CN112163325B - RV reducer service life prediction method based on digital twinning - Google Patents
RV reducer service life prediction method based on digital twinning Download PDFInfo
- Publication number
- CN112163325B CN112163325B CN202010947950.0A CN202010947950A CN112163325B CN 112163325 B CN112163325 B CN 112163325B CN 202010947950 A CN202010947950 A CN 202010947950A CN 112163325 B CN112163325 B CN 112163325B
- Authority
- CN
- China
- Prior art keywords
- reducer
- model
- data
- simulation
- real
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16H—GEARING
- F16H57/00—General details of gearing
- F16H57/02—Gearboxes; Mounting gearing therein
- F16H57/023—Mounting or installation of gears or shafts in the gearboxes, e.g. methods or means for assembly
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16H—GEARING
- F16H57/00—General details of gearing
- F16H57/08—General details of gearing of gearings with members having orbital motion
- F16H57/082—Planet carriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/25—Design optimisation, verification or simulation using particle-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Geometry (AREA)
- Human Resources & Organizations (AREA)
- Mechanical Engineering (AREA)
- Computer Hardware Design (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Biomedical Technology (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
RV reducer service life prediction method based on digital twinning comprises the following steps: (1) establishing a data acquisition system: the sensor is used for collecting data of working conditions and environmental parameters when the RV reducer actually works; (2) establishing a digital twin model of the RV reducer: establishing a digital twin model of the RV reducer by establishing each submodel of the digital twin model, so as to ensure the comprehensive analysis of real-time data; (3) predicting service life of RV reducer: the sub-model can obtain the predicted service life of the RV reducer through simulation; (4) dynamic real-time monitoring and feedback: the actual running condition of the RV reducer is monitored through real-time acquisition of data, and the reliability of twin model prediction is guaranteed. According to the invention, the virtual simulation technology and the agent model are combined to study the use of the RV reducer, so that the quick and accurate prediction of the service life of the RV reducer is realized, the experiment of a physical prototype is not needed, the study cost and time are saved, and the study efficiency is greatly improved.
Description
Technical Field
The invention relates to the field of mechanical life prediction and digital twin specific application.
Background
RV (Rotary Vector) is a novel transmission mode developed on the basis of cycloidal pin gear transmission, and the RV reducer is a precise reducer with high precision, has a structure of a two-stage crank type closed differential gear train, is widely applied to high-tech fields such as numerical control machine tools, aerospace, particularly industrial robot industries, and has the advantages of small size, light weight, compact structure, large transmission ratio range, large bearing capacity, high motion precision, high transmission efficiency and the like.
The service life is an important performance parameter of the RV reducer, the service life of the RV reducer is the place where the difference between the current country and the foreign country is the most obvious and the greatest, the accuracy of 5000 hours or even ten thousand hours can be ensured by using the 40E reducer, the finished product in China is only about 1000 hours, and the performance of each aspect can be obviously reduced. At present, compared with foreign RV reducers, the domestic RV reducer has a large gap, and is mainly characterized by short service life, poor transmission precision and the like. Therefore, the reliability life is one of the key factors restricting the development of domestic robots.
The RV reducers are very complex in structure, making prediction of their service life very difficult. The existing prediction method for the RV reducer is divided into the following 3 types:
the 3 modes have certain limitations, the first method can cause larger errors of the actual service life and the predicted service life, the second method is overlong in period, the third method combines the characteristics of the first two methods, the actual service life of the RV reducer can be predicted quickly and accurately, but the high cost is achieved, meanwhile, the interference of various external factors on the actual use of the RV reducer cannot be considered, and the defects still exist.
The life of an RV retarder is related to the fatigue characteristics and thermal behavior of its components. Under constant contact load, fatigue failure can occur to parts of the RV reducer along with the progress of work. Fatigue failure of the parts can affect the overall life of the RV retarder. Parts of the RV reducer include input shafts, planets, sun gears, crankshafts, needle bearings, cycloidal gears, pin teeth, etc., wherein the planets, cycloidal gears and pin teeth are key components that cause the RV reducer to fail, and the crankshaft is next (Yao Canjiang, wei Linghui, wang Hailong. FTA and FMEA based reliability analysis of RV reducers [ J ]. Modern manufacturing engineering, 2018 (01): 136-140.). It can be said that the fatigue life of the critical parts determines the overall service life of the RV retarder.
The study shows that the failure modes and reasons of the planet gears and the cycloidal gears of the RV reducer are as follows:
for fatigue characteristic analysis of mechanical parts, geometric information, load spectrum, S-N curve of materials and stress distribution rule of parts of an object model (Zhang hong RV reducer dynamics modeling and fatigue optimization analysis [ D ]. Nanjing aviation aerospace university, 2019.) need to be obtained first.
