physical space equipment works and a mathematical model constructed in the twin space.
The data processing module is used for optimizing and preprocessing the operation data.
The anomaly detection module is used for detecting the anomaly of the data based on an Isolation Forest (Isolation Forest) algorithm.
And the failure prediction module is used for predicting potential failures and potential safety hazards of the blade rotor test bed according to the established Bayesian network prediction model.
The physical space module and the twin space module are connected with the edge gateway module through Ethernet, and the physical space module and the twin space module are mainly connected through a communication protocol.
It should be noted that the interface type and data type of each experimental device are needed in the physical space module, and then the physical space module is connected with the edge gateway module through the ethernet, the edge computing gateway mainly undertakes preprocessing of the operation data collected from the physical space, performs data anomaly detection on the operation data through an anomaly detection model constructed in the twin space and realizes fault prediction on the blade rotor test bed through a network prediction model, the rationality and feasibility of the scheme are verified by acting the scheme on the twin space first and utilizing means such as dismounting process simulation, control algorithm and man-machine work efficiency, and the purpose of virtual-real combination is realized, the system reduces the data processing pressure born by the server at the twin space end, and the operation efficiency is integrally improved.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (2)
1. A predictive maintenance system of a blade rotor test bed based on digital twinning comprises a physical space module, a twinning space module and an edge gateway module, wherein the edge gateway module is an important medium for realizing the action of the twinning space on the physical space;
the physical space module is a set formed by people, machines, objects and environments thereof, and on one hand, provides static bottom layer data of the equipment, including equipment information and working parameters; on the other hand, dynamic data are collected through various sensors when the blade rotor test bed operates, the dynamic data comprise rotor signals, and the signals comprise rotor rotating speed and three-dimensional vibration; the blade signal comprises blade rotating speed and three-dimensional vibration; the lubricating signal comprises lubricating oil pressure, oil temperature and liquid level; the motor signal comprises the motor rotating speed and the motor torque; bearing signals, wherein the signals comprise oil film pressure, oil temperature, humidity and temperature in the surrounding environment;
the data collected in the physical space module is sent to the edge gateway module in real time;
the twin space module consists of a twin equipment module, a virtual scene module, a mathematical model module, a data storage module, a real-time monitoring module and a fault reporting module;
the twin equipment module is a blade rotor twin test bed constructed by three-dimensional modeling software and related experimental equipment and equipment;
the virtual scene module is built from a real scene, so that immersive experience can be brought;
the mathematical model module is used for constructing a data anomaly detection model and a network prediction model, effectively storing the data anomaly detection model and the network prediction model and sending the data anomaly detection model and the network prediction model to the edge gateway module;
the established mathematical model is not limited to realizing fault location on the edge gateway, and based on multidimensional external factors, the machine fault does not only contain one reason, and fault prediction of partial multidimensional factors can be realized in a twin space;
the data storage module is used for storing the information transmitted by the edge gateway module;
the real-time monitoring module is used for displaying data in a two-dimensional or three-dimensional form to realize real-time monitoring on data change;
the fault reporting module is used for transmitting fault information to the outside and making a solution in time;
the edge gateway module consists of an information receiving module, a data processing module, an abnormality detection module and a fault prediction module;
the information receiving module is used for receiving current operation data when the physical space equipment works and a mathematical model constructed in a twin space;
the data processing module is used for optimizing and preprocessing the running data;
the anomaly detection module is used for detecting the anomaly of the data based on the Isolation Forest algorithm;
the failure prediction module is used for predicting potential failures and potential safety hazards of the blade rotor test bed according to the established Bayesian network prediction model;
the physical space module, the twin space module and the edge gateway module are connected by Ethernet, and the physical space module and the twin space module are connected by a communication protocol.
