阅读说明：本技术 基于数字孪生的列车轴箱轴承寿命预测及故障维修系统 (Train axle box bearing life prediction and fault maintenance system based on digital twinning ) 是由 吕银辰 廖爱华 胡定玉 师蔚 于 2021-07-21 设计创作，主要内容包括：本发明涉及一种基于数字孪生的列车轴箱轴承寿命预测及故障维修系统,包括数据传输模块、数据储存及融合模块、虚拟空间、寿命预测模块、真实维修环境感知模块和信息交互模块,其中,寿命预测模块通过物理空间里的数据信息建立寿命预测模型；虚拟空间模块建立数字孪生模型、MR故障维修指导单元和故障分析与信息匹配单元；真实维修环境感知模块采集物理维修空间维修环境的数据信息；虚实融合模块用于生成混合现实维修指导解决方案；信息交互模块进行故障数据交互,实现指导维修人员进行故障维修,与现有技术相比,本发明具有提高寿命预测及故障维修的准确性、降低维修成本、解决维修现场指导资源浪费和维修人员维修知识欠缺的技术难题等优点。(The invention relates to a digital twin-based train axle box bearing service life prediction and fault maintenance system, which comprises a data transmission module, a data storage and fusion module, a virtual space, a service life prediction module, a real maintenance environment perception module and an information interaction module, wherein the service life prediction module establishes a service life prediction model through data information in a physical space; the virtual space module establishes a digital twin model, an MR fault maintenance guidance unit and a fault analysis and information matching unit; the real maintenance environment sensing module acquires data information of a physical maintenance space maintenance environment; the virtual-real fusion module is used for generating a mixed reality maintenance guidance solution; compared with the prior art, the invention has the advantages of improving the accuracy of service life prediction and fault maintenance, reducing the maintenance cost, solving the technical problems of resource waste guidance in the maintenance site and lack of maintenance knowledge of maintenance personnel, and the like.)

1. A digital twin-based train axlebox bearing life prediction and failure repair system is characterized in that the system comprises:

a data transmission module: the system is used for collecting and transmitting the axle box bearing of the train and off-line data and on-line data of train running in a physical space;

the data storage and fusion module: the system is used for receiving, storing, fusing and classifying the transmitted data;

a virtual space module: the system comprises a digital twin model, an MR fault maintenance guidance unit and a fault analysis and information matching unit, wherein the digital twin model is used for extracting information and characteristics of a train axial bearing, and the MR fault maintenance guidance unit and the fault analysis and information matching unit are fused to obtain a matching maintenance guidance solution;

a life prediction module: the device comprises three life prediction units and a data sorting and comparing unit, and is used for predicting the residual service life of the axle box bearing;

the real maintenance environment perception module: the system is used for collecting data information of a physical maintenance space maintenance environment;

and a virtual-real fusion module: to generate a mixed reality maintenance guidance solution;

the information interaction module: including the visual head-mounted device of life-span alarm unit and MR, the visual head-mounted device of life-span alarm unit is used for sending early warning information, and the visual head-mounted device of MR is used for receiving mixed reality maintenance guidance solution and alarm information.

2. The system according to claim 1, wherein the off-line data of the physical space comprises geometrical parameters, material parameters and fault defect data of the axle box bearing, and the on-line data comprises operation data of the vehicle and the axle box bearing.

3. The system for predicting the service life of the axle box bearing of the train and maintaining the fault based on the digital twin as claimed in claim 1, wherein the data transmission module comprises a sensor and a data collector, the sensor is used for collecting data required by the service life prediction module and the virtual space module, and the data collected by the data collector is transmitted to the data storage and fusion module for storage.

4. The system for predicting the service life of the axle box bearing of the train and maintaining the fault based on the digital twin as claimed in claim 1, wherein the data transmission module comprises an internet of things server and a data mapping device, the internet of things server is used for receiving the real-time state data of the axle box bearing transmitted by the data acquisition unit and preprocessing the data, and the preprocessing comprises data cleaning, feature extraction, identification and classification;

the data mapping device comprises a storage device and a database, and is used for integrating all data from modules such as a physical space and the like and carrying out deep fusion on the data on the basis, wherein the data in the database comprises sensor data, model prediction data, initial experiment data, historical life calculation data, historical geometric parameters, material parameter data and maintenance operation knowledge.

