Sliding bearing oil film distributed load identification method and system under time-varying uncertain factors

Abstract

The embodiment of the invention provides a sliding bearing oil film distributed load identification method and system under time-varying uncertain factors. The method is applied to the technical field of distributed dynamic load identification. The method comprises the following steps: extracting operation parameters of a transmission shaft system of a wind power gear box through finite element modeling and analysis; an axis track is collected, and a median function and an auto-covariance matrix are obtained through calculation based on the operation parameters; calculating to obtain a time-varying uncertain sliding bearing oil film distribution load median function based on the median function; the fluctuation interval of the time-varying uncertain sliding bearing oil film distribution load is obtained through calculation based on the auto-covariance matrix; and calculating the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load according to the median function and the fluctuation interval, and calculating the sliding bearing oil film distribution load. In this way, the range of the distributed dynamic load recognition technology is widened, the engineering practicability and universality of the load recognition technology are enhanced, and reasonable load conditions are provided for sliding bearing stability design and optimization.

Description

Technical Field

[0001] The present disclosure relates to the technical field of distributed dynamic load identification, and in particular to a method and system for identifying oil film distributed loads of sliding bearings under time-varying uncertain factors. Background Art

[0002] Sliding bearings are the most promising support design for the transmission shaft system of high-power wind turbine gearboxes. The oil film distribution load of the sliding bearings at the coupling interface with the gearbox changes over time due to the impact load of frequent start-stop and stop on the gearbox, yaw moment, wind rotor weight and unbalanced load and other non-torsion loads, resulting in oil film vortex and oscillation, which seriously affects the stability of the sliding bearings. It is an important basis for operating status monitoring and fault diagnosis of key components such as wind turbines. In the case that the mechanism of wind-fluid-solid coupling of wind power equipment has not been thoroughly studied, it is very difficult to theoretically obtain a complete and accurate time-effect calculation expression of the oil film distribution load during operation. The sliding bearings are built into the gearbox, and the changeable position and direction of the load bring difficulties to the arrangement of the sensor. The direct measurement method cannot be used to measure the oil film distribution load of the sliding bearings. The calculation inverse technology of load identification using testable response data is becoming an indispensable and important means of load acquisition in engineering practice.

[0003] However, in addition to being affected by the coupling interface load, the axis trajectory of wind turbine sliding bearings is also affected by the repeated propagation of multi-source uncertainty information such as bearing contact surface friction characteristics, oil film characteristics, transmission chain structure coupling deformation, manufacturing and installation errors, and measurement noise. The axis trajectory data sequence contains a large amount of variation information of constantly changing uncertainties. Effective monitoring of the operation and maintenance status of wind turbines requires accurate understanding of the dynamic fluctuations of the oil film distribution load of sliding bearings, and it is crucial to accurately identify the oil film distribution load of sliding bearings under time-varying uncertainties. However, the oil film distribution load of sliding bearings under time-varying uncertainties does not show a definite change law, resulting in irregular fluctuations; describing time-varying uncertainties through random processes requires obtaining a large number of samples of the axis trajectory, which is difficult and costly to obtain for wind turbines.

[0004] Therefore, providing a method for identifying the oil film distribution load of a sliding bearing considering time-varying uncertainties and a corresponding system is of great significance for understanding the current operation and maintenance status of wind turbines. Summary of the invention

[0005] The present invention provides a method and system for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors. The time-varying uncertain oil film distribution load of the sliding bearing is expressed as the sum of the median function of the oil film distribution load of the sliding bearing and the load uncertainty part, and then the identification of the oil film distribution load of the sliding bearing with time-varying uncertainty is decomposed into two types of deterministic problems: the identification of the median function of the original load and the identification of the autocovariance matrix of the load. This solves the technical problem that the existing technology cannot solve the irregular fluctuation of the oil film distribution load of the sliding bearing that affects the stable operation of key components of a wind turbine.

