A bearing micro-scale lubrication condition monitoring method integrating model and data

Abstract

The present invention discloses a bearing micro-scale lubrication state monitoring method that integrates models and data; first, determine the bearing geometry, surface roughness, friction coefficient and material parameters; second, combine CFD fluid-solid coupling simulation, ADAMS multi-body simulation and bearing dynamics and elastohydrodynamic lubrication coupling algorithm, comprehensively consider the influence of non-steady-state conditions, surface roughness and thermal effects to establish a bearing multi-domain simulation model; then, based on the multi-domain simulation model and the bearing test bench, respectively collect the simulation and test data such as the temperature of each component of the bearing, axial load, radial load, vibration acceleration, and the speed of the inner ring and cage of the bearing under the same working conditions; finally, by adjusting the lubricating oil flow in the multi-domain simulation model, the maximum mean difference between the bearing simulation and the test data meets the consistency requirement and the bearing twin simulation model is obtained, and the bearing micro-scale lubrication state is extracted based on the twin simulation model. The present invention can realize accurate monitoring of the micro-scale lubrication state of the bearing under variable working conditions.

Description

Technical Field

[0001] The present invention relates to the field of bearing lubrication state detection, and in particular to a bearing micro-scale lubrication state monitoring method that integrates models and data. Background Art

[0002] Monitoring the micro-lubrication state of bearings is crucial to analyzing the dynamic performance of bearings. The motion characteristics of high-speed bearings are easily affected by the lubrication state of bearings. Accurate bearing micro-lubrication state requires comprehensive consideration of non-steady-state conditions such as acceleration and deceleration, the influence of surface roughness and thermal effects to establish an accurate twin simulation model of the bearing. This requires the use of CFD fluid-solid coupling simulation, ADAMS multi-body dynamics simulation, and bearing dynamics and elastohydrodynamic lubrication coupling algorithms for analysis and modeling. Because related research involves multidisciplinary simulation and is difficult, there is currently no comprehensive research on related issues.

[0003] Regarding the research on the coupling algorithm of bearing dynamics and elastohydrodynamic lubrication, since the numerical analysis of the relevant theories is very complicated, most of the research on the coupling algorithm is to simplify one of the theories. The present invention combines ADAMS multi-body dynamics simulation and CFD fluid-solid coupling simulation to effectively reduce the complexity of the numerical solution of bearing dynamics, and can realize the coupling analysis of bearing dynamics and elastohydrodynamic lubrication theory without any simplification.

[0004] Since the contact surface of the bearing is not completely smooth and has a certain degree of roughness, and for high-speed bearings, the oil film thickness and surface roughness are in the same order of magnitude in most cases, the influence of surface roughness on the oil film characteristics cannot be ignored. At the same time, since the oil film characteristics are extremely unstable during variable working conditions such as acceleration and deceleration, it is easy to have a significant impact on the motion characteristics of the bearing. Therefore, the present invention comprehensively considers the influence of non-steady-state working conditions such as surface roughness, acceleration and deceleration, and thermal effects, and can realize accurate monitoring of the micro-scale lubrication state of the bearing under variable working conditions. Summary of the invention

[0005] Technical problem: In view of this, the purpose of the present invention is to provide a bearing micro-scale lubrication state monitoring method that integrates models and data, which can realize accurate monitoring of the bearing micro-scale lubrication state under variable working conditions and has high practical value and application prospects.

[0006] Technical solution: A bearing micro-scale lubrication state monitoring method integrating models and data of the present invention comprises:

[0007] Step S1, determining the geometric dimensions, surface roughness, friction coefficient and material parameters of each bearing component;

[0008] Step S2, based on the parameters obtained in step S1, combined with CFD fluid-solid coupling simulation, ADAMS multi-body dynamics simulation, and bearing dynamics and elastohydrodynamic lubrication coupling algorithm, comprehensively considering the influence of unsteady-state conditions of acceleration and deceleration, surface roughness and thermal effects, establish a bearing multi-domain simulation model;

[0009] Step S3, based on the bearing multi-domain simulation model and the bearing test bench obtained in step S2, respectively collect simulation and test data of the bearing inner ring temperature, outer ring temperature, rolling element temperature, cage temperature, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed under the same radial load, axial load and bearing inner ring speed conditions;

[0010] Step S4: Use the maximum mean difference to evaluate the simulation and test data obtained in step S3. If the consistency requirements are not met, it is necessary to adjust the lubricating oil flow of the bearing multi-domain simulation model in step S2. If the consistency requirements are met, the micro-scale lubrication state of the bearing is extracted based on the twin simulation model.

[0011] in,

[0012] The step S1 comprises:

[0013] Step S101, determining the geometric dimensions of each bearing component, specifically including:

[0014] Determine the inner diameter, outer diameter and width of the inner and outer rings of the bearing respectively; determine the diameter and length of the rolling element; determine the width, inner diameter, outer diameter and pocket size of the bearing cage; use Solidwork software to establish the bearing geometry model based on the obtained geometric dimensions, and use Hypermesh software to mesh the bearing to obtain a finite element model;

[0015] Step S102, determining the surface roughness of each bearing component:

[0016] Measure the surface roughness of the outer surface of the inner ring, the inner surface of the outer ring and the rolling element surface of the bearing in turn. Select N points evenly on each surface in turn, measure each point 10 times, and take the average value as the surface roughness of the inner ring and outer surface of the bearing Outer ring inner surface roughness and rolling element surface roughness

[0017] Step S103, determining the dynamic and static friction coefficient of the bearing:

[0018] The dynamic and static friction coefficient μ between the rolling element and the inner ring was measured under different inner ring speeds, axial loads, radial loads and lubrication conditions. i And the dynamic and static friction coefficient μ between the rolling element and the outer ring o ;

[0019] Step S104: Determine bearing material parameters.

[0020] The step S104 specifically includes:

[0021] Step S1041, determining the elastic modulus and Poisson's ratio of the bearing according to modal test and model correction technology;

[0022] Step S1042: Calculate the volumes of the inner ring, outer ring, cage and rolling element according to the bearing geometric model established in step S101, measure the mass of each bearing component, and calculate the density of each bearing component using the following formula:

[0023]

[0024] Where: n = a, b, c, d represent the bearing rolling element, inner ring, outer ring and cage respectively; m n is the mass of different bearing components; V n is the volume of different parts of the bearing; ρ n is the density of different parts of the bearing;

[0025] Step S1043: Determine the specific heats of the bearing rolling element and the inner ring, respectively. a and c b , determine the heat transfer coefficients of the bearing rolling elements and inner ring respectively as k a and k b .