5. The system for predicting the service life and repairing the fault of the train axle box bearing based on the digital twin according to claim 1, wherein the digital twin model in the virtual space module comprises a digital twin high-fidelity three-dimensional model and a digital twin fault characteristic three-dimensional model, and information and characteristics of the train axial bearing of the two digital twin models are extracted and sent to the service life predicting module.

6. The system for predicting the service life and maintaining the fault of the axle box bearing of the train based on the digital twin as claimed in claim 5, wherein the three service life prediction units of the service life prediction module are respectively a service life prediction unit based on Lundberg-Palmgren, a theoretical service life prediction unit based on the linear damage accumulation of a miner and an empirical service life prediction unit of the bearing built based on a fatigue test of the bearing with an initial fault;

the service life prediction unit based on the Lundberg-Palmgren comprises a bearing statics analysis model and a service life model based on the Lundberg-Palmgren, and is used for solving the distribution of internal load of the bearing to obtain the relation between the internal load and deformation of the bearing, specifically, the service life of each part of the bearing is firstly solved, and then the service life L of the whole bearing is obtained₁；

The service life prediction unit based on the mini linear damage accumulation theory obtains the real dynamic load by establishing a train-bearing coupling dynamic module to obtain the residual service life L of the bearing₂；

The bearing experience life prediction unit built based on the bearing fatigue test with the initial fault carries out the fatigue test with the initial fault on the bearing, a stripped bearing experience life model is built, the test collected data are processed, the information and the characteristics of the train axial bearing extracted by the virtual space module are compared with the test collected data, the bearing model with the initial fault and the highest similarity with the high-fidelity three-dimensional model is obtained, and then the closest experience life value L is obtained₃。

7. The system for predicting the life of an axle box bearing and repairing the failure of a train based on a digital twin as claimed in claim 6, wherein the data collating and comparing unit compares the life L of the whole train axle box bearing₁Residual service life L of bearing₂And an empirical life value L₃Comparing to obtain the minimum life prediction value L_minAnd input to the life alarm unit of the information interaction module,the service life alarm unit of the information interaction module is internally provided with an alarm value L, and when the service life is predicted to be the minimum value L_minAnd when the service life is greater than or equal to L, the service life alarm unit gives an alarm.

8. The system for predicting the service life of the axle box bearing of the train and maintaining the fault based on the digital twin as claimed in claim 1, wherein the MR maintenance guidance unit of the virtual space module establishes an MR maintenance guidance platform according to a maintenance guidance technical manual, then establishes a digital three-dimensional model, then establishes a maintenance standard schematic diagram, and finally forms an MR maintenance guidance animation, and the fault analysis and information matching unit first performs information identification and classification, then establishes a fault data set, then establishes a fault ID identification number, and finally matches a fault maintenance behavior.

9. The system for predicting the service life of the axle box bearing of the train and maintaining the fault based on the digital twin as claimed in claim 1, wherein the real maintaining environment sensing module is connected with an MR visualization head-mounted device of the information interaction module, and is used for performing target identification, scene reconstruction and pose calculation according to information acquired by a camera of the MR visualization head-mounted device.

10. The system for predicting the service life of the axle box bearing of the train and maintaining the fault based on the digital twin as claimed in claim 9, wherein the virtual-real fusion module performs virtual-real fusion on the matching maintenance guidance solution in the virtual space and the results of target recognition, scene reconstruction and pose calculation of the real environment sensing module in a manner of combining natural feature registration and artificial identification registration to generate a mixed-reality maintenance guidance solution and transmits the mixed-reality maintenance guidance solution to the MR visualization head-mounted device to guide maintenance personnel to perform fault maintenance.

Technical Field

The invention relates to the technical field of trains, in particular to a digital twin-based train axle box bearing service life prediction and fault maintenance system.