[0006] According to a first aspect of the present disclosure, a method for identifying oil film distribution load of a sliding bearing under time-varying uncertain factors is provided, comprising the following steps: Extract the operating parameters of the wind turbine gearbox transmission shaft system through finite element modeling and analysis; Collect the axis trajectory of the wind turbine gearbox transmission shaft system, and calculate the median function and autocovariance matrix of the axis trajectory based on the operating parameters; Based on the median function of the axis trajectory and using the distributed load identification method, the median function of the oil film distributed load of the time-varying uncertain sliding bearing is calculated; Based on the autocovariance matrix and using the concentrated load identification method, the fluctuation range of the time-varying uncertain oil film distribution load of the sliding bearing is calculated; The upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are calculated according to the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and the sliding bearing oil film distribution load is calculated.

[0007] According to the aspects described above and any possible implementation, an implementation is further provided, wherein the wind turbine gearbox transmission shaft system operating parameters include a mass matrix M, a stiffness matrix K, a damping matrix C and a Green's kernel function matrix G.

[0008] According to the above aspects and any possible implementation, an implementation is further provided, wherein the process of calculating the median function of the oil film distributed load of the time-varying uncertain sliding bearing based on the median function of the axis trajectory and using the distributed load identification method is as follows: The spatial distribution function of the time-varying uncertain oil film distribution load of the sliding bearing is reconstructed by fitting a set of independent basis functions using Chebyshev orthogonal polynomials. Based on the conventional distributed load identification method, the modal load is obtained, and the first-order mode with the largest contribution is used as the time history function of the time-varying uncertain sliding bearing oil film distributed load, and the median function of the time history function is obtained; The median function of the time-varying uncertain sliding bearing oil film distribution load can be obtained by multiplying the identified median function of the time history function and the spatial distribution function.

[0009] According to the above aspects and any possible implementation, an implementation is further provided, wherein the process of calculating the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load based on the autocovariance matrix and using the concentrated load identification method is as follows: Based on Mercer's theorem, the autocovariance matrix of the axis trajectory is spectrally decomposed to extract the eigenvalues ​​and eigenvectors of the autocovariance matrix of the axis trajectory; An equivalent eigenvector of the axis trajectory response is calculated based on the eigenvalue and the eigenvector, and an equivalent eigenvector of the load is identified based on a concentrated dynamic load identification method; Calculating an autocovariance matrix of an interval time history function based on an equivalent eigenvector of the load, and calculating a radius function of the time history function based on the autocovariance matrix of the interval time history function; The upper and lower boundaries of the time history function of the oil film distribution load are obtained based on the median function of the time history function and the radius function, and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load is obtained.

[0010] According to the above aspects and any possible implementation, an implementation is further provided, wherein the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are calculated based on the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and the process of calculating the sliding bearing oil film distribution load is: The time-varying uncertain sliding bearing oil film distribution load is analyzed to construct a functional relationship between the sliding bearing oil film distribution load and the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying sliding bearing oil film distribution load; Based on the functional relationship, the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are calculated; Based on the sliding bearing oil film distribution load characteristics, the sliding bearing oil film distribution load is obtained according to the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load.

[0011] According to the above aspects and any possible implementation, an implementation is further provided, wherein the process of performing an analysis on the time-varying uncertain sliding bearing oil film distribution load and constructing a functional relationship between the sliding bearing oil film distribution load and the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying sliding bearing oil film distribution load is: The interval orthogonal series expansion method is used to expand the time-varying uncertain oil film distribution load of sliding bearings. The expanded time-varying uncertain sliding bearing oil film distribution load is substituted into the wind turbine gear transmission shaft system dynamics differential equation, and the functional relationship between the sliding bearing oil film distribution load and the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying sliding bearing oil film distribution load is calculated.

[0012] According to the above aspects and any possible implementation, an implementation is further provided, wherein the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are: ; in, is the lower boundary of the oil film distribution load fluctuation range, is the upper boundary of the fluctuation range of the oil film distribution load, f(z) is the spatial distribution function of the time-varying uncertain sliding bearing oil film distribution load, is the median function of the time history function of the oil film distribution load, is the radius function of the time history function of the oil film distributed load, is the median function of the time-varying uncertain sliding bearing oil film distribution load, is the load fluctuation range of the time-varying uncertain sliding bearing oil film distribution.