The problem that the theory and reality are separated from each other in the existing RV reducer prediction technology is found, manufacturing errors are not considered in the method 1 for predicting the service life of the RV reducer, the service life of the RV reducer is not analyzed by combining with the working condition of the actual RV reducer, a prediction result is obtained by using theoretical parameters for simulation, and the accuracy of the prediction result is low; the method 2 and the method 3 only obtain the prediction result through field feedback or experimental data, and are long in time and high in cost and cannot meet actual production requirements for RV speed reducers of various types. The prediction method does not integrate multi-source data well to obtain more accurate prediction data, the service life prediction research of the RV reducer can be expanded to the fatigue life research of the RV reducer under the condition of matching all parts, related life parameters are solved through a simulation method, the model is used for prediction, and a more accurate life prediction value is obtained through real-time updating of the data.
The digital twinning technology is a technical means integrating the characteristics of multiple physics, multiple scales and multiple disciplines, and can well solve the problem of physical and virtual separation by establishing a virtual model with high fidelity and real-time synchronization characteristics to simulate, simulate and feed back the condition of a physical entity, and achieve accurate prediction effects through the acquisition and simulation analysis of physical data (Tao Fei, liu Weiran, liu Jianhua, liu Xiaojun, liu Jiang, qu Ting, hu Tianliang, zhang Zhina, peak directions, xu Wenjun, wang Junjiang, zhang Yingfeng, liu Zhenyu, li Hao, cheng Jiangfeng, qi Qinglin, zhang Meng, zhang He, inert aromatic substances, he Lirong, yi Wangmin, cheng Hui, digital twinning and application exploration [ J ]. Computer integrated manufacturing system 2018,24 (01): 1-18.
According to the method for predicting the service life of the RV reducer based on digital twinning, actual use data of the RV reducer is collected by using a sensor technology, a digital twinning model of the RV reducer is built according to a 1:1 size, effective information is processed by the data through a wavelet transformation method, and the processed effective information is input into each sub-model in the digital twinning model, wherein the sub-models comprise a geometric model, a dynamic model, a thermal model and a stress distribution model. The digital twin model obtains operation data of the next time period through dynamic simulation, and the data is subjected to fatigue simulation to obtain the fatigue life of the RV reducer, so that the service life of the RV reducer can be rapidly predicted. The simulated input and output data can be used for predicting the fatigue life of the key parts of the RV reducer by using the Kriging proxy model, the accuracy of the model is improved by continuously inputting the data, meanwhile, the reliability of a simulation result can be checked by collecting the RV reducer data in the next time period, and further, the accuracy of the life prediction of the RV reducer can be gradually improved by continuously improving the model. The fatigue model comparison analysis generated by the digital twin model of the RV reducer with different parameters is combined with the use of a particle swarm algorithm to obtain the RV reducer parameter combination with prolonged service life, so that the RV reducer parameters are optimally designed, and a basis is provided for the optimization of the RV reducer.
Disclosure of Invention
The invention aims to introduce a digital twin technology into the prediction of the service life of the RV reducer, improve the defects of the conventional prediction method of the service life of the RV reducer, and combine the digital twin technology to realize more accurate prediction of the service life of the RV reducer.
In order to achieve the purpose of the invention, the technical scheme adopted by the invention is as follows:
RV reducer service life prediction method based on digital twinning comprises the following steps:
(1) Establishing a data acquisition system: the method comprises the steps of acquiring data of working conditions and environmental parameters of the RV reducer in actual working by using a sensor, wherein the data comprise data of geometrical shapes, material properties, running environmental parameters, torque, return difference, disturbance and the like of all parts of the RV reducer, including a torque sensor, a vibration sensor, a temperature sensor and the like;
(2) Establishing a digital twin model of the RV reducer: and establishing a digital twin model of the RV reducer by establishing each submodel of the digital twin model, so as to ensure the comprehensive analysis of real-time data. The submodules of the digital twin model comprise a geometric model, a dynamics model, a thermal model and a stress distribution model. And acquiring real-time data through a data acquisition system, and uploading the data to the twin model. Processing the uploaded data, and performing simulation in each sub-model;
(3) RV reduction gear life prediction: the sub-model can obtain the predicted service life of the RV reducer through simulation. By recording the prediction conditions of the input parameters and the output life values of the RV reducer, after the RV reducer runs for a plurality of times, the life prediction can be performed by using a kriging proxy model instead of a simulation model. According to the data prediction of the sub-model operation and the actual operation, updating the data to each sub-model of the digital twin model in time through updating iteration of the data, and obtaining an accurate service life value of the RV reducer;
(4) Dynamic real-time monitoring and feedback: the actual running condition of the RV reducer is monitored through real-time acquisition of data, and the reliability of twin model prediction can be ensured. The actual running condition of the RV reducer can synchronize the twin model and the physical model in a data mode, and the simulation data and the actual data are integrated through one-time simulation to accurately monitor and predict the running condition of the RV reducer. The optimized RV reducer parameter combination can be obtained by applying the particle swarm algorithm, the service life of the RV reducer can be prolonged, and a data basis is provided for the design of the RV reducer.