[0026] The step S2 comprises:

[0027] Step S201, based on the parameters obtained in step S1, combined with the bearing dynamics and elastohydrodynamic lubrication analysis coupling algorithm, comprehensively considering the influence of non-steady-state conditions such as acceleration and deceleration, surface roughness and thermal effects, establish a bearing numerical calculation model; wherein the bearing dynamics analysis provides mechanical parameters and kinematic parameters as inputs of the elastohydrodynamic lubrication analysis, and the bearing elastohydrodynamic lubrication analysis feeds back the temperature and lubricating oil film distribution to the bearing dynamics analysis;

[0028] Step S202: Establish a CFD fluid-solid coupling simulation model based on the parameters obtained in step S1 and the heat generated by the bearing calculated by the transient energy equation in step S201:

[0029] Based on the parameters obtained in step S1, the Fluent software is used to establish a bearing fluid domain model and divide the grid, and a fluid domain model including the flow space inside the bearing is established. The RNGk-ε turbulence model is used to describe the flow of oil and gas two phases inside the cylindrical roller bearing; the multiple reference system method is used to describe the fluid movement inside the bearing, and the transient flow is converted into a steady state for approximate solution;

[0030] Boundary condition setting: set the inside of the bearing to a two-phase flow of lubricating oil and air, where air is the main phase and lubricating oil is the secondary phase; set the inner ring wall of the bearing to a rotating wall boundary, the outer ring wall of the bearing to a stationary wall boundary, and the retainer wall to a rotating wall boundary, and the lubricating oil inlet and outlet are distributed on both sides of the bearing; at the same time, set temperature boundary conditions for the inner and outer rings of the bearing and the rolling element wall in combination with the heat generated by the bearing calculated by the transient energy equation in step S201;

[0031] Step S203: Establish an ADAMS multi-body dynamics simulation model based on the parameters obtained in step S1:

[0032] The bearing finite element model obtained in step S101 is imported into the ADAMS multi-body dynamics simulation software, and the parameters such as the dynamic and static friction coefficient of the bearing, the elastic modulus and Poisson's ratio of each bearing component, and the density obtained in step S103 are imported into the ADAMS multi-body dynamics simulation model;

[0033] Step S204, combining the bearing numerical calculation model established in step S201, the CFD fluid-solid coupling simulation model established in step S202 and the ADAMS multi-body dynamics simulation model established in step S203 to perform a joint simulation to obtain the bearing contact micro-area temperature, mechanical parameters and motion parameters.

[0034] The step S201 specifically includes:

[0035] S2011, Elastohydrodynamic lubrication analysis of bearings

[0036] The elastohydrodynamic lubrication theory considering time-varying effects, surface roughness and thermal effects includes three parts: the transient Reynolds equation, the transient film thickness equation considering surface roughness and the transient energy equation. The specific calculation formula is as follows:

[0037] Transient Reynolds equation:

[0038]

[0039] Where: x j and j is the coordinate value of the jth rolling element; j = 1, 2, ..., Z is the number of rolling elements; η is the viscosity of the lubricating oil; ρ is the density of the lubricating oil; p j is the lubricating oil pressure at the jth rolling element; h j is the oil film thickness at the jth rolling element; is the partial differential operator; t is the time; U is the average velocity of the contact point of the friction pair along the x-axis; V is the average velocity of the contact point of the friction pair along the y-axis;

[0040] Transient film thickness equation considering surface roughness:

[0041]

[0042] Where: h j (x j ,y j ,t) is the jth rolling element with time t and position (x j ,y j ) is the oil film thickness related to time t; h0(t) is the film thickness at the center of the bearing related to time t; E' j is the equivalent elastic modulus of the jth rolling element; and are the jth rolling element along x j and j Surface roughness in the direction; is the surface roughness of the jth rolling element;

[0043] Transient energy equation:

[0044]

[0045] Where: c p is the specific heat capacity of the lubricating oil; T j is the temperature of the jth rolling element; k is the heat transfer coefficient of the lubricating oil; u is the velocity component of the fluid in the x-axis direction; v is the velocity component of the fluid in the y-axis direction; where the heat transfer equation between the rolling element and the inner ring of the bearing is as follows:

[0046]

[0047]

[0048] Where: U a and U b are the linear speeds of rolling element and inner ring of bearing respectively; z a and z b are the coordinate values ​​of the rolling element and the inner ring of the bearing along the z-axis respectively;

[0049] S2012, Bearing Dynamics Analysis

[0050] The dynamic analysis of the bearing requires the establishment of three parts: the rolling element dynamic equation, the cage dynamic equation and the bearing ring dynamic equation. The specific calculation formula is as follows:

[0051] Rolling element dynamics equation:

[0052]

[0053]

[0054]

[0055] Where: NP is the number of slices of the rolling element along its axis; W is the width of the slice; A is a constant; F c The centrifugal force on the rolling element; and F i j The jth rolling element bears the drag force of the inner raceway and the outer raceway respectively; and is the normal force on the jth rolling element received by the cage; and is the tangential drag force of the jth cage; and are the resistances on the surface and end of the jth rolling element respectively; is the drag force of the jth rolling element guide rib on the rolling element end surface; is the blocking torque of the jth rolling element rib on the rolling element; is the angular acceleration of the jth rolling element rotating around its own axis; is the angular velocity of the jth rolling element; m is the mass of the rolling element; r is the radius of the rolling element; R is the radius of the cage guide surface; g 1ij and g 2ij are the contact deformations of the i-th slice of the j-th rolling element with the inner raceway and the outer raceway respectively; l is the effective contact length of the rolling element;

[0056] Cage dynamic equation:

[0057]

[0058]

[0059]

[0060] Where: m cage is the mass of the cage; φ j is the azimuth angle of the jth rolling element; T CDO T is the traction torque on the outer surface of the cage; CDS is the side blocking torque of the cage; T land is the torque generated by the fluid dynamic pressure; I cage is the moment of inertia of the cage; α is the central angle of the cage pocket; W x and W y The loads along the x-axis and y-axis respectively; and The displacement acceleration of the cage along the x-axis and y-axis respectively;

[0061] Dynamic equation of the inner ring of the bearing:

[0062]

[0063] Where: and are the normal contact forces of the outer raceway and the inner raceway on the jth rolling element respectively; and are the friction torques between the jth rolling element and the outer ring raceway and the inner ring raceway respectively; F y and F z are the loads along the y-axis and z-axis respectively; M y and M z The moments along the y and z axes respectively.