[0013] According to a second aspect of the present disclosure, there is provided a sliding bearing oil film distribution load identification system under time-varying uncertain factors, comprising: a wind power gearbox transmission shaft system operation parameter acquisition module, an axis trajectory analysis module and a time-varying uncertain sliding bearing oil film distribution load identification module; The wind turbine gearbox transmission shaft system operating parameter acquisition module is used to extract the wind turbine gearbox transmission shaft system operating parameters through finite element modeling and analysis; The axis trajectory analysis module is used to collect the axis trajectory of the wind turbine gearbox transmission shaft system, and calculate the median function and autocovariance matrix of the axis trajectory based on the operating parameters; The time-varying uncertain sliding bearing oil film distribution load identification module is used to calculate the time-varying uncertain sliding bearing oil film distribution load according to the median function of the axis trajectory and the autocovariance matrix; According to the above aspects and any possible implementation, an implementation is further provided, wherein the time-varying uncertain sliding bearing oil film distribution load identification module includes a time-varying uncertain sliding bearing oil film distribution load median function calculation module, a time-varying uncertain sliding bearing oil film distribution load fluctuation range calculation module and a sliding bearing oil film distribution load identification module; The time-varying uncertain sliding bearing oil film distribution load median function calculation module is used to calculate the time-varying uncertain sliding bearing oil film distribution load median function based on the median function of the axis trajectory and using the distributed load identification method; The fluctuation range calculation module of the time-varying uncertain sliding bearing oil film distribution load is used to calculate the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load based on the autocovariance matrix and using the concentrated load identification method; The sliding bearing oil film distribution load identification module is used to calculate the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load according to the time-varying uncertain sliding bearing oil film distribution load median function and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and calculate the sliding bearing oil film distribution load.

[0014] Compared with the prior art, the present invention has the following technical effects: (1) The method of the present invention introduces a non-probabilistic interval process model to describe the uncertainty of the axis trajectory and the oil film distribution load of the sliding bearing in the form of a bounded set; at the same time, based on an orthogonal series expansion method of an interval process, the time-varying uncertain sliding bearing oil film distribution load is expressed as the sum of the median function of the sliding bearing oil film distribution load and the load uncertainty part, and then the identification of the time-varying uncertain sliding bearing oil film distribution load is decomposed into two types of deterministic problems: the identification of the original load median function and the identification of the load autocovariance matrix. This method can effectively deal with the time-varying uncertain distribution load identification problem existing in actual engineering, broaden the scope of load identification technology, and enhance the engineering practicality and extensiveness of load identification technology; thereby avoiding the problem that traditional probability-based methods require a large amount of test sample data, which is extremely costly or cannot be obtained at all.

[0015] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which: Figure 1 A schematic flow chart of a method for identifying oil film distribution load of a sliding bearing under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Figure 2 A schematic diagram of the structure of a sliding bearing oil film distribution load identification system under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Figure 3 A schematic diagram of a main shaft model of a high-power wind turbine gearbox is shown according to an embodiment of a method for identifying a sliding bearing oil film distribution load under time-varying uncertain factors according to an embodiment of the present disclosure; Figure 4 A schematic diagram of the median function of the axis trajectory at five measuring points of an embodiment of a method for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Figure 5 A schematic diagram showing the identification result of the median function of the time history function of an embodiment of a method for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Figure 6 A schematic diagram of a Chebyshev orthogonal polynomial reconstruction of a spatial distribution function is shown according to an embodiment of a method for identifying a sliding bearing oil film distribution load under time-varying uncertain factors according to an embodiment of the present disclosure; Figure 7 A schematic diagram of a median function and an error of a sliding bearing oil film distribution load under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Figure 8 A schematic diagram of the covariance matrix of the axis trajectory of an embodiment of a method for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Fig. 9 A schematic diagram of a covariance matrix of a time history function of an embodiment of a method for identifying a sliding bearing oil film distribution load under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Fig.10 A schematic diagram of the identification result of the radius function of the time history function of an embodiment of a method for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Fig.11 A specific schematic diagram of the autocorrelation coefficient of an embodiment of a method for identifying a sliding bearing oil film distribution load under time-varying uncertain factors according to an embodiment of the present disclosure is shown; Fig.12 A schematic diagram of the interval dynamics of a time-varying uncertain sliding bearing oil film distribution load according to an embodiment of a method for identifying a sliding bearing oil film distribution load under time-varying uncertain factors according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0018] Wind turbine gearboxes have more than a thousand load conditions. During operation, they will not only bear frequent start-stop impact loads, as well as non-torsion loads such as yaw moment, wind rotor weight and unbalanced load, but also complex structural coupling deformation, which makes the sliding bearing work in a harsh environment (heavy load, high temperature, sudden load, etc.), and multiple failure modes (deformation failure, fracture failure, wear failure, etc.) coexist. There are multi-source uncertainties such as speed, lubricating oil parameters, bearing surface quality and load in the lubrication process. Uncertain information is repeatedly propagated between subsystems. Coupled with the time-varying effects of material properties and boundary conditions, the axis trajectory data sequence contains a large amount of variation information of constantly changing uncertainties. There is no definite change rule for each measurement result. As a result, the identified sliding bearing oil film distribution load produces irregular fluctuations. At the same time, the axis trajectory test of the sliding bearing of a high-power wind turbine gearbox is expensive, and the serious lack of axis trajectory test data leads to low accuracy in testing the axis trajectory. The existing interval or probability distribution dynamic load identification methods are prone to accuracy problems such as overly conservative results or large deviations when based on low-precision axis trajectories, and often fail to meet the actual needs of sliding bearing stability fault prediction.