Further, the step (1) of establishing a data acquisition system comprises the following processes:
(11) And a sensor is additionally arranged in the RV reducer and used for recording dynamic data of the RV reducer in real time when the RV reducer is used. The sensor comprises the following types and uses:
(12) The signals generated by the sensors in actual use of the RV reducer are converted into spectrograms by extracting useful information through wavelet transformation technology;
compared with Fourier transformation, the time of each component can be obtained by utilizing wavelet transformation to receive the signal, the condition that the frequency of the signal changes along with the time is known, and the instantaneous frequency and the amplitude of each moment are analyzed in real time. When the signal is suddenly changed, the wavelet transformation can accurately show the change and reflect the change to a time frequency spectrum, so that the failure condition of each part of the RV reducer can be conveniently found.
(13) Uploading and classifying data;
the step (2) of establishing a digital twin model of the RV reducer comprises the following steps:
the digital twin model comprises a plurality of sub-models, namely a geometric model, a dynamics model, a thermal model, a stress analysis model and a fatigue model which are the same as the sizes of the parts of the physical model;
(21) Establishing a geometric model and a kinetic model of the RV reducer: modeling software is used for 1:1 modeling of the RV reducer, so that mapping can be formed between the RV reducer and the RV reducer entity, and simulation calculation is better carried out;
wherein, the step (22) of establishing the geometric model and the dynamic model comprises the following procedures:
(211) Collecting and uploading real-time data;
(212) Drawing a 3D model of the RV reducer according to real-time data 1:1 by using SolidWorks software to form a geometric model;
the parameter equation of the cycloidal gear tooth profile is as follows:
K 1 -short-amplitude coefficients;
e-eccentricity of cycloidal gear;
i H -cycloidal pin gear ratio;
-the angle through which the crank shaft rotates relative to the central sagittal axis of a needle tooth;
r p -the centre circle radius of the needle teeth;
r rp -needle tooth radius;
the data are all the data actually measured by the sensor.
(213) Introducing key parts of the RV reducer in the geometric model into an ANYSYS, selecting units, defining materials, dividing a network, and defining external nodes and a rigid region;
(214) The geometric model is led into ADAMS software, the key parts are replaced by partial flexible bodies, cycloidal gears, pin teeth, crankshafts and a planet carrier are defined as flexible bodies, the rest are set as rigid bodies, and constraint is added to form a dynamic model.
After SolidWorks modeling, the SolidWorks is stored into an intermediate format file: x_t is the pamasolid format, and then imported into ADAMS.
The material properties of each component are determined by the actual RV reducer material, taking RV-40E model as an example, and the material properties of each component are shown in the following table:
constraints of each component are added, as shown in the following table:
(215) Carrying out dynamics simulation according to the real-time measured input torque, return difference and other data of the RV reducer;
(216) And outputting a load spectrum of an output shaft of the RV reducer. Storing the real-time data, the simulation data and other necessary data, and preparing for the next data analysis;
(22) Establishing a RV reducer thermal model: and establishing different gear temperature fields of the RV reducer according to different gear load spectrums obtained by ADAMS simulation. Analyzing the influence of parameters such as geometric parameters, load, lubrication characteristics and the like of each gear on the temperature field of each gear;
(23) Establishing a stress analysis model of the RV reducer: according to the load spectrums of different parts obtained by ADAMS simulation of the stress sensor and the obtained data, the stress and stress distribution of each point in the object such as the mechanical part, the component and the like are analyzed and solved, and the stress concentration of dangerous points related to the failure of the mechanical part and the component and the peak stress and strain of the strain concentration part are determined.
(24) Establishing a fatigue model of the RV reducer: the fatigue model is established by combining the digital twin model and the actual RV reducer operation data, so that the actual fatigue state of the RV reducer is accurately reflected, and the future fatigue state is estimated.
(25) The data among the models supplement each other, meanwhile, the historical data are compared with the real-time data, when the data are different, errors are marked, and the data are displayed on an operation interface for processing;
the service life prediction of the RV reducer in the step (3) comprises the following steps:
(31) Inputting simulation results of each model in the digital twin model into MSC.Fatigue simulation software;
(32) The fatigue life simulation operation of the RV reducer outputs damage data of all parts, and predicts the fatigue life of all parts of the RV reducer;
(33) Comparing the failure standards of the RV reducer parts, and once one part fails, the RV reducer fails to reach the service life maximum value;
(34) Inputting the result into a damage evolution model;
(35) Gradually constructing a life prediction function of the RV reducer by adopting a Kriging proxy model, and establishing a high-precision life prediction proxy model of the RV reducer by continuously collecting input data and output data;
compared with a simulation model, the agent model has a plurality of advantages, firstly, the agent model can replace a complex and time-consuming numerical analysis model, secondly, the agent model can ensure that sample points are accurately converged to a real solution on the basis of historical data, and the agent model can ensure high approximation accuracy in an important area, particularly a real solution area and efficiently obtain a target solution in the face of a complex multidimensional problem.