[0064] The step S204 specifically includes:

[0065] Step S2041: Perform joint simulation based on the bearing numerical calculation model established in step S201 and the CFD fluid-solid coupling simulation model established in step S202:

[0066] According to the heat generated by the bearing calculated in step S201, it is used as the boundary condition of the CFD fluid-solid coupling simulation model established in step S202 to obtain the bearing temperature field distribution, lubricating oil distribution and lubricating oil film thickness; at the same time, the obtained parameters such as temperature and oil film thickness are fed back to the bearing numerical calculation model for coupling analysis;

[0067] Step S2042: Based on the joint simulation model established in step S2041, the joint simulation model is further combined with the ADAMS multi-body dynamics simulation model established in step S203 to obtain a bearing multi-domain simulation model:

[0068] The bearing numerical calculation model established in step S201 is written in Fortran language, and is further jointly simulated with the ADAMS multi-body dynamics simulation model established in step S203 in combination with the joint simulation result obtained in step S2041; wherein, the bearing contact micro-area temperature, mechanical parameters and motion parameters obtained by the joint simulation model of step S2041 are input into the ADAMS multi-body dynamics simulation model, and the interaction force and deformation extracted from the ADAMS multi-body dynamics simulation model are input into the joint simulation model of step S2041 for solution, thereby obtaining the bearing multi-domain simulation model.

[0069] The step S3 comprises:

[0070] Step S301, under certain radial load, axial load and bearing inner ring speed conditions, test data such as bearing inner ring temperature, outer ring temperature, rolling element temperature, cage temperature, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed are collected respectively:

[0071] Step S302, based on the obtained bearing multi-domain simulation model, under the same radial / axial load and bearing inner ring speed as step S301, respectively collect simulation data such as bearing inner ring temperature, outer ring temperature, rolling element temperature, cage temperature, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed.

[0072] The step S4 comprises:

[0073] Step S401: Use the maximum mean difference to evaluate the bearing simulation data and test data obtained in step S3:

[0074] Assume that the bearing simulation data and test data obtained in step S3 are as follows:

[0075] S={s1,s2,...,s 10}={T rs ,T os ,T is ,T cas ,W bas ,W brs ,a xs ,a ys ,a zs ,ω ins ,ω cages} (14)

[0076] E={e1,e2,...,e 10}={T re ,T oe ,T ie ,T cae ,W bae ,W bre ,a xe ,a ye ,a ze ,ω ine ,ω cagee} (15)

[0077] Where: S and E are the bearing simulation data set and test data set respectively; T rs , T os , T is and T cas are the temperatures of the bearing rolling element, outer ring, inner ring and cage measured by simulation; W bas and W brs are the axial and radial loads measured by simulation; a xs 、a ys and a zs are the vibration accelerations of the bearing in the x, y and z directions measured by simulation; ω ins and ω cagesare the speeds of the inner ring and cage of the bearing measured by simulation; T re , T oe , T ie and T cae are the temperatures of the bearing rolling element, outer ring, inner ring and cage measured in the test; W bae and W bre are the axial and radial loads measured in the test; a xe 、a ye and a ze are the vibration accelerations of the bearing in the x, y and z directions measured in the test; ω ine and ω cagee are the experimentally measured speeds of the inner ring and cage of the bearing, respectively; the empirical estimate of the distance between the two distributions defined by the maximum mean difference is:

[0078]

[0079] Where: ||·|| H represents the modulus of the vector in the universal reproducing kernel Hilbert space; φ(·) is the nonlinear mapping from the original feature space to the reproducing kernel Hilbert space; Dist(·) represents the maximum average difference distance between the two distributions;

[0080] Step S402: If the maximum mean difference obtained in step S401 does not meet the consistency requirement, the lubricating oil flow and viscosity parameters need to be adjusted. If the consistency requirement is met, a twin simulation model is obtained:

[0081] If the maximum mean difference obtained in step S401 is greater than 1, that is, the consistency requirement is not met, it is necessary to adjust the lubricating oil flow in the bearing multi-domain simulation model established in step S2042 until the maximum mean difference obtained in step S401 is less than or equal to 1. At this time, it can be considered that the established bearing multi-domain simulation model is a twin simulation model that is completely consistent with the motion state and lubrication characteristics of the test bearing;

[0082] Step S403: extracting the micro-scale lubrication state of the bearing based on the twin simulation model obtained in step S402:

[0083] Under different operating and lubrication conditions of the bearing, the temperature field distribution, oil film thickness and lubricating oil distribution at a certain microscopic scale inside the bearing are observed based on the bearing twin simulation model obtained in step S402 to achieve accurate monitoring of the microscale lubrication state of the bearing.

[0084] Beneficial effects: The advantages of the present invention are:

[0085] 1. The present invention comprehensively considers various factors such as non-steady-state working conditions such as acceleration and deceleration, surface roughness and thermal effects, and establishes a practical multi-domain simulation model of bearings, laying a foundation for the study of bearing micro-motion state and lubrication characteristics;

[0086] 2. Based on the established bearing twin simulation model, the present invention realizes accurate monitoring of the bearing lubrication status under variable working conditions, and has high practical value and application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 A schematic flow chart of a bearing micro-scale lubrication state monitoring method for integrating a model and data provided in Example 1;

[0088] Figure 2 The geometric structure diagram of the N208 cylindrical roller bearing used in Example 1;

[0089] Figure 3 The bearing dynamics and elastohydrodynamic lubrication analysis coupling algorithm steps provided in Example 1;

[0090] Figure 4 The CFD fluid-solid coupling simulation model provided in Example 1;