Background

Along with the acceleration of the Chinese train, the performance requirements on train parts are further improved. The axle box bearing is one of key components of a train running gear, and the performance of the axle box bearing directly influences the running safety of a train. Due to the special service environment of the axle box bearing, the axle box bearing generates failure modes such as contact fatigue, plastic deformation, retainer fracture and the like, and fracture damage occurs in severe cases, which can cause train derailment and serious accidents. Therefore, the method has very important significance for service life prediction and fault maintenance of the axle box bearing of the train.

The existing service life prediction method estimates the fatigue of a bearing under a certain load, and the estimation has a certain difference with the practical bearing application working condition, so that the working condition of a single axle box bearing cannot be deeply known and reduced, and the fatigue life prediction result is inconsistent with the practical condition; and the on-site train axle box bearing fault maintenance is time-consuming and labor-consuming, so that the dynamic load of the bearing which is closer to the reality is obtained by analyzing the actual working condition of the train-track, the fatigue life of the bearing is more accurately estimated, and an MR auxiliary guidance system is introduced, so that the technical problems of resource waste and maintenance knowledge shortage of maintenance personnel in the maintenance site are solved.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a digital twin-based train axle box bearing life prediction and fault repair system.

The purpose of the invention can be realized by the following technical scheme:

a digital twin-based train axlebox bearing life prediction and troubleshooting system, the system comprising:

a data transmission module: the system is used for collecting and transmitting the axle box bearing of the train and off-line data and on-line data of train running in a physical space;

the data storage and fusion module: the system is used for receiving, storing, fusing and classifying the transmitted data;

a virtual space module: the system comprises a digital twin model, an MR fault maintenance guidance unit and a fault analysis and information matching unit, wherein the digital twin model is used for extracting information and characteristics of a train axial bearing, and the MR fault maintenance guidance unit and the fault analysis and information matching unit are fused to obtain a matching maintenance guidance solution;

a life prediction module: the device comprises three life prediction units and a data sorting and comparing unit, and is used for predicting the residual service life of the axle box bearing;

the real maintenance environment perception module: the system is used for collecting data information of a physical maintenance space maintenance environment;

and a virtual-real fusion module: to generate a mixed reality maintenance guidance solution;

the information interaction module: including the visual head-mounted device of life-span alarm unit and MR, the visual head-mounted device of life-span alarm unit is used for sending early warning information, and the visual head-mounted device of MR is used for receiving mixed reality maintenance guidance solution and alarm information.

The off-line data of the physical space comprises geometric parameters, material parameters and fault defect data of the axle box bearing of the train, and the on-line data comprises operation data of the vehicle and the axle box bearing.

The data transmission module comprises a sensor and a data acquisition unit, the sensor is used for acquiring data required by the service life prediction module and the virtual space module, and the data acquired by the data acquisition unit is transmitted to the data storage and fusion module for storage.

The data transmission module comprises an Internet of things server and a data mapping device, wherein the Internet of things server is used for receiving real-time state data of the axle box bearing transmitted by the data acquisition unit and preprocessing the data, including data cleaning, feature extraction, identification and classification;

the data mapping device comprises a storage device and a database, and is used for integrating all data from modules such as a physical space and the like and carrying out deep fusion on the data on the basis, wherein the data in the database comprises sensor data, model prediction data, initial experiment data, historical life calculation data, historical geometric parameters, material parameter data and maintenance operation knowledge.

The digital twin model in the virtual space module comprises a digital twin high-fidelity three-dimensional model and a digital twin fault characteristic three-dimensional model, and information and characteristics of the train axial bearings of the two digital twin models are extracted and sent to the service life prediction module.