(36) When the reliability of the RV reducer life prediction proxy model meets the requirement, the proxy model can be directly used for predicting the RV reducer life, so that the simulation prediction time is saved;
the step (4) of dynamic real-time monitoring and feedback comprises the following processes:
(41) Recording real-time data;
(42) Recording simulation data;
(43) Recording of prediction data;
(44) Comparing the simulation calculation data with the real-time data;
(45) Comparing the simulation prediction data with real-time data;
(46) According to the different service lives of RV speed reducers with different parameters under different working conditions and environments, various parameters of the RV speed reducers are optimized by adopting a particle swarm algorithm to improve the service lives, the best RV speed reducer parameter suggestion can be given under a specific environment, and data reference is provided for the RV speed reducer parameter design.
The step (46) of parameter optimization design comprises the following steps:
(461) Determining an influence parameter and a value range thereof;
(462) Substituting the service life result into a particle swarm algorithm, and applying a kriging model to calculate the service life result;
(1) initializing a population
The dimension (array element number) of the particle swarm is 11, and the particle swarm is respectively a working load, a pin tooth radius, a pin tooth center circle radius, a crank axle eccentric distance, a cycloid wheel moving distance shape correction amount, a cycloid wheel equidistant shape correction amount, a cycloid wheel inner hole and rotating arm shaft sleeve gap, a crank axle and supporting shaft sleeve gap, an upper supporting shaft sleeve and planet carrier gap, a lower supporting shaft sleeve and a gland gap;
particle population size N: taking a general value of 200;
for each particle, there are two attributes:
position attribute:
X i (t)＝{x i,1 (t),x i,2 (t),Λ,x i,j (t)}； (2)
speed attribute:
V i (t)＝{v i,1 (t),v i,2 (t),Λ,v i,j (t)}； (3)
where t represents the t iteration (t meeting), i represents that the number of this particle is i, j represents the dimension of the search space, j=2 (searching in the plane) for the particle;
(2) calculating individual and global optimum
The optimal position searched for by the ith particle so far is called the individual extremum:
P best ＝(p i1 ,p i2 ,Λ,p ij ),i＝1,2,ΛN (4)
the optimal position searched so far for the whole population of particles is the global extremum:
g best ＝(g 1 ,g 2 ,Λ,g j ) (5)
(3) after finding the two optimal values, the particle swarm updates its own speed and position
The speed update formula:
wherein c 1 Is a self-learning factor, c 2 Is global learning factor, which is constant factor, taking c 1 ＝c 2 ＝2.5；r 1 ,r 2 Is [0,1]The random number in the range is uniform, so that the randomness of the particle flight is increased;is the particle velocity, and it was found that v was set max And the effect of adjusting the inertial weight is equivalent, so v max The method is generally used for initializing and setting, and takes the value of 10; lambda is a compression factor for controlling the final convergence and increasing the convergence rate, wherein +.>
The location update formula:
x ij (t+1)＝x ij +v ij (t+1) (8)
(4) it is determined whether convergence criteria for the algorithm are met. If yes, the operation is ended; if not, go to (2).
The RV reducer service life prediction method based on digital twinning has the following beneficial effects:
1. the method solves the problems of low service life prediction precision, high cost and long duration of the existing RV reducer, and provides a new method for further improving the service life prediction of the RV reducer.
2. The modeling simulation avoids measurement errors in the artificial measurement experiment process, and real-time data is adopted for processing and analysis, so that the prediction result is more accurate.
3. According to the method, the virtual simulation technology and the agent model are combined to study the use of the RV reducer, so that the service life of the RV reducer is rapidly and accurately predicted, a test of a physical prototype is not needed, study cost and time are saved, and study efficiency is greatly improved.
Drawings
FIG. 1 is a parametric model diagram of various parts of the RV reducer of the present invention; FIG. 1-a is a planet carrier, FIG. 1-b is a gland, FIG. 1-c is a pin housing, FIG. 1-d is a planet wheel, FIG. 1-e is a sun wheel, FIG. 1-f is a pin, FIG. 1-g is a crankshaft, FIG. 1-h is a cycloid wheel, FIG. 1-i is a cylindrical roller bearing (upper bearing sleeve), FIG. 1-j is a tapered roller bearing (lower bearing sleeve), and FIG. 1-k is a main bearing;
FIG. 2 is a three-dimensional model of the RV reducer of the present invention;
FIG. 3 is an exploded view of the RV reducer of the present invention;
FIG. 4 is a diagram of a step of predicting RV reducer life according to the present invention;
FIG. 5 is a flow chart of the invention for creating a dynamic simulation model;
FIG. 6 is a schematic diagram of the dynamics simulation process of the present invention;
FIG. 7 is a kinetic simulation result of the present invention;
FIG. 8 is a flow chart of life prediction by creating a proxy model in accordance with the present invention;
FIG. 9 is a flow chart of optimization of the particle swarm algorithm of the present invention;
FIG. 10 is a graphical representation of the service life prediction technique of the RV reducer of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
The invention aims to design a method for predicting the service life of an RV reducer based on digital twinning, overcomes the defects of the conventional method for predicting the service life of the RV reducer, and combines the digital twinning technology to perform more accurate service life prediction on the RV reducer.