[0091] Figure 5 The joint simulation idea provided in Example 1;

[0092] Figure 6 is the bearing vibration acceleration signal provided in Example 1, wherein a is the vibration acceleration signal in the x direction, b is the vibration acceleration signal in the y direction, and c is the vibration acceleration signal in the z direction;

[0093] Figure 7 This is the bearing micro-scale lubrication state monitoring result provided in Example 1, wherein a is the oil film pressure distribution, b is the oil film thickness distribution, and c is the oil film temperature distribution. DETAILED DESCRIPTION

[0094] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0095] Example 1

[0096] See also Figure 1-Figure 7 This embodiment provides a bearing micro-scale lubrication state monitoring method that integrates models and data. The process of the method is as follows: Figure 1 As shown, this embodiment uses N208 cylindrical roller bearing as an example to verify the algorithm performance. Figure 2 The geometric structure diagram of N208 cylindrical roller bearing is shown in Figure 1. The strategy specifically includes the following steps:

[0097] Step S1, determining the geometric dimensions, surface roughness, friction coefficient and material parameters of each bearing component;

[0098] Specifically, in this embodiment, step S1 includes:

[0099] Step S101, determining the geometric dimensions of each bearing component, specifically including:

[0100] Determine the inner diameter, outer diameter and width of the inner and outer rings of the bearing respectively; determine the diameter and length of the rolling element; determine the width, inner diameter, outer diameter and pocket size of the bearing cage; use Solidwork software to establish the bearing geometry model based on the obtained geometric dimensions, and use Hypermesh software to mesh the bearing to obtain a finite element model; the dimensions of the N208 bearing are shown in Table 1:

[0101] Table 1 N208 bearing dimensions

[0102]

[0103] Step S102, determining the surface roughness of each bearing component:

[0104] Measure the surface roughness of the outer surface of the inner ring, the inner surface of the outer ring and the rolling element surface of the bearing in turn. Select N points evenly on each surface in turn, measure each point 10 times, and take the average value as the surface roughness of the inner ring and outer surface of the bearing Outer ring inner surface roughness and rolling element surface roughness The measured values ​​of the inner surface roughness of the bearing outer ring are shown in Table 2, the measured values ​​of the surface roughness of the bearing rolling element are shown in Table 3, and the measured values ​​of the outer surface roughness of the bearing inner ring are shown in Table 4.

[0105] Table 2 Bearing outer ring inner surface roughness values

[0106]

[0107] Table 3 Bearing rolling element surface roughness values

[0108]

[0109] Table 4 Roughness values ​​of outer surface of bearing inner ring

[0110]

[0111] Step S103, determining the dynamic and static friction coefficient of the bearing:

[0112] The dynamic and static friction coefficient μ between the rolling element and the inner ring was measured under different inner ring speeds, axial loads, radial loads and lubrication conditions. i And the dynamic and static friction coefficient μ between the rolling element and the outer ring o ; The measured friction coefficient results are shown in Table 5:

[0113] Table 5 Measurement results of dynamic and static friction coefficient of bearings

[0114]

[0115] Step S104, determining bearing material parameters:

[0116] (1) The elastic modulus and Poisson's ratio of each bearing component are determined based on modal tests and model correction technology; the corrected material parameters of each bearing component are shown in Table 6:

[0117] Table 6 Corrected material parameters of bearing components

[0118]

[0119]

[0120] (2) Calculate the volumes of the inner ring, outer ring, cage and rolling element according to the bearing geometric model established in step S101, measure the mass of each bearing component, and calculate the density of each bearing component using the following formula:

[0121]

[0122] Where: n = a, b, c, d represent the bearing rolling element, inner ring, outer ring and cage respectively; m n is the mass of different bearing components; V n is the volume of different parts of the bearing; ρ n is the density of different parts of the bearing; the density of each bearing component is shown in Table 7:

[0123] Table 7 Density of bearing components

[0124]

[0125] (3) Determine the specific heat of the bearing rolling element and inner ring, respectively: c a and c b , determine the heat transfer coefficients of the bearing rolling elements and inner ring respectively as k a and k b ; The correlation coefficients are shown in Table 8:

[0126] Table 8 Specific heat and thermal conductivity of bearing rolling elements and inner rings

[0127]

[0128] Step S2, based on the parameters obtained in step S1, combined with CFD fluid-solid coupling simulation, ADAMS multi-body dynamics simulation, and bearing dynamics and elastohydrodynamic lubrication coupling algorithm, comprehensively considering the influence of non-steady-state conditions such as acceleration and deceleration, surface roughness and thermal effects, establish a bearing multi-domain simulation model;

[0129] Specifically, in this embodiment, step S2 includes:

[0130] Step S201: Based on the parameters obtained in step S1, the bearing dynamics and elastohydrodynamic lubrication analysis coupling algorithm are combined to comprehensively consider the influence of non-steady-state conditions such as acceleration and deceleration, surface roughness and thermal effects to establish a bearing numerical calculation model. The bearing dynamics analysis provides mechanical parameters and kinematic parameters as inputs of the elastohydrodynamic lubrication analysis. The bearing elastohydrodynamic lubrication analysis feeds back the temperature and lubricating oil film distribution to the bearing dynamics analysis. The coupling solution steps are as follows: Figure 3 As shown; the specific contents include:

[0131] (1) Analysis of bearing elastohydrodynamic lubrication

[0132] The elastohydrodynamic lubrication theory considering time-varying effects, surface roughness and thermal effects includes three parts: the transient Reynolds equation, the transient film thickness equation considering surface roughness and the transient energy equation. The specific calculation formula is as follows:

[0133] Transient Reynolds equation:

[0134]

[0135] Where: x j and j is the coordinate value of the jth rolling element; j = 1, 2, ..., 14 is the number of rolling elements; η is the viscosity of the lubricating oil; ρ is the density of the lubricating oil; p j is the lubricating oil pressure at the jth rolling element; h j is the oil film thickness at the jth rolling element; is the partial differential operator; t is the time; U is the average velocity of the contact point of the friction pair along the x-axis; V is the average velocity of the contact point of the friction pair along the y-axis;

[0136] Transient film thickness equation considering surface roughness:

[0137]

