[go: up one dir, main page]

CN111159827A - High-speed rail bearing probability fatigue reliability assessment method - Google Patents

High-speed rail bearing probability fatigue reliability assessment method Download PDF

Info

Publication number
CN111159827A
CN111159827A CN201911376682.5A CN201911376682A CN111159827A CN 111159827 A CN111159827 A CN 111159827A CN 201911376682 A CN201911376682 A CN 201911376682A CN 111159827 A CN111159827 A CN 111159827A
Authority
CN
China
Prior art keywords
speed rail
damage
fatigue
rail bearing
probability
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.)
Granted
Application number
CN201911376682.5A
Other languages
Chinese (zh)
Other versions
CN111159827B (en
Inventor
张小玲
刘嘉庆
张鸿鑫
凌丹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201911376682.5A priority Critical patent/CN111159827B/en
Publication of CN111159827A publication Critical patent/CN111159827A/en
Application granted granted Critical
Publication of CN111159827B publication Critical patent/CN111159827B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

本发明公开了一种高铁轴承概率疲劳可靠性评估方法，针对L10寿命公式修正复杂，本发明的方法首先对材料的分散性和载荷的随机性进行量化，基于数论选点方法选出随机因素代表点；其次根据选出的代表点，建立高铁轴承的三维有限元仿真模型，进一步分析随机因素影响下轴承的接触应力，并建立最大接触应力的BP神经网络代理模型；然后通过载荷谱编制和材料的S‑N曲线，运用有限元仿真方法分析轴承的疲劳寿命，进一步根据疲劳损伤累积原理计算轴承的累积损伤和小于临界损伤的可靠性，为高精度、高效率的刻画高铁轴承的失效物理和疲劳寿命评估提供一种有效方法。

Figure 201911376682

The invention discloses a probabilistic fatigue reliability evaluation method for high-speed rail bearings, which is complicated to correct the L10 life formula. The method of the invention first quantifies the dispersion of materials and the randomness of loads, and selects random factors based on the point selection method of number theory. Representative points; secondly, according to the selected representative points, establish a three-dimensional finite element simulation model of the high-speed rail bearing, further analyze the contact stress of the bearing under the influence of random factors, and establish a BP neural network proxy model for the maximum contact stress; The S-N curve of the material, the finite element simulation method is used to analyze the fatigue life of the bearing, and the cumulative damage of the bearing and the reliability less than the critical damage are calculated according to the fatigue damage accumulation principle, and the failure physics of the high-speed rail bearing is described with high precision and high efficiency. and fatigue life assessment provides an effective method.