FIG. 5 is a flow chart for creating a simulation model, comprising the steps of:
(221) Drawing a 3D model of the RV reducer by using SolidWorks software 1:1 to form a geometric model;
(222) Key parts are imported into ANYSYS, units are selected, materials are defined, networks are divided, and external nodes and rigid areas are defined;
(223) The simulation model is led into ADAMS software, the key parts are replaced by partial flexible bodies, the cycloid gear, the pin gear, the crank shaft and the planet carrier are defined as flexible bodies, and the rest are set as rigid bodies to form the dynamic model.
FIG. 8 is a flowchart of a kriging proxy model life prediction. Firstly, after a response value is obtained through multiple dynamics analysis and life analysis in a digital twin model, substituting sensor data and the response value into a kriging proxy model for curve fitting, and establishing an initial proxy model; and then, the accuracy of the proxy model is improved in the optimization of the historical data and the proxy model, on one hand, the acquisition of the real-time data of the sensor can input more data into the proxy model, and the model is continuously corrected. And on the other hand, the parameter combinations around the fitting curve are sampled through hypercube sampling, and the parameter combinations are substituted into a dynamic model in the digital twin model for analog simulation, so that a response value is obtained, and the accuracy of the proxy model is further improved. Finally, the high-precision proxy model can rapidly predict the service life of the RV reducer, and a solving method is also provided for the realization of the algorithm in the next feedback.
FIG. 9 is a roadmap of the RV reducer life prediction technique, with the overall prediction process being a closed loop. And generating a corresponding dynamic model, a thermal model and a stress model through real-time data and working conditions of the physical model, filtering data received by a sensor, transmitting the data to a multi-body dynamic model, receiving working condition data measured by the physical model by the multi-body dynamic model, simulating by combining the filtered real-time part size to obtain predicted simulation data, entering a prediction stage, predicting the service life of the RV reducer by combining service life analysis software and a kriging proxy model, comparing a predicted value with an actual measured value, and updating a digital twin model. And finally, guiding the operation of the physical model, and optimally designing the RV reducer. In fig. 9, the sensor data includes: working load, needle tooth radius, needle tooth center circle radius, crank axle eccentric distance, cycloid gear shift correction, cycloid gear equidistant correction, gap between cycloid gear inner hole and rotary arm shaft sleeve, gap between crank axle and support shaft sleeve, gap between upper support shaft sleeve and planet carrier, gap between lower support shaft sleeve and gland, etc.
Claims (4)
1. RV reducer service life prediction method based on digital twinning comprises the following steps:
(1) Establishing a data acquisition system: the method comprises the steps that data acquisition is carried out on working conditions and environmental parameters of the RV reducer in actual working by using a sensor, wherein the working conditions and the environmental parameters comprise a torque sensor, a vibration sensor and a temperature sensor, and the acquired contents comprise data such as geometric shapes, material properties, running environmental parameters, torque, return difference and disturbance of each part of the RV reducer;
(2) Establishing a digital twin model of the RV reducer: establishing a digital twin model of the RV reducer by establishing each submodel of the digital twin model, so as to ensure the comprehensive analysis of real-time data; the submodules of the digital twin model comprise a geometric model, a dynamics model, a thermal model and a stress distribution model; acquiring real-time data through a data acquisition system, and uploading the data to a twin model; processing the uploaded data, and performing simulation in each sub-model; the method specifically comprises the following steps:
the digital twin model comprises a plurality of sub-models, namely a geometric model, a dynamics model, a thermal model, a stress analysis model and a fatigue model which are the same as the sizes of the parts of the physical model;
(21) Establishing a geometric model and a kinetic model of the RV reducer: modeling software is used for 1:1 modeling of the RV reducer, so that mapping can be formed between the RV reducer and the RV reducer entity, and simulation calculation is better carried out;
wherein, the step (22) of establishing the geometric model and the dynamic model comprises the following procedures:
(211) Collecting and uploading real-time data;
(212) Drawing a 3D model of the RV reducer according to real-time data 1:1 by using SolidWorks software to form a geometric model;
the parameter equation of the cycloidal gear tooth profile is as follows:
K 1 -short-amplitude coefficients;
e-eccentricity of cycloidal gear;
i H -cycloidal pin gear ratio;
-the angle through which the crank shaft rotates relative to the central sagittal axis of a needle tooth;
r p -the centre circle radius of the needle teeth;
r rp -needle tooth radius;
the data are all the data actually measured by the sensor;
(213) Introducing key parts of the RV reducer in the geometric model into an ANYSYS, selecting units, defining materials, dividing a network, and defining external nodes and a rigid region;
(214) Introducing the geometric model into ADAMS software, replacing part of flexible bodies of key parts, defining cycloid gears, pin teeth, crankshafts and a planet carrier as flexible bodies, setting the rest as rigid bodies, and adding constraint to form a dynamic model;
after SolidWorks modeling, the SolidWorks is stored into an intermediate format file: x_t is a pamaroid format, and then is imported into ADAMS;