[0138] Where: h j (x j ,y j ,t) is the jth rolling element with time t and position (x j ,yj ) is the oil film thickness related to time t; h0(t) is the film thickness at the center of the bearing related to time t; E' j is the equivalent elastic modulus of the jth rolling element; and are the jth rolling element along x j and j Surface roughness in the direction; is the surface roughness of the jth rolling element;

[0139] Transient energy equation:

[0140]

[0141] Where: c p is the specific heat capacity of the lubricating oil; T j is the temperature of the jth rolling element; k is the heat transfer coefficient of the lubricating oil; u is the velocity component of the fluid in the x-axis direction; v is the velocity component of the fluid in the y-axis direction; where the heat transfer equation between the rolling element and the inner ring of the bearing is as follows:

[0142]

[0143]

[0144] Where: U a and U b are the linear speeds of rolling element and inner ring of bearing respectively; z a and z b are the coordinate values ​​of the rolling element and the inner ring of the bearing along the z-axis respectively;

[0145] (2) Bearing dynamics analysis

[0146] The dynamic analysis of the bearing requires the establishment of three parts: the rolling element dynamic equation, the cage dynamic equation and the bearing ring dynamic equation. The specific calculation formula is as follows:

[0147] Rolling element dynamics equation:

[0148]

[0149]

[0150]

[0151] Where: NP is the number of slices of the rolling element along its axis; W is the width of the slice; A is a constant; F c The centrifugal force on the rolling element; and F i j The jth rolling element bears the drag force of the inner raceway and the outer raceway respectively; and The normal force acting on the jth rolling element by the cage; and The tangential drag force of the cage on the jth rolling element; and are the resistances on the surface and end of the jth rolling element respectively; is the drag force of the jth rolling element guide rib on the rolling element end surface; is the blocking torque of the jth rolling element rib on the rolling element; is the angular acceleration of the jth rolling element rotating around its own axis; is the angular velocity of the jth rolling element; m is the mass of the rolling element; r is the radius of the rolling element; R is the radius of the cage guide surface; g 1ij and g 2ij are the contact deformations of the i-th slice of the j-th rolling element with the inner raceway and the outer raceway respectively; l is the effective contact length of the rolling element;

[0152] Cage dynamic equation:

[0153]

[0154]

[0155]

[0156] Where: m cage is the mass of the cage; φ j is the azimuth angle of the jth rolling element; T CDO T is the traction torque on the outer surface of the cage; CDS is the side blocking torque of the cage; T land is the torque generated by the fluid dynamic pressure; I cage is the moment of inertia of the cage; α is the central angle of the cage pocket; W x and W y The loads along the x-axis and y-axis respectively; and The displacement acceleration of the cage along the x-axis and y-axis respectively;

[0157] Dynamic equation of the inner ring of the bearing:

[0158]

[0159] Where: and are the normal contact forces of the outer raceway and the inner raceway on the jth rolling element respectively; and are the friction torques between the jth rolling element and the outer ring raceway and the inner ring raceway respectively; Fy and F z are the loads along the y-axis and z-axis respectively; M y and M z The moments along the y-axis and z-axis respectively;

[0160] Step S202: Establish a CFD fluid-solid coupling simulation model based on the parameters obtained in step S1 and the heat generated by the bearing calculated by the transient energy equation in step S201:

[0161] Based on the parameters obtained in step S1, the Fluent software is used to establish a bearing fluid domain model and divide the grid, and a fluid domain model including the flow space inside the bearing is established. The RNGk-ε turbulence model is used to describe the flow of oil and gas two phases inside the cylindrical roller bearing; the multiple reference system method is used to describe the fluid movement inside the bearing, and the transient flow is converted into a steady state for approximate solution;

[0162] In the boundary condition setting: the inside of the bearing is set to a two-phase flow of lubricating oil and air, in which air is the main phase and lubricating oil is the secondary phase; the inner ring wall of the bearing is set to a rotating wall boundary, the outer ring wall of the bearing is set to a stationary wall boundary, the retainer wall is set to a rotating wall boundary, and the lubricating oil inlet and outlet are distributed on both sides of the bearing; at the same time, combined with the heat generated by the bearing calculated by the transient energy equation in step S201, the temperature boundary conditions are set for the inner and outer rings of the bearing and the wall of the rolling element; the fluid boundary conditions are set according to the lubricating oil parameters; the CFD fluid-solid coupling simulation model is as follows Figure 4 As shown; lubricating oil parameters are shown in Table 9:

[0163] Table 9 Lubricating oil parameters

[0164]

[0165] In the boundary condition setting, the initial temperature of the lubricating oil and the ambient temperature are 300K, and the initial temperature of the bearing is 320K.

[0166] Step S203: Establish an ADAMS multi-body dynamics simulation model based on the parameters obtained in step S1:

[0167] The bearing finite element model obtained in step S101 is imported into the ADAMS multi-body dynamics simulation software, and the parameters such as the dynamic and static friction coefficient of the bearing, the elastic modulus and Poisson's ratio of each bearing component, and the density obtained in step S103 are imported into the ADAMS multi-body dynamics simulation model;

[0168] Step S204: Combine the numerical calculation model of the bearing established in step S201, the CFD fluid-solid coupling simulation model established in step S202, and the ADAMS multi-body dynamics simulation model established in step S203 to perform joint simulation to obtain the contact micro-area temperature, mechanical parameters, and motion parameters of the bearing. The idea of ​​joint simulation is as follows: Figure 5 As shown; the specific steps are as follows:

[0169] (1) Perform joint simulation based on the bearing numerical calculation model established in step S201 and the CFD fluid-solid coupling simulation model established in step S202:

[0170] According to the heat generated by the bearing calculated in step S201, it is used as the boundary condition of the CFD fluid-solid coupling simulation model established in step S202 to obtain the bearing temperature field distribution, lubricating oil distribution and lubricating oil film thickness; at the same time, the obtained parameters such as temperature and oil film thickness are fed back to the bearing numerical calculation model for coupling analysis;

[0171] (2) The joint simulation model established in step (1) is further combined with the ADAMS multi-body dynamics simulation model established in step S203 to obtain a bearing multi-domain simulation model:

[0172] The bearing numerical calculation model established in step S201 is written in Fortran language, and is further co-simulated with the ADAMS multi-body dynamics simulation model established in step S203 in combination with the co-simulation result obtained in step (1); wherein the bearing contact micro-area temperature, mechanical parameters and motion parameters obtained in the co-simulation model of step (1) are input into the ADAMS multi-body dynamics simulation model, and the interaction force and deformation extracted from the ADAMS multi-body dynamics simulation model are input into the co-simulation model of step (1) for solution, thereby obtaining a multi-domain simulation model of the bearing;

[0173] Step S3, based on the bearing multi-domain simulation model and the bearing test bench obtained in step S2, respectively collect simulation and test data such as temperature parameters of bearing components, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed under the same radial / axial load and bearing inner ring speed conditions;

[0174] Specifically, in this embodiment, step S3 includes:

[0175] Step S301, collecting test data such as temperature parameters of bearing components, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed under certain radial load, axial load and bearing inner ring speed conditions:

[0176] Step S302: Based on the multi-domain simulation model of the bearing obtained in step S204, simulation data such as temperature parameters of the bearing components, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed are collected under the same radial / axial load and bearing inner ring speed as in step S301:

[0177] Step S4, using the maximum mean difference to evaluate the bearing simulation data and the test data obtained in step S3, if the consistency requirement is not met, it is necessary to adjust the lubricating oil flow of the bearing multi-domain simulation model in step S2, if the consistency requirement is met, extract the bearing micro-scale lubrication state based on the twin simulation model;

[0178] Specifically, in this embodiment, step S4 includes:

[0179] Step S401: Use the maximum mean difference to evaluate the bearing simulation data and test data obtained in step S3:

[0180] Assume that the bearing simulation data and test data obtained in step S3 are as follows:

[0181] S={s1,s2,…,s 10}={T rs ,T os ,T is ,T cas ,W bas ,W brs ,a xs ,a ys ,a zs ,ω ins ,ω cages} (14)

[0182] E={e1,e2,…,e 10}={T re ,T oe ,T ie ,T cae ,W bae ,W bre ,a xe ,a ye ,a ze ,ω ine ,ω cagee} (15)

[0183] Where: S and E are the bearing simulation data set and test data set respectively; T rs , T os , T is and T cas are the temperatures of the bearing rolling element, outer ring, inner ring and cage measured by simulation; W bas and W brs are the axial and radial loads measured by simulation; a xs 、a ys and a zs are the vibration accelerations of the bearing in the x, y and z directions measured by simulation; ω ins and ω cagesare the speeds of the inner ring and cage of the bearing measured by simulation; T re , T oe , T ie and T cae are the temperatures of the bearing rolling element, outer ring, inner ring and cage measured in the test; W bae and W bre are the axial and radial loads measured in the test; a xe 、a ye and a ze are the vibration accelerations of the bearing in the x, y and z directions measured in the test; ω ine and ω cagee are the experimentally measured speeds of the inner ring and cage of the bearing, respectively; the empirical estimate of the distance between the two distributions defined by the maximum mean difference is:

[0184]

[0185] Where: ||·|| H represents the modulus of the vector in the universal reproducing kernel Hilbert space; φ(·) is the nonlinear mapping from the original feature space to the reproducing kernel Hilbert space; Dist(·) represents the maximum average difference distance between the two distributions; the measured bearing vibration signal is as follows Figure 6 As shown;

[0186] Step S402: If the maximum mean difference obtained in step S401 does not meet the consistency requirement, the lubricating oil flow and viscosity parameters need to be adjusted. If the consistency requirement is met, a twin simulation model is obtained:

[0187] If the maximum mean difference obtained in step S401 is greater than 1, that is, the consistency requirement is not met, then the lubricating oil flow in the bearing multi-domain simulation model established in step S2042 needs to be adjusted until the maximum mean difference obtained in step S401 is less than or equal to 1. At this time, it can be considered that the established bearing multi-domain simulation model is a twin simulation model that is completely consistent with the motion state and lubrication characteristics of the test bearing; the maximum mean difference calculated in this embodiment is 0.603, which meets the consistency requirement;

[0188] Step S403: extracting the micro-scale lubrication state of the bearing based on the twin simulation model obtained in step S402:

[0189] Under different operation and lubrication conditions of the bearing, the temperature field distribution, oil film thickness and lubricating oil distribution at a certain microscopic scale inside the bearing are observed based on the bearing twin simulation model obtained in step S402, so as to accurately monitor the microscale lubrication state of the bearing. The monitoring results of the microscale lubrication state of the bearing are as follows: Figure 7 shown.

[0190] The matters not described in detail in the present invention are all known technologies to those skilled in the art.