Figure 201911376682

Description

High-speed rail bearing probability fatigue reliability assessment method
Technical Field
The invention belongs to the field of reliability of mechanical parts, and particularly relates to a high-speed rail bearing probability fatigue reliability assessment method.
Background
At present, China has become the world with the highest mileage, driving speed and traffic volume of high-speed railways. As a basic key part of a high-speed rail, the localization of a bearing is a key problem which needs to be solved by the high-speed rail technology in China. The bearing of the high-speed rail main transmission case bears high rotating speed, random load and strong impact in the operation process, and the physical failure is complex. L in international standard ISO rolling bearing rated dynamic load and rated service life10The lifetime is a test and calculation under a predetermined condition. The influence of random eccentric load, improper installation, material property dispersion and other factors on the L in the specific operation of the bearing10And correcting the life formula. The failure mechanism of the fatigue pitting corrosion of the high-speed rail rolling bearing is that under the continuous action of alternating contact stress, local surface metal of a rolling body or a raceway falls off, so that the bearing generates vibration and noise to fail. The bearing failure physics is described from the failure mechanism of the bearing based on the fatigue damage accumulation principle, and a data basis and a technical support can be provided for sample performance test and reliability evaluation in the bearing research and development process. However, the fatigue life is dispersed due to manufacturing errors, material property dispersion and load randomness, so that the error of estimating and predicting the bearing life according to a fixed value is large.
Disclosure of Invention
For L10The service life formula is complex to correct, load randomness and material dispersibility are not considered on the basis of a fatigue damage accumulation principle, and the invention provides a high-speed rail bearing probability fatigue reliability evaluation method.
The technical scheme of the invention is as follows: a high-speed rail bearing probability fatigue reliability assessment method comprises the following steps:
s1, quantifying the dispersibility and random working conditions of the high-speed rail bearing material, and determining a main failure mode and a failure mechanism;
s2, calculating random radial force and axial force borne by the bearing according to the uncertainty information of the high-speed rail bearing material and the random working condition quantized in the step S1, and selecting a random factor representative point;
s3, establishing a finite element model of the high-speed rail bearing according to the random factor representative points selected in the step S2, carrying out statics analysis, and analyzing the maximum contact stress at each random factor representative point.
And S4, establishing a neural network proxy model of the maximum contact stress under the random factors according to the maximum contact stress at each random factor representative point in the step S3.
And S5, compiling a load spectrum by adopting a rain flow counting method, and analyzing the fatigue life under each stress level by using a finite element simulation method according to the S-N curve of the material.
S6, establishing an S-N curve of the high-speed rail bearing according to the fatigue life of each stress level in the step S5, calculating the fatigue life and probability distribution under each working condition, and calculating the accumulated damage probability distribution and parameters thereof;
and S7, calculating the probability distribution and parameters of the critical accumulated damage, and further calculating the reliability of reaching the critical damage.
Further, step S2 selects a random factor representative point based on a number theory point selection method.
Further, the step S4 is to establish a BP neural network proxy model of the maximum contact stress of the bearing under random factors.
Further, in step S5, a load spectrum is specifically compiled by a rain flow counting method.
Further, the step S2 specifically includes the following sub-steps:
s21, calculating random radial force and axial force borne by the bearing according to the uncertainty information of the high-speed rail bearing material and the random working condition quantized in the step S1;
s22, obtaining a S-dimensional generated vector (N, h) by a grid Point (GLP) set in a standard independent space1,h2,...,hS) Further, a unit hypercube [ 01 ] is obtained from the formula (1)]SInner point set:
Figure RE-GDA0002375670480000021
wherein N is the number of point sets to be constructed, hSIs Fibonacci sequence, and the Fibonacci sequence is defined by a recursive method as follows: h isj＝hj-1+hj-2,h0＝h11(n ═ 1, 2.., S), int (·) denotes taking the largest integer no less than or equal to that in parentheses;
s23, taking the limit of the normalized random variable as L, and carrying out scaling and translation transformation on the point set generated in the step S22 by using the formula (2) to obtain a square [ -L, L]SInner uniformly distributed point set:
θj,k＝2(xj,k-0.5)L,(k＝1,2,…,N,j＝1,2,…,S) (2)
and S24, transforming the sample points of the standard independent space to the sample points corresponding to the original space by utilizing Nataf inverse transformation, wherein the transformed sample points are input representative points.
Further, the step S3 specifically includes the following sub-steps:
s31, establishing a three-dimensional model of the high-speed rail bearing in three-dimensional drawing software;
s32, importing the high-speed rail bearing three-dimensional model established in the step S31 into ANSYS WORKBENCH;