the material properties of each component are determined by the actual RV reducer material, taking RV-40E model as an example, and the material properties of each component are shown in the following table:
constraints of each component are added, as shown in the following table:
(215) Carrying out dynamics simulation according to the real-time measured input torque, return difference and other data of the RV reducer;
(216) Outputting a load spectrum of an output shaft of the RV reducer; storing the real-time data, the simulation data and other necessary data, and preparing for the next data analysis;
(22) Establishing a RV reducer thermal model: establishing different gear temperature fields of the RV reducer according to different gear load spectrums obtained by ADAMS simulation; analyzing the influence of parameters such as geometric parameters, load, lubrication characteristics and the like of each gear on the temperature field of each gear;
(23) Establishing a stress analysis model of the RV reducer: according to the load spectrums of different parts obtained by ADAMS simulation of the stress sensor and the obtained data, analyzing and solving stress and stress distribution of each point in the object such as the mechanical part, the component and the like, and determining the stress concentration of dangerous points related to failure of the mechanical part and the component and the peak stress and strain of the strain concentration part;
(24) Establishing a fatigue model of the RV reducer: the fatigue model is established by combining the digital twin model and the actual RV reducer operation data, so that the actual fatigue state of the RV reducer is accurately reflected, and the future fatigue state is estimated;
(25) The data among the models supplement each other, meanwhile, the historical data are compared with the real-time data, when the data are different, errors are marked, and the data are displayed on an operation interface for processing;
(3) RV reduction gear life prediction: the sub-model can obtain the predicted service life of the RV reducer through simulation; through recording the prediction conditions of the input parameters and the output life values of the RV reducer, after the RV reducer runs for a plurality of times, a kriging proxy model can be used for replacing a simulation model to predict the life; according to the data prediction of the sub-model operation and the actual operation, updating the data to each sub-model of the digital twin model in time through updating iteration of the data, and obtaining an accurate service life value of the RV reducer;
(4) Dynamic real-time monitoring and feedback: the actual running condition of the RV reducer is monitored through real-time acquisition of data, so that the reliability of twin model prediction can be ensured; the actual running condition of the RV reducer can synchronize the twin model with the physical model in a data mode, and the simulation data and the actual data are integrated through one-time simulation to accurately monitor and predict the running condition of the RV reducer; the optimized RV reducer parameter combination can be obtained by applying the particle swarm algorithm, the service life of the RV reducer can be prolonged, and a data basis is provided for the design of the RV reducer.
2. The method for predicting the service life of the RV reducer based on digital twinning as defined in claim 1, wherein the method comprises the following steps: the step (1) of establishing a data acquisition system comprises the following processes:
(11) A sensor is additionally arranged in the RV reducer and used for recording dynamic data of the RV reducer in real time when the RV reducer is used; the sensor comprises the following types and uses:
(12) The signals generated by the sensors in actual use of the RV reducer are converted into spectrograms by extracting useful information through wavelet transformation technology;
compared with Fourier transformation, the time of each component can be obtained by utilizing wavelet transformation to receive the signal, the condition that the frequency of the signal changes along with the time is known, and the instantaneous frequency and the amplitude of each moment are analyzed in real time; when the signal is suddenly changed, the wavelet transformation can accurately show the change and reflect the change to a time frequency spectrum, so that the failure condition of each part of the RV reducer can be found conveniently;
(13) And uploading and classifying the data.
3. The method for predicting the service life of the RV reducer based on digital twinning as defined in claim 1, wherein the method comprises the following steps: the service life prediction of the RV reducer in the step (3) comprises the following steps:
(31) Inputting simulation results of each model in the digital twin model into MSC.Fatigue simulation software;
(32) The fatigue life simulation operation of the RV reducer outputs damage data of all parts, and predicts the fatigue life of all parts of the RV reducer;
(33) Comparing the failure standards of the RV reducer parts, and once one part fails, the RV reducer fails to reach the service life maximum value;
(34) Inputting the result into a damage evolution model;
(35) Gradually constructing a life prediction function of the RV reducer by adopting a Kriging proxy model, and establishing a high-precision life prediction proxy model of the RV reducer by continuously collecting input data and output data;
compared with a simulation model, the agent model has a plurality of advantages, firstly, the agent model can replace a complex and time-consuming numerical analysis model, secondly, the agent model can ensure that sample points are accurately converged to a real solution on the basis of historical data, and the agent model can ensure high approximation accuracy in an important area, particularly a real solution area and efficiently obtain a target solution in the face of a complex multidimensional problem;
(36) When the reliability of the RV reducer life prediction agent model meets the requirement, the agent model can be directly used for predicting the life of the RV reducer, so that the simulation prediction time is saved.