[0191] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A bearing micro-scale lubrication state monitoring method integrating model and data, characterized in that: The method includes: Step S1, determining the geometric dimensions, surface roughness, friction coefficient and material parameters of each bearing component; Step S2, based on the parameters obtained in step S1, combined with CFD fluid-solid coupling simulation, ADAMS multi-body dynamics simulation, and bearing dynamics and elastohydrodynamic lubrication coupling algorithm, comprehensively considering the influence of unsteady-state conditions of acceleration and deceleration, surface roughness and thermal effects, establish a bearing multi-domain simulation model; Step S3, based on the bearing multi-domain simulation model and the bearing test bench obtained in step S2, respectively collect simulation and test data of the bearing inner ring temperature, outer ring temperature, rolling element temperature, cage temperature, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed under the same radial load, axial load and bearing inner ring speed conditions; Step S4, using the maximum mean difference to evaluate the simulation and test data obtained in step S3, if the consistency requirement is not met, it is necessary to adjust the lubricating oil flow of the bearing multi-domain simulation model in step S2, if the consistency requirement is met, extract the bearing micro-scale lubrication state based on the twin simulation model; The step S1 comprises: Step S101, determining the geometric dimensions of each bearing component, specifically including: Determine the inner diameter, outer diameter and width of the inner and outer rings of the bearing respectively; determine the diameter and length of the rolling element; determine the width, inner diameter, outer diameter and pocket size of the bearing cage; use Solidworks software to establish the bearing geometry model based on the obtained geometric dimensions, and use Hypermesh software to mesh the bearing to obtain a finite element model; Step S102, determining the surface roughness of each bearing component: Measure the surface roughness of the outer surface of the inner ring, the inner surface of the outer ring and the rolling element surface of the bearing in turn. Select N points evenly on each surface in turn, measure each point 10 times, and take the average value as the surface roughness of the inner ring and outer surface of the bearing. Outer ring inner surface roughness and rolling element surface roughness Step S103, determining the dynamic and static friction coefficient of the bearing: The dynamic and static friction coefficient μ between the rolling element and the inner ring was measured under different inner ring speeds, axial loads, radial loads and lubrication conditions. i And the dynamic and static friction coefficient μ between the rolling element and the outer ring o ; Step S104, determining bearing material parameters; The step S2 comprises: Step S201, based on the parameters obtained in step S1, combined with the bearing dynamics and elastohydrodynamic lubrication analysis coupling algorithm, comprehensively considering the non-steady-state conditions of acceleration and deceleration, the influence of surface roughness and thermal effect, a bearing numerical calculation model is established; wherein the bearing dynamics analysis provides mechanical parameters and kinematic parameters as inputs of the elastohydrodynamic lubrication analysis, and the bearing elastohydrodynamic lubrication analysis feeds back the temperature and lubricating oil film distribution to the bearing dynamics analysis; Step S202: Establish a CFD fluid-solid coupling simulation model based on the parameters obtained in step S1 and the heat generated by the bearing calculated by the transient energy equation in step S201: Based on the parameters obtained in step S1, the Fluent software is used to establish a bearing fluid domain model and divide the grid, and a fluid domain model including the flow space inside the bearing is established. The RNGk-ε turbulence model is used to describe the flow of oil and gas two phases inside the cylindrical roller bearing; the multiple reference system method is used to describe the fluid movement inside the bearing, and the transient flow is converted into a steady state for approximate solution; Boundary condition setting: set the inside of the bearing to a two-phase flow of lubricating oil and air, where air is the main phase and lubricating oil is the secondary phase; set the inner ring wall of the bearing to a rotating wall boundary, the outer ring wall of the bearing to a stationary wall boundary, and the retainer wall to a rotating wall boundary, and the lubricating oil inlet and outlet are distributed on both sides of the bearing; at the same time, set temperature boundary conditions for the inner and outer rings of the bearing and the rolling element wall in combination with the heat generated by the bearing calculated by the transient energy equation in step S201; Step S203: Establish an ADAMS multi-body dynamics simulation model based on the parameters obtained in step S1: The bearing finite element model obtained in step S101 is imported into the ADAMS multi-body dynamics simulation software, and the dynamic and static friction coefficient of the bearing, the elastic modulus and Poisson's ratio of each bearing component and the density parameters obtained in step S103 are input into the ADAMS multi-body dynamics simulation model; Step S204, combining the bearing numerical calculation model established in step S201, the CFD fluid-solid coupling simulation model established in step S202, and the ADAMS multi-body dynamics simulation model established in step S203 to perform a joint simulation to obtain the bearing contact micro-area temperature, mechanical parameters, and motion parameters; The step S4 comprises: Step S401: Use the maximum mean difference to evaluate the bearing simulation data and test data obtained in step S3: Assume that the bearing simulation data and test data obtained in step S3 are as follows: S={s1,s2,...,s 10 }={T rs ,T os ,T is ,T cas ,W bas ,W brs ,a xs ,a ys ,a zs ,ω ins ,ω cages } (14) E={e1,e2,...,e 10 }={T re ,T oe ,T ie ,T cae ,W bae ,W bre ,a xe ,a ye ,a ze ,ω ine ,ω cagee } (15) Where: S and E are the bearing simulation data set and test data set respectively; T rs 、T os 、T is and T cas are the temperatures of the bearing rolling element, outer ring, inner ring and cage measured by simulation; W bas and W brs are the axial and radial loads measured by simulation; a xs 、a ys and a zs are the vibration accelerations of the bearing in the x, y and z directions measured by simulation; ω ins and ω cages are the speeds of the inner ring and cage of the bearing measured by simulation; T re 、T oe 、T ie and T cae W are the temperatures of the bearing rolling element, outer ring, inner ring and cage measured in the test; bae and W bre are the axial and radial loads measured in the test; a xe 、a ye and a ze are the vibration accelerations of the bearing in the x, y and z directions measured in the test; ω ine and ω cagee are the experimentally measured speeds of the inner ring and cage of the bearing, respectively; the empirical estimate of the distance between the two distributions defined by the maximum mean difference is: Where: ||·|| H represents the modulus of the vector in the universal reproducing kernel Hilbert space; φ(·) is the nonlinear mapping from the original feature space to the reproducing kernel Hilbert space; Dist(·) represents the maximum average difference distance between the two distributions; Step S402: If the maximum mean difference obtained in step S401 does not meet the consistency requirement, the lubricating oil flow and viscosity parameters need to be adjusted. If the consistency requirement is met, a twin simulation model is obtained: If the maximum mean difference obtained in step S401 is greater than 1, that is, the consistency requirement is not met, it is necessary to adjust the lubricating oil flow in the bearing multi-domain simulation model established in step S2 until the maximum mean difference obtained in step S401 is less than or equal to 1. At this time, it can be considered that the established bearing multi-domain simulation model is a twin simulation model that is completely consistent with the motion state and lubrication characteristics of the test bearing; Step S403: extracting the micro-scale lubrication state of the bearing based on the twin simulation model obtained in step S402: Under different operating and lubrication conditions of the bearing, the temperature field distribution, oil film thickness and lubricating oil distribution at a certain microscopic scale inside the bearing are observed based on the bearing twin simulation model obtained in step S402 to achieve accurate monitoring of the microscale lubrication state of the bearing.

2. A bearing micro-scale lubrication state monitoring method integrating model and data according to claim 1, characterized in that: The step S104 specifically includes: Step S1041, determining the elastic modulus and Poisson's ratio of the bearing according to modal test and model correction technology; Step S1042: Calculate the volumes of the inner ring, outer ring, cage and rolling element according to the bearing geometric model established in step S101, measure the mass of each bearing component, and calculate the density of each bearing component using the following formula: Where: n = a, b, c, d represent the bearing rolling element, inner ring, outer ring and cage respectively; m n is the mass of different bearing components; V n is the volume of different parts of the bearing; ρ n is the density of different parts of the bearing; Step S1043: Determine the specific heats of the bearing rolling element and the inner ring, respectively. a and c b , determine the heat transfer coefficients of the bearing rolling elements and inner ring respectively as k a and k b .