s33, pre-processing the model imported in the step S32 such as network division, contact setting, constraint setting and the like;
s34, carrying out statics analysis on the model processed in the step S33 to obtain the maximum contact stress at each random factor representative point.
Further, the step S4 specifically includes the following sub-steps:
s41, setting the structural parameters of the neural network by taking the random factor representative points selected in the step S3 as input and the maximum contact stress at each representative point obtained in the step S34 as output;
and S42, training the neural network established in the step S41, and establishing a BP neural network agent model of the maximum contact stress under the action of random factors of the high-speed rail bearing.
Further, the step S5 specifically includes the following sub-steps:
s51, compiling a load spectrum by a rain flow counting method;
and S52, analyzing the fatigue life of the high-speed rail bearing under each stress level by using a finite element simulation method according to the load spectrum and the S-N curve of the material.
Further, the step S6 specifically includes the following sub-steps:
s61, according to the fatigue life of each stress level in the step S5, the fatigue life of each working condition is calculated according to the established S-N curve of the high-speed rail bearing, and the fatigue life distribution is evaluated.
S62, converting the probability distribution of the fatigue life into the probability distribution of the damage by adopting a one-to-one probability density conversion method according to the fatigue life and the distribution thereof calculated in S61:
fn(Nfi)dNfi＝fD(Di)dDi(3)
wherein f isn(. is a probability density function of fatigue life, fD(. is a probability density function of damage, NfiIs the fatigue life at the i-th stress level, DiDamage at the ith stress level.
And S63, calculating the cumulative damage distribution and the distribution parameters thereof according to the cumulative damage formula (4) and the damage distributions.
Figure RE-GDA0002375670480000031
Wherein k represents the stress order, NfiDenotes the fatigue life at the i-th stress level, niRepresenting the actual number of cycles of the ith stress stage, DSExpressed as the accumulation of damage, r, at a certain number of cycles for each stress leveliIs the stress level correlation coefficient.
Further, the step S7 specifically includes the following sub-steps:
s71, calculating the critical damage and the distribution thereof according to the formula (5):
Figure RE-GDA0002375670480000032
wherein N ismaxRepresenting the fatigue life value corresponding to the maximum first-order load in the applied load series, c and d are material constants, rmaxFor rate of lesion development, mmaxNumber of damaged nuclei for the material, DCRIs a critical fatigue damage.
S72, calculating the fatigue reliability by using a Monte Carlo simulation method according to the formula (6):
R＝P(ZD＝DCR-DS≥0) (6)
wherein D isSExpressed as the accumulation of damage at a certain number of cycles per stress level, DCRFor critical fatigue damage, P (-) is the probability that the bearing will not fail.
The invention has the beneficial effects that: for L10The method for evaluating the probability fatigue reliability of the high-speed rail bearing comprises the steps of firstly quantifying the dispersibility of materials and the randomness of loads, and selecting random factor representative points based on a number theory point selection method; secondly, establishing a three-dimensional finite element simulation model of the high-speed rail bearing according to the selected representative points, further analyzing the contact stress of the bearing under the influence of random factors, and establishing a BP neural network agent model of the maximum contact stress; and then, the fatigue life of the bearing is analyzed by a finite element simulation method through load spectrum compilation and an S-N curve of a material, the accumulated damage of the bearing and the reliability of reaching critical damage are further calculated according to the fatigue damage accumulation principle, and an effective method is provided for high-precision and high-efficiency failure physics depicting of the high-speed rail bearing and fatigue life assessment.
Drawings
Fig. 1 is a schematic flow chart of a high-speed rail bearing probability fatigue reliability evaluation method according to an embodiment of the invention.
Fig. 2 is a schematic view of a high-speed rail bearing according to an embodiment of the present invention, in which (a) is a three-dimensional model diagram of the high-speed rail bearing, and (b) is a maximum contact stress cloud diagram of the high-speed rail bearing.
FIG. 3 is a diagram of a neural network architecture for maximum contact stress in accordance with an embodiment of the present invention.
FIG. 4 is an S-N curve of GCr15 bearing steel according to an embodiment of the present invention.
Fig. 5 is a fatigue life probability distribution diagram of the high-speed rail bearing according to the embodiment of the invention during acceleration.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In order to improve the fatigue damage accumulation evaluation accuracy, the dispersion of material performance and the randomness of load need to be considered. In the absence of enough data to describe the dispersibility of the material, the elastic modulus of the material is described by intervals, the random working condition of the bearing is described by random variables, and how to select a test design point under the mixing uncertainty is a key problem. The scheme provides a high-speed rail bearing probability fatigue life assessment method under mixed uncertainty. The method has the advantages that the problem of test design point selection of multidimensional mixed uncertainty factors can be solved through number theory point selection; and a proxy model of the random factor representative point and the contact stress is established based on the BP neural network, so that the efficiency of probability fatigue life evaluation can be improved. The method can be an effective way for accurately depicting the high-speed rail bearing failure physics and improving the bearing performance evaluation precision, and the specific flow is shown in figure 1 and comprises the following steps:
s1, quantifying the dispersibility and random working conditions of the high-speed rail bearing material, and determining a main failure mode and a failure mechanism;
s2, calculating random radial force and axial force borne by the bearing according to the uncertainty information of the high-speed rail bearing material and the random working condition quantized in the step S1, and selecting random factor representative points based on a number theory point selection method;
s3, establishing a finite element model of the high-speed rail bearing according to the random factor representative points selected in the step S2, carrying out statics analysis, and analyzing the maximum contact stress at each random factor representative point.
S4, establishing a BP neural network proxy model of the maximum contact stress under the random factors according to the maximum contact stress at each random factor representative point in the step S3.
And S5, compiling a load spectrum by adopting a rain flow counting method, and analyzing the fatigue life under each stress level by using a finite element simulation method according to the S-N curve of the material.
S6, establishing an S-N curve of the bearing according to the fatigue life of each stress level in the step S5, and calculating the fatigue life and probability distribution under each working condition, and the accumulated damage probability distribution and parameters thereof;
and S7, calculating the probability distribution and parameters of the critical accumulated damage, and further calculating the reliability of reaching the critical damage.
In step S1, the embodiment of the present invention takes a high-speed rail bearing as an example, quantifies randomness and random conditions of the performance of the high-speed rail bearing material, and determines a main failure mode and a failure mechanism. The method specifically comprises the following steps:
s11, the uncertainty of the material performance of the high-speed rail bearing is described by interval variables, and the elastic modulus of each component material of the bearing is shown in Table 1;
s12, determining three random working conditions of the high-speed rail bearing in the invention: acceleration, uniform speed and deceleration. Respectively corresponding to high-speed rail outbound, operation and inbound;
s13, determining the main failure mode of the high-speed rail bearing as fatigue pitting;
s14, the failure mechanism of fatigue pitting corrosion is that under the action of load, contact stress is generated between the rolling body and the inner and outer raceways. When the bearing rotates, the contact stress is changed circularly, and after the bearing works for a plurality of times, the local surface metal of the rolling body or the raceway falls off, so that the bearing generates vibration and noise to lose efficacy.
TABLE 1 uncertainty description of high-speed rail bearing materials
Component part Material density/kg m-3 Modulus of elasticity/GPa Poisson ratio Material
Inner ring 7830 [215 224] 0.3 GCr15
Outer ring 7830 [215 224] 0.3 GCr15
Roller 7830 [215 224] 0.3 GCr15
Holding rack 1370 [6 10] 0.3 GRPA66
In step S2, according to the quantitative information of uncertainty of the high-speed rail bearing material and the random working condition determined in step S1, calculating the random radial force and the random axial force borne by the bearing, and selecting a random factor representative point based on a number theory point selection method specifically includes the following sub-steps:
s21, calculating according to the random working conditions determined in the step S1 to obtain the distribution of random radial force and axial force born by the bearing as shown in a table 2;
TABLE 2 radial and axial forces (KN) to which the bearing is subjected under different operating conditions
Figure RE-GDA0002375670480000051
Quantifying the elastic modulus of the inner ring, the outer ring and the roller and the elastic modulus of the retainer by using interval variables for uncertainty of the high-speed rail bearing material in the steps S22 and S21; the axial and radial forces experienced by the bearing are described by randomly distributed variables. Therefore, in the standard independent space, 4-dimensional generated vectors (N, h) are obtained by means of a grid Point (GLP) set1,h2,h3,h4) Further, a unit hypercube [ 01 ] is obtained from the formula (7)]SInner point set:
Figure RE-GDA0002375670480000052
wherein, N is the number of point sets to be constructed, N is 30, hSIs Fibonacci sequence, and the Fibonacci sequence is defined by a recursive method as follows: h isj＝hj-1+hj-2,h0＝h11(n ═ 1, 2.., S), int (·) denotes taking the largest integer equal to or less than that in parentheses.
S23, taking the limit of the normalized random variable as L, where L is 2 in the present invention, and performing scaling and translation transformation on the point set generated in step S22 using equation (8) to obtain a square [ -L, L]SInner uniformly distributed point set:
θj,k＝2(xj,k-0.5)L,(k＝1,2,…,N,j＝1,2,…,S) (8)
and S24, transforming the sample points of the standard independent space to the sample points corresponding to the original space by utilizing Nataf inverse transformation, wherein the transformed sample points are input representative points.