4. The method for predicting the service life of the RV reducer based on digital twinning as defined in claim 1, wherein the method comprises the following steps: the step (4) of dynamic real-time monitoring and feedback comprises the following processes:
(41) Recording real-time data;
(42) Recording simulation data;
(43) Recording of prediction data;
(44) Comparing the simulation calculation data with the real-time data;
(45) Comparing the simulation prediction data with real-time data;
(46) According to the different service lives of RV speed reducers with different parameters under different working conditions and environments, various parameters of the RV speed reducers are optimized by adopting a particle swarm algorithm to improve the service life, the best RV speed reducer parameter suggestion can be given under a specific environment, and a data reference is provided for the RV speed reducer parameter design;
the step (46) of parameter optimization design comprises the following steps:
(461) Determining an influence parameter and a value range thereof;
(462) Substituting the service life result into a particle swarm algorithm, and applying a kriging model to calculate the service life result;
(1) initializing a population
The dimension (array element number) of the particle swarm is 11, and the particle swarm is respectively a working load, a pin tooth radius, a pin tooth center circle radius, a crank axle eccentric distance, a cycloid wheel moving distance shape correction amount, a cycloid wheel equidistant shape correction amount, a cycloid wheel inner hole and rotating arm shaft sleeve gap, a crank axle and supporting shaft sleeve gap, an upper supporting shaft sleeve and planet carrier gap, a lower supporting shaft sleeve and a gland gap;
particle population size N: taking a general value of 200;
for each particle, there are two attributes:
position attribute:
X i (t)＝{x i,1 (t),x i,2 (t),…,x i,j (t)}； (2)
speed attribute:
V i (t)＝{v i,1 (t),v i,2 (t),…,v i,j (t)}； (3)
where t represents the t iteration (t meeting), i represents that the number of this particle is i, j represents the dimension of the search space, j=2 (searching in the plane) for the particle;
(2) calculating individual and global optimum
The optimal position searched for by the ith particle so far is called the individual extremum:
P best ＝(p i1 ,p i2 ,…,p ij ),i＝1,2,…N (4)
the optimal position searched so far for the whole population of particles is the global extremum:
g best ＝(g 1 ,g 2 ,…,g j ) (5)
(3) after finding the two optimal values, the particle swarm updates its own speed and position
The speed update formula:
v ij (t+1)＝λ·v ij (t)+c 1 r 1 [p ij (t)-x ij (t)]+c 2 r 2 [p gj (t)-x ij (t)] (6)
wherein c 1 Is a self-learning factor, c 2 Is global learning factor, which is constant factor, taking c 1 ＝c 2 ＝2.5；r 1 ,r 2 Is [0,1]The random number in the range is uniform, so that the randomness of the particle flight is increased;is the particle velocity, and it was found that v was set max And the effect of adjusting the inertial weight is equivalent, so v max The method is generally used for initializing and setting, and takes the value of 10; lambda is a compression factor for controlling the final convergence and increasing the convergence rate, wherein +.>
The location update formula:
x ij (t+1)＝x ij +v ij (t+1) (8)
(4) determining whether convergence criteria of an algorithm are met; if yes, the operation is ended; if not, go to (2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010947950.0A CN112163325B (en) | 2020-09-10 | 2020-09-10 | RV reducer service life prediction method based on digital twinning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010947950.0A CN112163325B (en) | 2020-09-10 | 2020-09-10 | RV reducer service life prediction method based on digital twinning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112163325A CN112163325A (en) | 2021-01-01 |
| CN112163325B true CN112163325B (en) | 2024-03-29 |
Family
ID=73858840
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010947950.0A Active CN112163325B (en) | 2020-09-10 | 2020-09-10 | RV reducer service life prediction method based on digital twinning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112163325B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112417742B (en) * | 2021-01-22 | 2021-04-23 | 浙江中自庆安新能源技术有限公司 | Gearbox life dynamic evaluation method and system based on digital twin model |
| CN112859789B (en) * | 2021-01-29 | 2024-02-06 | 重庆邮电大学 | CFD-based method and system for constructing digital twin body of data center |
| CN112943905B (en) * | 2021-03-12 | 2024-02-23 | 佛山职业技术学院 | Noise reduction method and noise reduction system for electric automobile gearbox |
| CN112883519B (en) * | 2021-03-19 | 2023-03-14 | 河南科技大学 | Service life prediction method and device for cycloid wheel |
| CN113111454B (en) * | 2021-04-01 | 2024-08-23 | 浙江工业大学 | RV reducer dynamic transmission error optimization method based on Kriging model |
| CN113190886A (en) * | 2021-04-28 | 2021-07-30 | 青岛赛普克有限元科技发展有限公司 | Equipment health monitoring method based on rapid simulation digital twinning technology |
| CN113535385B (en) * | 2021-06-21 | 2025-01-24 | 广州杰赛科技股份有限公司 | Virtual network function deployment method, device, equipment and storage medium |