3. The method for monitoring the micro-scale lubrication state of a bearing by integrating a model and data according to claim 1, characterized in that: The step S201 specifically includes: S2011, Elastohydrodynamic lubrication analysis of bearings The elastohydrodynamic lubrication theory considering time-varying effects, surface roughness and thermal effects includes three parts: the transient Reynolds equation, the transient film thickness equation considering surface roughness and the transient energy equation. The specific calculation formula is as follows: Transient Reynolds equation: Where: x j and j is the coordinate value of the jth rolling element; j = 1, 2, ..., Z, where Z is the number of rolling elements; η is the viscosity of the lubricating oil; ρ is the density of the lubricating oil; p j is the lubricating oil pressure at the jth rolling element; h j is the oil film thickness at the jth rolling element; is the partial differential operator; t is the time; U is the average velocity of the contact point of the friction pair along the x-axis; V is the average velocity of the contact point of the friction pair along the y-axis; Transient film thickness equation considering surface roughness: Where: h j (x j ,y j ,t) is the jth rolling element with time t and position (x j ,y j ) is the oil film thickness related to time t; h0(t) is the film thickness at the center of the bearing related to time t; E' j is the equivalent elastic modulus of the jth rolling element; and are the jth rolling element along x j and j Surface roughness in the direction; is the surface roughness of the jth rolling element; Transient energy equation: Where: c p is the specific heat capacity of lubricating oil; T j is the temperature of the jth rolling element; k is the heat transfer coefficient of the lubricating oil; u is the velocity component of the fluid in the x-axis direction; v is the velocity component of the fluid in the y-axis direction; where the heat transfer equation between the rolling element and the inner ring of the bearing is as follows: Where: C a , C b are the specific heat capacities of the rolling elements and the inner ring of the bearing; U a and U b are the linear speeds of rolling element and inner ring of bearing respectively; z a and z b are the coordinate values ​​of the rolling element and the inner ring of the bearing along the z-axis respectively; S2012, Bearing Dynamics Analysis The dynamic analysis of the bearing requires the establishment of three parts: the rolling element dynamic equation, the cage dynamic equation and the bearing ring dynamic equation. The specific calculation formula is as follows: Rolling element dynamics equation: Where: NP is the number of slices of the rolling element along its axis; W is the width of the slice; A is a constant; F c The centrifugal force on the rolling element; and F i j The jth rolling element bears the drag force of the inner raceway and the outer raceway respectively; and The normal force acting on the jth rolling element by the cage; and The tangential drag force of the cage on the jth rolling element; and are the resistances on the surface and end of the jth rolling element respectively; is the drag force of the jth rolling element guide rib on the rolling element end surface; is the blocking torque of the jth rolling element rib on the rolling element; is the angular acceleration of the jth rolling element rotating around its own axis; is the angular velocity of the jth rolling element; m is the mass of the rolling element; r is the radius of the rolling element; R is the radius of the cage guide surface; g 1ij and g 2ij are the contact deformations of the i-th slice of the j-th rolling element with the inner raceway and the outer raceway, i = 1, 2, ..., NP; l is the effective contact length of the rolling element; Cage dynamic equation: Where: m cage is the mass of the cage; φ j is the azimuth angle of the jth rolling element; T CDO T is the traction torque on the outer surface of the cage; CDS is the side blocking torque of the cage; T land is the torque generated by the fluid dynamic pressure; I cage is the moment of inertia of the cage; α is the central angle of the cage pocket; W x and W y The loads along the x-axis and y-axis respectively; and The displacement acceleration of the cage along the x-axis and y-axis respectively; Dynamic equation of the inner ring of the bearing: Where: and are the normal contact forces of the outer raceway and the inner raceway on the jth rolling element respectively; and are the friction torques between the jth rolling element and the outer ring raceway and the inner ring raceway respectively; F y and F z are the loads along the y-axis and z-axis respectively; M y and M z The moments along the y and z axes respectively.

4. The method for monitoring the micro-scale lubrication state of a bearing by integrating a model and data according to claim 1, characterized in that: The step S204 specifically includes: Step S2041: Perform joint simulation based on the bearing numerical calculation model established in step S201 and the CFD fluid-solid coupling simulation model established in step S202: According to the heat generated by the bearing calculated in step S201, it is used as the boundary condition of the CFD fluid-solid coupling simulation model established in step S202 to obtain the bearing temperature field distribution, lubricating oil distribution and lubricating oil film thickness; at the same time, the obtained temperature and oil film thickness parameters are fed back to the bearing numerical calculation model established in step S201 for coupling analysis; Step S2042: Based on the joint simulation model established in step S2041, the joint simulation model is further combined with the ADAMS multi-body dynamics simulation model established in step S203 to obtain a bearing multi-domain simulation model: The bearing numerical calculation model established in step S201 is written in Fortran language, and is further jointly simulated with the ADAMS multi-body dynamics simulation model established in step S203 in combination with the joint simulation result obtained in step S2041; wherein, the bearing contact micro-area temperature, mechanical parameters and motion parameters obtained by the joint simulation model of step S2041 are input into the ADAMS multi-body dynamics simulation model, and the interaction force and deformation extracted from the ADAMS multi-body dynamics simulation model are input into the joint simulation model of step S2041 for solution, thereby obtaining the bearing multi-domain simulation model.

5. A bearing micro-scale lubrication state monitoring method integrating model and data according to claim 4, characterized in that: The step S3 comprises: Step S301, collecting test data of bearing inner ring temperature, outer ring temperature, rolling element temperature, cage temperature, axial load, radial load, vibration acceleration, and bearing inner ring and cage speeds under certain radial load, axial load, and bearing inner ring speed conditions; Step S302, based on the bearing multi-domain simulation model obtained in step S2042, respectively collects simulation data of the bearing inner ring temperature, outer ring temperature, rolling element temperature, cage temperature, axial load, radial load, vibration acceleration, and bearing inner ring and cage speed under the same radial load, axial load and bearing inner ring speed as step S301.

Images