2. A digital twin based predictive repair method for a bladed rotor test stand using the digital twin based predictive repair system according to claim 1, comprising the steps of:
s1: creating a twin space corresponding to a physical entity in a physical space;
s2: collecting operating data of the blade rotor test bed during working, and forming operating data with unified data structure and data type after data optimization and data fusion;
s3: establishing a data anomaly detection model, inputting operation data with unified data structure and data type into the model for data anomaly detection, and removing the abnormal data if the abnormal data occurs;
s4: establishing a network prediction model, predicting potential faults of the blade rotor test bed, and formulating a fault solution according to fault types;
s5: applying a fault solution to the twin space for detecting the rationality and feasibility of the solution;
s6: judging whether the fault is solved in the twin space through dismounting process simulation, a control algorithm and human-computer work efficiency;
s7: if the fault is solved, maintaining the physical entity in the physical space, otherwise, re-formulating the scheme;
in step S1, a twin space of the blade rotor test stand is established by three-dimensional modeling software and simulation software;
in step S2, the upper computer sends an acquisition instruction to the data acquisition instrument, and the data is acquired by the sensor;
in the data anomaly detection in step S3, using the Isolation Forest algorithm, by giving a string of continuous variable data sets, randomly generating an Isolation Tree, where each node on the Tree either strings two child nodes or forms a single leaf node, and then performing data detection according to the formed isolated Tree, walking the data on the isolated Tree once, and recording the leaf node where the data falls, where the data is generally rare and is quickly distributed to the leaf nodes, so as to determine whether the data is anomalous data according to the height from the leaf node to the root node of the Tree;
establishing a network prediction model in the step S4, namely, establishing a bayesian network prediction model to realize the prediction of a potential fault of the blade rotor test bed, establishing the bayesian network prediction model by obtaining historical operating data of the equipment, predicting the potential fault and potential safety hazard of the test equipment, obtaining current operating data after data processing and data anomaly detection, realizing the prediction of the overall reliability of the blade rotor test bed and the rapid positioning of the fault according to the current operating data and the bayesian network prediction model, and determining the type and time of the fault, thereby making a fault solution;
the fault types in step S4 include bearing wear, oil film temperature being too high, or rotor vibration instability fault types;
the bearing abrasion fault can determine the sequence of disassembling test equipment according to the disassembling process simulation so as to determine the bearing abrasion part;
the oil temperature can be reduced by replacing lubricating oil or reducing the rotating speed of the bearing when the temperature of the oil film is too high;
the unstable reason of the rotor vibration is the external tiny fluctuation or the unstable factor of the rotor speed, and the rotor vibration can be stabilized by adjusting the rotor speed and stabilizing the frequency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011006102.6A CN112162543B (en) | 2020-09-23 | 2020-09-23 | Blade rotor test bed predictive maintenance method and system based on digital twinning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011006102.6A CN112162543B (en) | 2020-09-23 | 2020-09-23 | Blade rotor test bed predictive maintenance method and system based on digital twinning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112162543A CN112162543A (en) | 2021-01-01 |
| CN112162543B true CN112162543B (en) | 2022-02-01 |
Family
ID=73863411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011006102.6A Active CN112162543B (en) | 2020-09-23 | 2020-09-23 | Blade rotor test bed predictive maintenance method and system based on digital twinning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112162543B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779769A (en) * | 2021-08-18 | 2021-12-10 | 国网浙江省电力有限公司舟山供电公司 | Cable test digital twinning system and working method thereof |
| CN113858146B (en) * | 2021-12-02 | 2022-05-13 | 徐州安普瑞特能源科技有限公司 | Workbench device |