[0019] The interval process model uses intervals rather than precise probability distribution to describe the variation range of parameters at any time. The uncertainty of dynamic parameters is described by two boundary curves over the entire time history, which can greatly reduce the dependence on the sample size of parameters, thereby providing an effective mathematical tool for time-varying uncertainty modeling and corresponding structural uncertainty analysis in the absence of experimental information. The method of the present invention introduces a non-probabilistic interval process model for the time-varying uncertain parameters in the identification of the oil film distribution load of the sliding bearing, and describes the uncertainty of the axis trajectory and the oil film distribution load of the sliding bearing in the form of a bounded set. The uncertainty of the oil film distribution load of the sliding bearing is described using the median function and the radius function. Through an orthogonal series expansion method of an interval process, the time-varying uncertainty oil film distribution load of the sliding bearing is expressed as the sum of the median function of the oil film distribution load of the sliding bearing and the uncertainty part of the load, and then the identification of the time-varying uncertainty oil film distribution load of the sliding bearing is decomposed into two types of deterministic problems: the identification of the median function of the original load and the identification of the autocovariance matrix of the load. For the identification of the median function of the oil film distribution load of the sliding bearing, the median function of the axis trajectory is used as input. This is actually a conventional distributed dynamic load identification problem, which can be solved by conventional distributed dynamic load identification methods. For the identification of the uncertainty of the oil film distributed load of the sliding bearing, based on an orthogonal series expansion method of an interval process, through the mapping relationship between the measured axis trajectory covariance matrix and the equivalent sliding bearing oil film concentrated dynamic load covariance matrix, it can be transformed into a problem of identifying the covariance matrix of the equivalent concentrated uncertainty dynamic load based on the response covariance matrix. The square root of the diagonal elements of the load covariance matrix is ​​the interval radius value of the load at different times. The upper and lower boundary functions of the oil film distributed load of the sliding bearing can be obtained by linearly superimposing the median function of the oil film distributed load of the sliding bearing and the interval radius value at different times. The present invention proposes a method for identifying the oil film distributed load of the sliding bearing under time-varying uncertain factors. The upper and lower boundary functions of the oil film distributed load of the sliding bearing can be identified by sampling a small amount of axis trajectory, and the dynamic fluctuation of the oil film distributed load of the sliding bearing can be well monitored, and the operating status of the wind turbine can be accurately evaluated. Thereby providing reasonable load conditions for the stability design and optimization of the sliding bearing.

[0020] This patent develops a method for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors. In view of the time-varying uncertain parameters in the identification of the oil film distribution load of the sliding bearing, a non-probabilistic interval process model is introduced to describe the uncertainty of the axis trajectory and the oil film distribution load of the sliding bearing in the form of a bounded set. The uncertainty of the oil film distribution load of the sliding bearing is described using the median function and the radius function. Through an orthogonal series expansion method of an interval process, the time-varying uncertainty oil film distribution load of the sliding bearing is expressed as the sum of the median function of the oil film distribution load of the sliding bearing and the uncertainty part of the load, and then the identification of the oil film distribution load of the sliding bearing with time-varying uncertainty is decomposed into two types of deterministic problems: the identification of the median function of the original load and the identification of the autocovariance matrix of the load. It can greatly reduce the dependence on the sample size of the axis trajectory, thereby providing an effective mathematical tool for time-varying uncertainty modeling and corresponding structural uncertainty analysis with lack of experimental information. The method of the present invention broadens the scope of distributed dynamic load identification technology. It enhances the engineering practicality and extensiveness of load identification technology.