| CN116205101A (en) * | 2023-02-10 | 2023-06-02 | 上海益实智能科技有限公司 | A method and device for parametric design of cigarette factory equipment based on digital twin |
| CN116579217A (en) * | 2023-05-30 | 2023-08-11 | 兰州理工大学 | A Fatigue Life Prediction Method for Control Valve Flow-Induced Vibration Based on Digital Twin |
| CN116611189B (en) * | 2023-07-18 | 2023-10-13 | 苏州迈卡格自动化设备有限公司 | Digital twin stacker metal structure fatigue life analysis method and device |
| CN116611190B (en) * | 2023-07-20 | 2023-10-03 | 宁波东力传动设备有限公司 | Design method of lightweight multi-stage speed reducer |
| CN117168802B (en) * | 2023-08-14 | 2024-06-18 | 苏州长木传动科技有限公司 | Method for detecting performance life of harmonic speed reducer |
| CN117150938B (en) * | 2023-10-31 | 2024-01-26 | 南京鸿雁讯通信息科技有限公司 | Residual life prediction method of induced draft fan based on digital twin |
| CN118395802B (en) * | 2024-05-22 | 2025-02-18 | 中国长江三峡集团有限公司 | Motor temperature stress field proxy model construction method, device, equipment and medium |
| CN118657030B (en) * | 2024-08-19 | 2024-11-15 | 启东市海生冲压件有限公司 | Mold life prediction method based on simulation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045608A (en) * | 2019-04-02 | 2019-07-23 | 太原理工大学 | Based on the twin mechanical equipment component structural dynamic state of parameters optimization method of number |
| CN110532626A (en) * | 2019-07-31 | 2019-12-03 | 西安交通大学 | Based on the twin aeroplane engine main bearing method for predicting residual useful life of number |
| CN110900307A (en) * | 2019-11-22 | 2020-03-24 | 北京航空航天大学 | Numerical control machine tool cutter monitoring system driven by digital twin |
-
2020
- 2020-09-10 CN CN202010947950.0A patent/CN112163325B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045608A (en) * | 2019-04-02 | 2019-07-23 | 太原理工大学 | Based on the twin mechanical equipment component structural dynamic state of parameters optimization method of number |
| CN110532626A (en) * | 2019-07-31 | 2019-12-03 | 西安交通大学 | Based on the twin aeroplane engine main bearing method for predicting residual useful life of number |
| CN110900307A (en) * | 2019-11-22 | 2020-03-24 | 北京航空航天大学 | Numerical control machine tool cutter monitoring system driven by digital twin |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112163325A (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112163325B (en) | RV reducer service life prediction method based on digital twinning | |
| Liang et al. | The process correlation interaction construction of Digital Twin for dynamic characteristics of machine tool structures with multi-dimensional variables | |
| Liu et al. | Data-driven thermal error compensation of linear x-axis of worm gear machines with error mechanism modeling | |
| Stavropoulos | Digitization of manufacturing processes: From sensing to twining | |
| CN113111454B (en) | RV reducer dynamic transmission error optimization method based on Kriging model | |
| Zhang et al. | An adaptive grinding method for precision-cast blades with geometric deviation | |
| An et al. | Tool life prediction based on Gauss importance resampling particle filter | |
| CN113504768A (en) | High-precision product digital twin computability method for assembly quality prediction | |
| Yucesan et al. | Physics-informed digital twin for wind turbine main bearing fatigue: Quantifying uncertainty in grease degradation | |
| Tiainen et al. | DIGITAL TWIN AND VIRTUAL SENSOR FOR A ROTOR SYSTEM. | |
| Li et al. | Clamping force prediction based on deep spatio-temporal network for machining process of deformable parts | |
| CN115994323A (en) | Equipment fault diagnosis model establishment method and equipment fault diagnosis method | |
| CN112990601B (en) | Worm gear machining accuracy self-healing system and method based on data mining | |
| Liu et al. | Data-driven thermal error modeling based on a novel method of temperature measuring point selection | |
| Wang et al. | Digital twin-driven fault diagnosis service of rotating machinery | |
| Yan | Modeling and kinematics simulation of plane ten-bar mechanism of warp knitting machine based on simcape/multibody | |
| Ribeiro et al. | SimuStruct: Simulated Structural Plate with Holes Dataset with Machine Learning Applications | |
| CN118309787B (en) | Transmission efficiency strengthening method and device for speed reducer | |
| Kalinkina et al. | Development of computer-aided design system of elastic sensitive elements of automatic metering devices | |
| Proteau et al. | CNC machining quality prediction using variational autoencoder: A novel industrial 2 TB dataset | |
| CN117454530B (en) | Digital twinning-based automobile body part modeling and detecting method and system | |
| Duan et al. | Intelligent fault diagnosis of 3D printers based on reservoir computing | |
| CN112365022A (en) | Engine bearing fault prediction method based on multiple stages | |
| Sun et al. | Analysis and inference of stream of dimensional errors in multistage machining process based on an improved semiparametric model | |
| Silvestri et al. | Experimental results of a self-learning compensation system for high precision manufacturing |
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 |