The three service life prediction units of the service life prediction module are respectively a service life prediction unit based on Lundberg-Palmgren, a linear damage accumulation theoretical service life prediction unit based on a mini and a bearing empirical service life prediction unit built based on a bearing fatigue test with an initial fault;

the service life prediction unit based on the Lundberg-Palmgren comprises a bearing statics analysis model and a service life model based on the Lundberg-Palmgren, and is used for solving the distribution of internal load of the bearing to obtain the relation between the internal load and deformation of the bearing, specifically, the service life of each part of the bearing is firstly solved, and then the service life L of the whole bearing is obtained₁；

The service life prediction unit based on the mini linear damage accumulation theory obtains the real dynamic load by establishing a train-bearing coupling dynamic module to obtain the residual service life L of the bearing₂；

The bearing experience life prediction unit built based on the bearing fatigue test with the initial fault carries out the fatigue test with the initial fault on the bearing, a stripped bearing experience life model is built, the test collected data are processed, the information and the characteristics of the train axial bearing extracted by the virtual space module are compared with the test collected data, the bearing model with the initial fault and the highest similarity with the high-fidelity three-dimensional model is obtained, and then the closest experience life value L is obtained₃；

Said numberAccording to the arrangement comparison unit, the service life L of the whole bearing₁Residual service life L of bearing₂And an empirical life value L₃Comparing to obtain the minimum life prediction value L_minAnd inputting the service life of the information interaction module into an alarm unit of the information interaction module, wherein an alarm value L is arranged in the alarm unit of the service life of the information interaction module, and when the minimum value L of the service life is predicted_minAnd when the service life is greater than or equal to L, the service life alarm unit gives an alarm.

The MR maintenance guidance unit of the virtual space module establishes an MR maintenance guidance platform according to a maintenance guidance technical manual, then establishes a digital three-dimensional model, further establishes a maintenance standard schematic diagram, and finally forms an MR maintenance guidance animation.

The real maintenance environment sensing module is connected with the MR visual head-mounted device of the information interaction module and used for carrying out target identification, scene reconstruction and pose calculation according to information collected by a camera of the MR visual head-mounted device.

And the virtual-real fusion module performs virtual-real fusion on the matching maintenance guidance solution in the virtual space and the results of target identification, scene reconstruction and pose calculation of the real environment perception module in a mode of combining natural feature registration and artificial identification registration to generate a mixed reality maintenance guidance solution and transmits the mixed reality maintenance guidance solution to the MR visual head-mounted device to guide maintenance personnel to perform fault maintenance.

Compared with the prior art, the invention has the following advantages:

firstly, under the background of a digital twin technology, a service life prediction module is combined with three service life prediction units at the same time, various factors influencing the service life of a bearing are fully considered, and the reliability and the accuracy of the service life prediction of the bearing are improved.

The system is introduced when the axle box bearing is in fault maintenance, maintenance personnel can complete maintenance operation according to a maintenance guide scheme given by the system, the maintenance personnel do not need to have professional maintenance knowledge and watch a maintenance technical manual, and the technical problems of resource waste guidance and maintenance knowledge shortage of the maintenance personnel on a maintenance site are solved.

The service life prediction model established by the train vehicle and axial bearing data acquired by the sensor in real time has real-time performance, strong adaptability to trains running on different lines, more reliable prediction result and quicker feedback.

Drawings

FIG. 1 is a schematic diagram of a physical space;

FIG. 2 is a schematic diagram of a data transmission module;

FIG. 3 is a schematic diagram of a data storage and fusion module;

FIG. 4 is a schematic diagram of a virtual space module;

FIG. 5 is a schematic diagram of a bench test principle;

FIG. 6 is a schematic diagram of a life prediction module;

FIG. 7 is a schematic diagram of a connection structure between a life prediction module and an information interaction module;

fig. 8 is a schematic diagram of a connection structure of the virtual-real fusion module and the virtual space and real maintenance environment sensing module.

FIG. 9 is a schematic diagram of the system of the present invention

Detailed Description

The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.