| CN114970890A (en) * | 2022-04-06 | 2022-08-30 | 北京航空航天大学 | Production system cross-layer preventive maintenance decision method based on digital twin drive |
| CN118132544A (en) * | 2023-12-29 | 2024-06-04 | 宁夏大学 | Method and system for processing coal chemical industry safety diagnosis data |
| CN118244744B (en) * | 2024-05-29 | 2024-09-06 | 济南二机床集团有限公司 | Remote diagnosis and debugging system and method for stamping line |
| CN118565828A (en) * | 2024-08-02 | 2024-08-30 | 中国矿业大学 | Digital twin state monitoring method for carrier roller bearing of belt conveyor with cloud edge cooperation |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109116751A (en) * | 2018-07-24 | 2019-01-01 | 西安西电电气研究院有限责任公司 | Digitization system and its construction method based on the twin technology of number |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190317459A1 (en) * | 2018-04-13 | 2019-10-17 | Honeywell International Inc. | Predictive reactor effluent air cooler maintenance |
| CN109613895A (en) * | 2018-11-12 | 2019-04-12 | 中国电子科技集团公司第三十八研究所 | A kind of intelligence production line number twinned system |
| CN110262284B (en) * | 2019-06-24 | 2020-10-09 | 江苏科瑞德智控自动化科技有限公司 | Lithium battery pole piece rolling mill control method and system based on digital twinning technology |
| CN110492186B (en) * | 2019-07-02 | 2021-10-01 | 北京航空航天大学 | Power battery module management method based on cloud control technology |
| CN111413887B (en) * | 2020-03-17 | 2021-06-01 | 浙江大学 | Digital twin system of complex product assembly line |
| CN111475889B (en) * | 2020-03-30 | 2022-12-06 | 西安交通大学 | Physical and data-driven aeroengine digital twin modeling method |
-
2020
- 2020-09-23 CN CN202011006102.6A patent/CN112162543B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109116751A (en) * | 2018-07-24 | 2019-01-01 | 西安西电电气研究院有限责任公司 | Digitization system and its construction method based on the twin technology of number |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112162543A (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112162543B (en) | Blade rotor test bed predictive maintenance method and system based on digital twinning | |
| CN111596604B (en) | Intelligent fault diagnosis and self-healing control system and method for engineering equipment based on digital twinning | |
| CN110900307A (en) | Numerical control machine tool cutter monitoring system driven by digital twin | |
| CN110168459A (en) | Integrated digital for industrial plants is twin | |
| CN106525415B (en) | Health state evaluation system and method for transmission chain of wind turbine generator | |
| JP5151556B2 (en) | Process analysis apparatus, process analysis method, and process analysis program | |
| CN107247230A (en) | A kind of electric rotating machine state monitoring method based on SVMs and data-driven | |
| CN114723285B (en) | Power grid equipment safety evaluation prediction method | |
| CN112487910A (en) | Fault early warning method and system for nuclear turbine system | |
| US11604447B2 (en) | Condition monitoring system | |
| CN114239708B (en) | Combustion engine abnormity detection method based on quality control chart theory | |
| CN114911206A (en) | Intelligent prediction maintenance system of coal preparation plant based on Internet of things | |
| CN112418455A (en) | Equipment failure prediction and health management system | |
| CN113742855B (en) | Fault prediction method, system, electronic equipment and readable storage medium | |
| CN110160783A (en) | A kind of jacquard sliding bearing monitoring system and method based on the twin technology of number | |
| CN113420465B (en) | Hydraulic support life cycle management method based on digital twin model | |
| CN106762343A (en) | The diagnostic method of the hydraulic generator set thrust bearing failure based on online data | |
| CN115328068A (en) | Digital twinning system applied to industrial production | |
| CN117846894A (en) | Data processing method, early warning device, equipment and medium of wind turbine generator | |
| CN117872969A (en) | Intelligent electromechanical machining system based on digital twinning | |
| WO2022113494A1 (en) | Learning device, learning method, and failure prediction system | |
| CN115618735A (en) | Facility structure health monitoring method based on digital twins and related device | |
| US20230111036A1 (en) | Integrated drilling dysfunction prediction | |
| CN112836570A (en) | Equipment anomaly detection method utilizing Gaussian noise | |
| Chhabria et al. | Predictive Maintenance using Machine Learning on Industrial Water Pumps |
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 |