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Reference Figure 1 As shown, this embodiment provides a method for identifying the oil film distribution load of a sliding bearing under time-varying uncertain factors, comprising the following steps: S101. Extract the operating parameters of the wind turbine gearbox transmission shaft system through finite element modeling and analysis.

[0023] In this embodiment, the wind turbine gearbox transmission shaft system operating parameters include a mass matrix M, a stiffness matrix K, a damping matrix C, and a Green's kernel function matrix G.

[0024] S102, collecting the axis trajectory of the wind turbine gearbox transmission shaft system, and calculating the median function and autocovariance matrix of the axis trajectory based on the operating parameters.

[0025] S103. Based on the median function of the axis trajectory and using a distributed load identification method, a median function of the oil film distributed load of the time-varying uncertain sliding bearing is calculated.

[0026] In this embodiment, the median function of the axis trajectory Based on the conventional distributed load identification method, the modal load is obtained, and the first-order mode with the largest contribution is used as the time history function of the time-varying uncertain sliding bearing oil film distributed load, that is, the median function of the time history function is obtained. ; The spatial distribution function of the time-varying uncertain oil film distribution load of the sliding bearing is reconstructed by fitting a set of independent basis functions using Chebyshev orthogonal polynomials. By multiplying the identified time history function and spatial distribution function, we can obtain the median function of the time-varying uncertain sliding bearing oil film distribution load: .

[0027] Specifically, in this embodiment, the identification of the oil film distribution load of the sliding bearing based on a determined axis trajectory can be performed based on a conventional distributed load identification method. By converting the oil film distribution load of the sliding bearing into a time-space independent distributed load process, it can be expressed as .in, Represents the real time history function of the oil film distribution load, The real space distribution function representing the oil film distribution load, It is a function of the position variable z and the time variable t.

[0028] Conventional distributed load identification methods mainly use the Order effective modal load To use the equivalent time history function, and to use Chebyshev orthogonal polynomials to linearly fit the spatial distribution function, we can (1) Where: Equivalent time history function The real time history function A ratio Relationship, c k is a ratio, and the fitting spatial distribution function and real space distribution function Also a ratio relation, c k The reciprocal of , , and denote the i-th generalized Chebyshev orthogonal polynomial and coefficient respectively, and m denotes the total number of terms of the generalized Chebyshev orthogonal polynomial.

[0029] From formula (1), we can know that the oil film distribution load time history function and spatial distribution function The contribution to the axis trajectory is different. Time history function It mainly affects the form of the axis trajectory, while the spatial distribution function Mainly affects the amplitude of the axis trajectory.

[0030] The oil film distributed load of the sliding bearing can be regarded as a single-source concentrated dynamic load acting on the wind turbine gearbox transmission shaft system in a certain spatial distribution form. At each point in the action area, the oil film distributed load of the sliding bearing has the same time history form, and they reach the maximum or minimum value at the same time.

[0031] Therefore, the time-varying uncertainty factors affect the dynamic fluctuation of the time history function of the oil film distribution load of the sliding bearing. That is, the fluctuation range of the time history function of the oil film distribution load caused by the time-varying uncertainty factors, and its upper and lower boundaries can be expressed as . are the median function and radius function of the time history function respectively. Therefore, the time-varying uncertain sliding bearing oil film distribution load can be transformed into: (2) in, is the lower boundary of the oil film distribution load fluctuation range, It is the upper boundary of the oil film distribution load fluctuation range.

[0032] That is, based on the conventional distributed load identification method, the median value of the oil film distributed load of the sliding bearing is identified through a deterministic axis trajectory. .

[0033] S104. Based on the autocovariance matrix and using a concentrated load identification method, a fluctuation range of the time-varying uncertain sliding bearing oil film distribution load is calculated.