Examples

As shown in FIG. 9, the invention provides a train axle box bearing service life prediction and fault maintenance system based on digital twin, which comprises a data transmission module, a data storage and analysis module, a service life prediction module, a virtual space module, a real maintenance environment sensing module, a virtual-real combination module, an information interaction module and other modules, wherein the system is used for realizing the timely prediction of the residual service life of an axle box bearing, converting digital information into image information by using a digital twin technology, displaying the image information in a three-dimensional and visual manner, extracting information characteristics of an axial bearing of a train, building an MR fault maintenance guidance unit and a fault analysis and information matching unit, fusing the two units to obtain a matching maintenance guidance solution, transmitting the matching maintenance guidance solution to an MR visual head-mounted device for displaying, providing a train axle box bearing maintenance scheme for maintenance personnel, and connecting the service life prediction module with a virtual space, an initial fault bearing experience life model is established in the life prediction module, an axle box bearing dynamic high-fidelity model and a fault characteristic model are established in a virtual space to acquire model information and characteristics, running state monitoring of the axle box bearing is completed, monitoring data and experiment acquisition data are compared, then the closest experience life value is obtained, finally the closest experience life value is transmitted to the man-machine interaction module through the information interaction module, and finally a maintenance worker makes a response according to the information obtained by the maintenance worker.

As shown in fig. 1, the physical space is divided into offline data and online data, where the offline data includes journal bearing geometric parameters, material parameters, and journal bearing fault defect data, and specifically, the journal bearing geometric parameters includes bearing inner and outer diameters, bearing width, two rows of axial distances, roller effective length, outer ring contact angle, and other data; the journal box bearing material parameters comprise density, Poisson's ratio, yield strength and the like, and the journal box bearing fault defect data comprise data of inner and outer ring peeling, roller peeling and the like. The online data comprises the running data of the vehicle and the axle box bearing, specifically the data of vehicle load, axle box bearing rotating speed, temperature, train running track spectrum and the like.

As shown in fig. 2, the data transmission module is used for acquiring data required by the service life prediction module and the virtual space, and mainly comprises a sensor and a data acquisition unit, the sensor specifically comprises a displacement sensor, a vibration acceleration sensor, a revolution speed sensor, a temperature sensor and other devices, wherein the displacement sensor is used for acquiring vehicle displacement, framework displacement, axial displacement and the like, the vibration acceleration sensor is used for acquiring vehicle vertical acceleration, framework vertical acceleration, axle box vertical acceleration and the like, the revolution speed sensor is used for measuring the revolution speeds of bearings and wheels, the temperature sensor is used for acquiring bearing temperature information, and the information is acquired by the data acquisition unit and then transmitted to the internet of things server and the database of the data fusion and storage module for processing.

As shown in fig. 3, the data fusion and storage module includes an internet of things server and a data mapping device, where the internet of things server is used to receive real-time status data of the axle box bearing transmitted by the data collector and preprocess the data, that is, data cleaning, feature extraction and identification, classification, and the like; the data mapping device comprises a storage device and a database, is mainly used for integrating all data from modules such as a physical space and the like, carries out deep fusion on the data on the basis, comprises sensor data, all historical data (including historical fault data and historical geometric parameters) of the axle box bearing, model prediction data, material parameters, initial experiment data, historical life calculation data, maintenance operation knowledge and the like, and keeps the consistency, integrity and real-time performance of the data.

As shown in fig. 4, a digital twin model is established in the virtual space, including bearing fault feature models and bearing high-fidelity three-dimensional models of all fault forms, and information features of the high-fidelity models and the fault feature models are collected in the virtual space and transmitted to the service life prediction module. In addition, the virtual space also comprises an MR maintenance guidance unit and a fault analysis and information matching unit, the MR maintenance guidance unit establishes an MR maintenance guidance platform according to a maintenance guidance technical manual, further establishes a digital three-dimensional model, further establishes a maintenance standard schematic diagram, and finally forms an MR maintenance guidance animation; the fault analysis and information matching unit firstly carries out information identification and classification, then establishes a fault data set, then establishes a fault ID identification number and finally matches fault maintenance behaviors. And finally, fusing the MR fault maintenance guidance unit and the fault analysis and information matching unit in the virtual space to obtain a matching maintenance guidance solution, and storing all data into a data fusion and storage module.

As shown in fig. 6 and 7, the life prediction module requests data from the data fusion and storage module and then performs parametric modeling, and three life prediction units are introduced, namely a Lundberg-Palmgren life model and a mini-based linear damage modelThe theoretical life model is accumulated, the bearing experience life model is built by the initial bearing residual life evaluation method, life analysis is carried out in an all-round mode, and the life prediction reliability is improved. Wherein the three life prediction units obtain three life prediction information L₁、L₂、L₃The minimum value L of the three is sorted and compared by the data sorting and comparing unit_minInputting a service life alarm unit in the information interaction module, wherein the information interaction module comprises a service life alarm unit in which a bearing service life alarm value L is set, and when the L is input by the service life prediction module_minAnd when the service life is less than or equal to L, the service life early warning unit gives an alarm.