In step S3, a finite element model of the high-speed rail bearing is built according to the random factor representative points selected in step S2, and a statics analysis is performed to analyze the maximum contact stress at each random factor representative point. The method specifically comprises the following steps:
s31, establishing a three-dimensional model of the high-speed rail bearing in three-dimensional drawing software, as shown in figure 2 (a);
s32, importing the high-speed rail bearing three-dimensional model established in the step S31 into ANSYS WORKBENCH;
s33, pre-processing the model imported in the step S32 such as network division, contact setting, constraint setting and the like;
s34, the model processed in step S33 is subjected to statics analysis, and the maximum contact stress at each random factor representative point is obtained, as shown in fig. 2 (b).
In step S4, a BP neural network proxy model of the maximum contact stress under random factors is established according to the maximum contact stress at each random factor representative point in step S3.
S41, setting the structural parameters of the neural network by taking the random factor representative points selected in the step S3 as input and the maximum contact stress at each representative point obtained in the step S34 as output, as shown in FIG. 3;
and S42, training the neural network established in the step S41, and establishing a BP neural network agent model of the maximum contact stress under the action of random factors.
In step S5, a load spectrum is constructed by rain flow counting, and fatigue life at each stress level is analyzed by finite element simulation based on the S-N curve of the material (as shown in FIG. 4). The method specifically comprises the following steps:
s51, compiling a load spectrum by a rain flow counting method;
and S52, analyzing the fatigue life of the high-speed rail bearing under each stress level by using a finite element simulation method according to the load spectrum and the S-N curve of the material.
In step S6, an S-N curve of the bearing is created based on the fatigue life at each stress level in step S5, and the fatigue life and probability distribution (fig. 5 is the fatigue life probability distribution at the time of acceleration of the high-speed rail bearing) under each condition are calculated and the cumulative damage probability distribution and parameters thereof are calculated. The method specifically comprises the following steps:
s61, according to the fatigue life of each stress level in the step S5, the fatigue life of each working condition is calculated according to the established S-N curve of the high-speed rail bearing, and the fatigue life distribution is evaluated.
S62, converting the probability distribution of the fatigue life into the probability distribution of the damage by adopting a one-to-one probability density conversion method according to the fatigue life and the distribution thereof calculated in S61:
fn(Nfi)dNfi＝fD(Di)dDi(9)
wherein f isn(. is a probability density function of fatigue life, fD(. is a probability density function of damage, NfiIs the fatigue life at the i-th stress level, DiDamage at the ith stress level.
S63, calculating the cumulative damage distribution and its distribution parameters from the cumulative damage formula (10) and the distribution of each damage.
Figure RE-GDA0002375670480000071
Wherein k represents the stress order, NfiDenotes the fatigue life at the i-th stress level, niRepresenting the actual number of cycles of the ith stress stage, DSExpressed as the accumulation of damage, r, at a certain number of cycles for each stress leveliIs a stress correction factor.
In step S7, the critical cumulative damage probability distribution and parameters are calculated, and the reliability of reaching the critical damage is further calculated. The method specifically comprises the following steps:
s71, calculating the critical damage and the distribution thereof according to the formula (11):
Figure RE-GDA0002375670480000072
wherein N ismaxRepresenting the fatigue life value corresponding to the maximum first-order load in the applied load series, c,d is the material constant, rmaxFor rate of lesion development, mmaxNumber of damaged nuclei for the material, DCRIs a critical fatigue damage. It is proposed here to take
Figure RE-GDA0002375670480000073
S72, calculating the fatigue reliability by using a Monte Carlo simulation method according to the formula (12):
R＝P(ZD＝DCR-DS≥0) (12)
wherein D isSExpressed as the accumulation of damage at a certain number of cycles per stress level, DCRFor critical fatigue damage, p (-) is the probability that the bearing will not fail.
In summary, it can be seen that for L10The service life formula is modified complicatedly, and the probability fatigue reliability evaluation method for the high-speed rail bearing selects random factor representative points based on a number theory point selection method by quantifying the dispersibility of materials and the randomness of loads; secondly, establishing a three-dimensional finite element simulation model of the high-speed rail bearing according to the selected representative points, further analyzing the contact stress of the bearing under the influence of random factors, and establishing a BP neural network proxy model of the contact stress of the bearing under uncertain factors; and then analyzing the fatigue life of the bearing under each stress level by load spectrum compilation and finite element simulation methods, further calculating the accumulated damage of the bearing and the reliability of the accumulated damage smaller than the critical damage according to the fatigue damage accumulation principle, and providing an effective method for accurately depicting the failure physics and predicting the life.