[0034] Specifically, in this embodiment, for the linear system of the wind turbine gearbox transmission shaft system, the uncertainty part of the axis trajectory is Essentially the uncertain part of the oil film distribution load The uncertain part of the oil film distribution load under time-varying uncertain factors It mainly comes from the fluctuation of the oil film distribution load time history function. When the input oil film distribution load time history function is about the standard interval variable In the form of an infinite series, theoretically, the uncertainty part of the axis trajectory is It should also be an infinite series about standard interval variables. According to the dynamic convolution relation, The axis trajectory caused by this can be expressed as: (3) The autocovariance function of the axis trajectory is: (4) in, represents the autocovariance function of the interval time history function, is the autocovariance function of the axis trajectory, t is a certain time, τ is a certain time interval, and the g() function is the Green function from the time history function action point to the axis trajectory response measurement point under unit impact load. , is the eigenvalue and corresponding characteristic function of the time history correlation coefficient function, For another time, For another time interval.

[0035] Then, the discretization of equation (4) is performed to obtain the autocovariance matrix model of the time history function of the oil film distribution load as follows: (5) in, is the Green kernel function matrix of the wind turbine gearbox transmission shaft system. The Green kernel function matrix G can be obtained by measurement or finite element calculation when the load position and the measuring point position are known. T is the transposed matrix. The covariance matrix of the axis trajectory can be obtained by correlation ellipsoid modeling of the measured axis trajectory samples.

[0036] According to Mercer's theorem, the autocovariance matrix of the axis trajectory is spectrally decomposed: , where the equivalent eigenvector of the axis trajectory response is , and Represents the autocovariance matrix of the axis trajectory Correspondingly, the spectral decomposition of the autocovariance matrix of the interval time history function is: Then formula (5) can be transformed into , then (6) in, is the equivalent eigenvector of the axis trajectory response, is the load equivalent eigenvector.

[0037] By observing formula (6), it is not difficult to find that it is similar to the positive problem model of concentrated dynamic load identification ( ) are consistent, where u is the displacement response vector and F is the external excitation load vector. The identification of can be handled based on the concentrated dynamic load identification method. Substitution This method realizes the accurate inverse calculation of the covariance matrix of the equivalent interval time history function. , we can get the radius function of the time history function The fluctuation range of the time-varying uncertain sliding bearing oil film distribution load can be obtained.

[0038] S105, calculating the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load according to the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and calculating the sliding bearing oil film distribution load.

[0039] In view of the lack of sample information of the axis trajectory, the construction of the interval process model only requires that the upper and lower boundaries of the fluctuation of the time-varying uncertainty parameters and the time correlation information are known. Compared with statistical information such as probability cumulative distribution function or probability density function, the upper and lower boundaries of the axis trajectory of parameter fluctuations are often easy to obtain in actual engineering and can be based on less sample data or engineering experience. Under the representation of the interval process model, in the non-random vibration analysis of the wind turbine gearbox transmission shaft system structure, the oil film distribution load of the sliding bearing is an uncertain quantity that varies with time and is described by an interval process, that is, , and are the lower and upper boundary functions of the time-varying uncertain sliding bearing oil film distribution load, respectively. The autocorrelation coefficient function is , is the oil film distribution load set. The interval orthogonal series expansion method is used to represent the time-varying uncertain sliding bearing oil film distribution load, and we have (7) in and They are respectively the time-varying uncertain sliding bearing oil film distribution load The median function and radius function of ; is a standard interval variable, satisfying ; and The autocorrelation coefficient function The eigenvalues ​​and eigenfunctions of .

[0040] The time-varying uncertain sliding bearing oil film distribution load described by equation (7) Substituting into the wind turbine gearbox transmission shaft system dynamics differential equation, we can get: (8) in, , and denote the mass, damping and stiffness matrices of the structure respectively, , , are the acceleration, velocity and displacement responses of the structure. Time-varying uncertain sliding bearing oil film distribution load is a position variable and time variables function.