The method comprises the steps of considering the surface of a raceway more based on a Lundberg-Palmgren life model, establishing a bearing statics model in ANSYS (American society of dynamics) by taking bearing load as pure radial load, pure axial load or combination of the pure radial load and the pure axial load, solving distribution of the load in the bearing to obtain the relation between the load in the bearing and deformation, introducing a probability statistics method, firstly obtaining the service life of each part of the bearing, and then obtaining the service life L of the whole bearing₁。

A life prediction unit based on a mini's linear damage accumulation theory establishes a train-bearing coupling dynamics module, in the process of simplifying a dynamics system, a train body, a framework, an axle box, a wheel pair and the like are regarded as rigid bodies, the mass center of the rigid body is regarded as a geometric center, a suspension device connecting all the rigid bodies is simplified into linear springs and damping, and in the actual working process of the axle box bearing, the inner ring and the axle, the outer ring and the axle box are in interference fit. In order to accurately analyze the internal load of the axle box bearing, the inner ring and the axle are regarded as a mass whole, the outer ring and the axle box are regarded as a mass whole, and the internal load is transmitted between the inner ring and the axle box through the rollers. According to the internal load and deformation relationship found in the ANSYS static model, the contact between the roller and the inner and outer races can be regarded as a spring damping system. Calculating the bearing life under each dynamic load after obtaining the dynamic load, and assuming front and back loads F in the load time course₁And F₂Is linearly changed, and the load step length is t₀To calculate the equivalent load over this period of time, Pa is introducedObtaining the equivalent load F according to the lmgren-Miner rule_mCalculating the residual service life L of the bearing by using an ISO281 correction service life calculation method₂。

Based on a bearing fatigue test with an initial fault, a bearing experience life prediction unit is built, the relation between the vibration characteristic and the temperature characteristic of the bearing with the initial fault and the fault characteristic and the residual service life of the bearing can be obtained, and the residual service life of the bearing can be predicted more reliably. For a bearing test with an initial fault, building a bearing experience life model by using collected vibration signals, temperature signals and a defect size change rule, processing data collected by the test, connecting a life prediction module with a virtual space, building a digital twin model in the virtual space, extracting information and characteristics of the built axle box bearing dynamic high-fidelity model and a fault characteristic model in the virtual space, monitoring the running state of the axle box bearing and comparing the data collected by the test to obtain an initial fault bearing model with the highest similarity with the high-fidelity three-dimensional model, and further obtaining the closest experience life value L₃。

As shown in fig. 5, the figure is a simplified diagram of a bench test principle, a bearing fatigue test with an initial fault is carried out, the bench adopts a hydraulic system to carry out radial and axial load loading on the bearing, the radial loading is up to 120KN, the axial loading is up to 10KN, the shaft diameter is 85-340 mm, a main shaft is dragged and rotated by a motor, and the maximum rotating speed reaches 1500 r/min.

As shown in fig. 8, the information interaction module comprises a MR visualization head-mounted device, and the real maintenance environment perception module is further connected with the MR visualization head-mounted device of the information interaction module, the system is used for carrying out target identification, scene reconstruction and pose calculation according to information acquired by a camera of the MR visual head-mounted device, a virtual-real fusion module acquires that an MR fault maintenance guidance unit and a fault analysis and information matching unit are fused in a virtual space to obtain a matching maintenance guidance solution, and obtaining the results of target recognition, scene reconstruction and pose calculation of the environment sensing module, performing virtual-real fusion by the virtual-real fusion module through a method of combining natural feature registration and artificial identification registration to obtain fusion maintenance information, and the train axle box bearing is transmitted to an MR visual head-mounted device for displaying, and a train axle box bearing maintenance scheme is provided for maintenance personnel.

While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.