Claims (10)

1. A high-speed rail bearing probability fatigue reliability assessment method comprises the following steps:
s1, quantifying the dispersibility and random working conditions of the high-speed rail bearing material, and determining a main failure mode and a failure mechanism;
s2, calculating random radial force and axial force borne by the bearing according to the uncertainty information of the high-speed rail bearing material and the random working condition quantized in the step S1, and selecting a random factor representative point;
s3, establishing a finite element model of the high-speed rail bearing according to the random factor representative points selected in the step S2, carrying out statics analysis, and analyzing the maximum contact stress at each random factor representative point.
S4, establishing a BP neural network proxy model of the maximum contact stress under the random factors according to the maximum contact stress at each random factor representative point in the step S3.
And S5, compiling a load spectrum, and analyzing the fatigue life under each stress level by using a finite element simulation method according to the S-N curve of the material.
S6, establishing an S-N curve of the bearing according to the fatigue life of each stress level in the step S5, and calculating the fatigue life and probability distribution under each working condition, and the accumulated damage probability distribution and parameters thereof;
and S7, calculating the probability distribution and parameters of the critical accumulated damage, and further calculating the reliability of reaching the critical damage.
2. The method for evaluating the probability fatigue reliability of the high-speed rail bearing according to claim 1, wherein the step S2 is to select the random factor representative point based on a number theory point selection method.
3. The method as claimed in claim 1, wherein the step S4 is performed to establish a BP neural network proxy model of maximum contact stress under random factors.
4. The method for evaluating the probabilistic fatigue reliability of a high-speed rail bearing according to claim 1, wherein the step S5 is to compile a load spectrum by a rain flow counting method.
5. The method for evaluating the probability fatigue reliability of the high-speed rail bearing according to claim 2, wherein the step S2 specifically comprises the following sub-steps:
s21, calculating random radial force and axial force borne by the bearing according to the uncertainty information of the high-speed rail bearing material and the random working condition quantized in the step S1;
s22, in the standard independent space,obtaining an S-dimensional generated vector (N, h) by means of a set of Good Lattice Points (GLP)1,h2,...,hS) Further, a unit hypercube [ 01 ] is obtained from the formula (1)]SInner point set:
Figure RE-FDA0002375670470000011
wherein N is the number of point sets to be constructed, hSIs Fibonacci sequence, and the Fibonacci sequence is defined by a recursive method as follows: h isj＝hj-1+hj-2,h0＝h11(n ═ 1, 2.., S), int (·) denotes taking the largest integer no less than or equal to that in parentheses;
s23, taking the limit of the normalized random variable as L, and carrying out scaling and translation transformation on the point set generated in the step S22 by using the formula (2) to obtain a square [ -L, L]SInner uniformly distributed point set:
θj,k＝2(xj,k-0.5)L,(k＝1,2,…,N,j＝1,2,…,S) (2)
and S24, transforming the sample points of the standard independent space to the sample points corresponding to the original space by utilizing Nataf inverse transformation, wherein the transformed sample points are input representative points.
6. The method for evaluating the probability fatigue reliability of the high-speed rail bearing according to claim 5, wherein the step S3 specifically comprises the following sub-steps:
s31, establishing a three-dimensional model of the high-speed rail bearing in three-dimensional drawing software;
s32, importing the high-speed rail bearing three-dimensional model established in the step S31 into ANSYS WORKBENCH;
s33, pre-processing the model imported in the step S32 such as network division, contact setting, constraint setting and the like;
s34, carrying out statics analysis on the model processed in the step S33 to obtain the maximum contact stress at each random factor representative point.
7. The method for evaluating the probability fatigue reliability of the high-speed rail bearing according to claim 6, wherein the step S4 specifically comprises the following sub-steps:
s41, setting the structural parameters of the neural network by taking the random factor representative points selected in the step S3 as input and the maximum contact stress at each representative point obtained in the step S34 as output;
and S42, training the neural network established in the step S41, and establishing a BP neural network agent model of the maximum contact stress under the action of random factors.
8. The method for assessing the probability fatigue reliability of the high-speed rail bearing according to claim 7, wherein the step S5 specifically comprises the following sub-steps:
s51, compiling a load spectrum by a rain flow counting method;
and S52, analyzing the fatigue life of the high-speed rail bearing at each stress level according to the load spectrum and the S-N curve of the material.
9. The method for assessing the probability fatigue reliability of the high-speed rail bearing according to claim 8, wherein the step S6 specifically comprises the following substeps:
s61, according to the fatigue life of each stress level in the step S5, the fatigue life of each working condition is calculated according to the established S-N curve of the high-speed rail bearing, and the fatigue life distribution is evaluated.
S62, converting the probability distribution of the fatigue life into the probability distribution of the damage by adopting a one-to-one probability density conversion method according to the fatigue life and the distribution thereof calculated in S61:
fn(Nfi)dNfi＝fD(Di)dDi(3)
wherein f isn(. is a probability density function of fatigue life, fD(. is a probability density function of damage, NfiIs the fatigue life at the i-th stress level, DiDamage at the ith stress level.
And S63, calculating the cumulative damage distribution and the distribution parameters thereof according to the cumulative damage formula (4) and the damage distributions.
Figure RE-FDA0002375670470000031
Wherein k represents the stress order, NfiDenotes the fatigue life at the i-th stress level, niRepresenting the actual number of cycles of the ith stress stage, DSExpressed as the accumulation of damage, r, at a certain number of cycles for each stress leveliIs the stress level correlation coefficient.
10. The method for assessing the probability fatigue reliability of the high-speed rail bearing according to claim 9, wherein the step S7 specifically comprises the following sub-steps:
s71, calculating the critical damage and the distribution thereof according to the formula (5):
Figure RE-FDA0002375670470000032
wherein N ismaxRepresenting the fatigue life value corresponding to the maximum first-order load in the applied load series, c and d are material constants, rmaxFor rate of lesion development, mmaxNumber of damaged nuclei for the material, DCRIs a critical fatigue damage.
S72, calculating the fatigue reliability by using a Monte Carlo simulation method according to the formula (6):
R＝P(ZD＝DCR-DS≥0) (6)
wherein D isSExpressed as the accumulation of damage at a certain number of cycles per stress level, DCRFor critical fatigue damage, P (-) is the probability that the bearing will not fail.
CN201911376682.5A 2019-12-27 2019-12-27 High-speed rail bearing probability fatigue reliability assessment method Expired - Fee Related CN111159827B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911376682.5A CN111159827B (en) 2019-12-27 2019-12-27 High-speed rail bearing probability fatigue reliability assessment method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911376682.5A CN111159827B (en) 2019-12-27 2019-12-27 High-speed rail bearing probability fatigue reliability assessment method