[0041] At this time, in the dynamic differential equation, only is an interval variable, and , , and are all deterministic functions or values. , represents the uncertainty part of the time-varying uncertain sliding bearing oil film distribution load. Then equation (8) can be expressed as: (9) Therefore, the axis trajectory can be regarded as a median function of the time-varying uncertain sliding bearing oil film distribution load and the uncertainty part or Considering that the wind turbine gearbox transmission shaft system is linear and time-invariant, its axis trajectory can be regarded as the median function The median value of the axis trajectory caused and The uncertainty of the axis trajectory caused by The sum of Therefore, formula (9) can be expressed separately as: (10) According to formula (10), the load identification of oil film distribution in time-varying uncertain sliding bearings can be decomposed into two types of deterministic problems: The median load Identification of the random uncertain part of the axis trajectory The uncertain part of the oil film distribution load Identification of statistical features.

[0042] like Figure 2 As shown, this embodiment also provides a sliding bearing oil film distribution load identification system under time-varying uncertain factors, including: a wind power gearbox transmission shaft system operation parameter acquisition module 1, an axis trajectory analysis module 2 and a time-varying uncertain sliding bearing oil film distribution load identification module 3; The wind turbine gearbox transmission shaft system operating parameter acquisition module 1 is used to extract the wind turbine gearbox transmission shaft system operating parameters through finite element modeling and analysis; The axis trajectory analysis module 2 is used to collect the axis trajectory of the wind turbine gearbox transmission shaft system, and calculate the median function and autocovariance matrix of the axis trajectory based on the operating parameters; The time-varying uncertain sliding bearing oil film distribution load identification module 3 is used to calculate the time-varying uncertain sliding bearing oil film distribution load according to the median function of the axis trajectory and the autocovariance matrix; The time-varying uncertain sliding bearing oil film distribution load identification module 3 includes a time-varying uncertain sliding bearing oil film distribution load median function calculation module 31, a time-varying uncertain sliding bearing oil film distribution load fluctuation interval calculation module 32 and a sliding bearing oil film distribution load identification module 33; The time-varying uncertain sliding bearing oil film distribution load median function calculation module 31 is used to calculate the time-varying uncertain sliding bearing oil film distribution load median function based on the median function of the axis trajectory and using the distributed load identification method; The fluctuation range calculation module 32 of the time-varying uncertain sliding bearing oil film distribution load is used to calculate the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load based on the autocovariance matrix and using the concentrated load identification method; The sliding bearing oil film distribution load identification module 33 is used to calculate the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load according to the time-varying uncertain sliding bearing oil film distribution load median function and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and calculate the sliding bearing oil film distribution load.

[0043] Embodiment: This embodiment constructs a high-power wind turbine gear main shaft model and uses the technical solution of the present invention to determine the time-varying uncertain sliding bearing oil film distribution load to verify the validity of the patent of the present invention. The specific process is as follows: like Figure 3 The main shaft model of a high-power wind turbine gearbox is shown in Figure 1. The material properties of the main shaft of a high-power wind turbine gearbox are the same, and its elastic modulus is GPa, Poisson's ratio is , the density is kg / m 3 The oil film pressure in the interval independent of time and space acts vertically on points 1-5 of the wind turbine gearbox axis along the y direction, and its spatial distribution function is , , the median function of the uncertainty time history is , , the radius is 200, and the autocorrelation coefficient function is: , set the sampling time interval to 0.005s.

[0044] By sampling the time history function of the oil film distribution load of the sliding bearing 200 times, 200 samples of the oil film distribution load of the sliding bearing are obtained. These sample loads are applied to the main shaft structure one by one, and the corresponding axis trajectory response is calculated and mixed with 5% level of measurement noise as the measurement response of the axis trajectory. Then the median function of the axis trajectory under the action of these 200 sample loads is calculated, and the ellipsoid correlation model of the axis trajectory random variables at each time is established to obtain the covariance matrix of the axis trajectory.

[0045] like Figure 4As shown, this embodiment identifies the median function of the oil film distribution load of the time-varying uncertain sliding bearing according to the median function of the axis trajectory at the selected five measuring points according to the conventional dynamic distributed load identification method, completes the identification of the modal load after the modal transformation, and identifies the median function of the oil film distribution load of the time-varying uncertain sliding bearing, wherein the first order with the largest contribution is selected as the median function of the time-varying uncertain sliding bearing oil film distribution load time history function (such as Figure 5 ), and reconstruct the spatial distribution function through Chebyshev orthogonal polynomials (as Figure 6 ), Figure 7 The median function and error of the time-varying uncertain sliding bearing oil film distribution load are shown, and the error is only 0.62%, which shows the effectiveness of the identification of the median function of the time-varying uncertain sliding bearing oil film distribution load.