Publications (2)

Publication Number Publication Date
CN111159827A true CN111159827A (en) 2020-05-15
CN111159827B CN111159827B (en) 2021-11-30

Family

ID=70558540

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911376682.5A Expired - Fee Related CN111159827B (en) 2019-12-27 2019-12-27 High-speed rail bearing probability fatigue reliability assessment method

Country Status (1)

Country Link
CN (1) CN111159827B (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111896255A (en) * 2020-08-12 2020-11-06 上海理工大学 A rapid evaluation method for the service life of wheel hub bearings based on multi-axis random road loads
CN112560196A (en) * 2020-12-08 2021-03-26 福建工程学院 Bearing fatigue life prediction method for hoop-type giant tire unloader
CN113408166A (en) * 2021-06-07 2021-09-17 扬州大学 Reliability analysis method for self-adaptive ear height shifting-receiving type intelligent breeding system
CN113946998A (en) * 2021-10-22 2022-01-18 福建工程学院 Calculation method of fatigue reliability of tire unloader bearing based on heterodimension interference model
WO2022052333A1 (en) * 2020-09-08 2022-03-17 中国电器科学研究院股份有限公司 Polymer material service life prediction method based on environmental big data and machine learning
CN114509262A (en) * 2022-01-25 2022-05-17 中国铁建重工集团股份有限公司 Acceleration test method for main bearing of heading machine
CN114755028A (en) * 2022-03-17 2022-07-15 武汉理工大学 Test bed for reliability and durability of whole vehicle, control method and medium
CN114936494A (en) * 2022-05-23 2022-08-23 华东理工大学 Data physical fusion driven high-temperature component reliability evaluation method and system
CN116822083A (en) * 2023-06-27 2023-09-29 上海交通大学 Heavy-duty bearing impact damage numerical simulation reproduction method and system
CN112131797B (en) * 2020-09-28 2024-02-02 衢州学院 A spindle bearing life prediction and reliability assessment method based on stress analysis

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130006542A1 (en) * 2010-03-16 2013-01-03 Ntn Corporation Assessment of shear fatigue property of rolling contact metal material and estimation of fatigue limit maximum contact pressure using same assessment
US20150254382A1 (en) * 2011-09-16 2015-09-10 Sentient Corporation Method and system for predicting surface contact fatigue life
CN107103140A (en) * 2017-04-28 2017-08-29 电子科技大学 A kind of time-dependent fatigue reliability analysis method based on bilinearity accumulated damage
CN109960845A (en) * 2018-11-19 2019-07-02 南京林业大学 Design method for finite life of drive axle housing based on bench test
CN110147624A (en) * 2019-05-24 2019-08-20 重庆大学 A kind of Gear Contact Prediction method for fatigue life based on loading spectrum
CN110232452A (en) * 2019-06-12 2019-09-13 中国神华能源股份有限公司 Repair method and system based on track car team state of the art

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130006542A1 (en) * 2010-03-16 2013-01-03 Ntn Corporation Assessment of shear fatigue property of rolling contact metal material and estimation of fatigue limit maximum contact pressure using same assessment
US20150254382A1 (en) * 2011-09-16 2015-09-10 Sentient Corporation Method and system for predicting surface contact fatigue life
CN107103140A (en) * 2017-04-28 2017-08-29 电子科技大学 A kind of time-dependent fatigue reliability analysis method based on bilinearity accumulated damage
CN109960845A (en) * 2018-11-19 2019-07-02 南京林业大学 Design method for finite life of drive axle housing based on bench test
CN110147624A (en) * 2019-05-24 2019-08-20 重庆大学 A kind of Gear Contact Prediction method for fatigue life based on loading spectrum
CN110232452A (en) * 2019-06-12 2019-09-13 中国神华能源股份有限公司 Repair method and system based on track car team state of the art

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
XIAOLING ZHANG等: "A Multi-Team Competitive Optimization Algorithm for Bearing Fault Diagnosis", 《IEEE XPLORE》 *
李仁兴等: "随机载荷下滚动轴承系统疲劳可靠性分析", 《机床与液压》 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11835419B2 (en) 2020-08-12 2023-12-05 University Of Shanghai For Science And Technology Method for rapid assessment of service life of hub bearing based on multiaxial random road loading
CN111896255A (en) * 2020-08-12 2020-11-06 上海理工大学 A rapid evaluation method for the service life of wheel hub bearings based on multi-axis random road loads
WO2022052333A1 (en) * 2020-09-08 2022-03-17 中国电器科学研究院股份有限公司 Polymer material service life prediction method based on environmental big data and machine learning
CN112131797B (en) * 2020-09-28 2024-02-02 衢州学院 A spindle bearing life prediction and reliability assessment method based on stress analysis
CN112560196B (en) * 2020-12-08 2022-06-07 福建工程学院 Bearing fatigue life prediction method for hoop-type giant tire unloader
CN112560196A (en) * 2020-12-08 2021-03-26 福建工程学院 Bearing fatigue life prediction method for hoop-type giant tire unloader
CN113408166B (en) * 2021-06-07 2023-05-23 扬州大学 Reliability analysis method of adaptive spike Gao Bo receptor type intelligent breeding system
CN113408166A (en) * 2021-06-07 2021-09-17 扬州大学 Reliability analysis method for self-adaptive ear height shifting-receiving type intelligent breeding system
CN113946998A (en) * 2021-10-22 2022-01-18 福建工程学院 Calculation method of fatigue reliability of tire unloader bearing based on heterodimension interference model
CN113946998B (en) * 2021-10-22 2025-05-27 福建理工大学 Calculation method of fatigue reliability of tire unloader bearing based on heterodimensional interference model
CN114509262A (en) * 2022-01-25 2022-05-17 中国铁建重工集团股份有限公司 Acceleration test method for main bearing of heading machine
CN114755028A (en) * 2022-03-17 2022-07-15 武汉理工大学 Test bed for reliability and durability of whole vehicle, control method and medium
CN114936494A (en) * 2022-05-23 2022-08-23 华东理工大学 Data physical fusion driven high-temperature component reliability evaluation method and system
CN114936494B (en) * 2022-05-23 2025-02-11 华东理工大学 High temperature component reliability assessment method and system driven by data-physics fusion
CN116822083A (en) * 2023-06-27 2023-09-29 上海交通大学 Heavy-duty bearing impact damage numerical simulation reproduction method and system

Also Published As

Publication number Publication date
CN111159827B (en) 2021-11-30

Similar Documents

Publication Publication Date Title
CN111159827B (en) High-speed rail bearing probability fatigue reliability assessment method
Tian et al. A neural network approach for remaining useful life prediction utilizing both failure and suspension histories
CN106248381B (en) A dynamic prediction method of rolling bearing life based on multi-feature and phase space
Zhu et al. A simulation-data-driven subdomain adaptation adversarial transfer learning network for rolling element bearing fault diagnosis
CN111711820A (en) A Compression Sampling Method for Cracked Images Based on Generative Adversarial Networks
CN110096805B (en) Bridge structure parameter uncertainty quantification and transfer method based on improved self-service method
CN119614844B (en) Detection method and system for copper wire multi-pass rolling intermediate annealing
CN112068519A (en) A quality control method and device for product assembly process
CN117332688A (en) Prediction method and system for fatigue crack growth of X80 pipeline steel under random loading
CN120006787A (en) A servo control method for axial force of foundation pit steel support based on digital twin
CN119476016A (en) Mechanical structure reliability analysis method and system based on active learning agent model
Triviño et al. Damage detection and localization at the jacket support of an offshore wind turbine using transformer models
CN115186701A (en) Bearing life prediction method, device, electronic device and storage medium
CN114792054A (en) Method for determining reliability of chassis bearing part
CN104462836B (en) The System in Small Sample Situation for being segmented step plus strategy based on the complete period accelerates failure to develop experimental method
CN115902620B (en) A method and apparatus for enhanced diagnosis of motor faults based on recurrent generative adversarial networks
CN110414086B (en) Sensitivity-based comprehensive stress acceleration factor calculation method
CN120875113A (en) River water quality prediction method and system based on machine learning coupling hydrologic model
Zhao et al. Fuzzy scalar quantisation based on hidden Markov model and application in fault diagnosis of wind turbine
CN118133439A (en) A component quality control method and system based on digital twin technology
Gao et al. Dynamic reliability modeling of gear transmission system considering multiple failure modes based on copula functions
CN117591906A (en) A dynamic carbon emission factor calculation method based on coal-fired unit operating data
Orkisz et al. On improved evolutionary algorithms application to the physically based approximation of experimental data
CN113779866A (en) A cost calculation method and system for high-voltage switch GIS cylinder
Sun et al. Research on Bearing Vibration Signal Processing Based on MCMC-SPERMAN Algorith

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20211130

CF01 Termination of patent right due to non-payment of annual fee