[0046] Next, the covariance matrix of the equivalent time history function, radius function and autocorrelation coefficient function are identified based on the covariance matrix of the axis trajectory. Since the original time history function is equivalent to the first-order modal load, the covariance matrix of the response at two measuring points is selected as the input for the identification of the time-varying uncertain time history function. Figure 8 shows the covariance matrix of the axis trajectory at these two measuring points.

[0047] Through spectral decomposition, the covariance matrix of the axis trajectory can be expressed as the superposition of the product of its equivalent eigenvector and the vector transpose. The eigenvalues ​​and eigenvectors of the autocovariance function of the axis trajectory are extracted and calculated. , for each axis trajectory response equivalent eigenvector , there must be a corresponding load equivalent eigenvector satisfy According to equation (6), Based on the concentrated dynamic load identification method, ,pass Compute the autocovariance function of a time history function , thus obtaining the radius function of the time history function The covariance matrix of the time history function is as follows: Fig. 9 As shown in Figure 10. Due to the influence of the ill-posedness in the inverse process, the covariance matrix of the identified time history function has a certain error. Taking the square root of the diagonal elements of the covariance matrix of the identified time history function, the radius function of the time history function is obtained, as shown in Figure 10. It shows a certain fluctuation, but its general trend is similar. The autocorrelation coefficient function of the equivalent identified time history function is consistent with the autocorrelation coefficient function of the time history function, which shows that the recognition result is effective. Fig.11 FIG12 shows the interval dynamics of the time-varying uncertain sliding bearing oil film distribution load.

[0048] It can be seen that the time-varying uncertain sliding bearing oil film distribution load identification method based on the interval process model and interval orthogonal series expansion of the present invention decomposes the time-varying uncertain sliding bearing oil film distribution load identification into two types of deterministic problems: identification of the original load median function and identification of the load autocovariance matrix. It can effectively deal with the time-varying uncertain distribution load identification problem existing in actual engineering. It broadens the scope of load identification technology and enhances the engineering practicality and extensiveness of load identification technology; at the same time, the non-probabilistic interval model describes the time-varying uncertainty of the axis trajectory and the sliding bearing oil film distribution load in the form of a bounded set, avoiding the problem that the traditional probability-based method requires a large amount of test sample data and has extremely high data costs. It reduces the particularity and difficulty of the time-varying uncertainty of the oil film distribution load.

[0049] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0050] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0051] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors, characterized in that: The following steps are involved: Extract the operating parameters of the wind turbine gearbox transmission shaft system through finite element modeling and analysis; Collecting the axis trajectory of the wind turbine gearbox transmission shaft system, and calculating the median function and autocovariance matrix of the axis trajectory based on the operating parameters; Based on the median function of the axis trajectory and using a distributed load identification method, a median function of the oil film distributed load of the time-varying uncertain sliding bearing is calculated; Based on the autocovariance matrix and using a concentrated load identification method, a time-varying uncertain sliding bearing oil film distribution load fluctuation range is calculated; The upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are calculated according to the time-varying uncertain sliding bearing oil film distribution load median function and the time-varying uncertain sliding bearing oil film distribution load fluctuation range, and the sliding bearing oil film distribution load is calculated.

2. The method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors according to claim 1 is characterized in that: The wind turbine gearbox transmission shaft system operating parameters include a mass matrix M, a stiffness matrix K, a damping matrix C and a Green's kernel function matrix G.

3. The method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors according to claim 1 is characterized in that: The process of calculating the median function of the oil film distributed load of the time-varying uncertain sliding bearing based on the median function of the axis trajectory and using the distributed load identification method is as follows: The spatial distribution function of the time-varying uncertain oil film distribution load of the sliding bearing is reconstructed by fitting a set of independent basis functions using Chebyshev orthogonal polynomials. Based on the conventional distributed load identification method, the modal load is obtained, and the first-order mode with the largest contribution is used as the time history function of the time-varying uncertain sliding bearing oil film distributed load, and the median function of the time history function is obtained; The median function of the identified time history function and the spatial distribution function are multiplied to obtain the median function of the time-varying uncertain sliding bearing oil film distribution load.

4. The method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors according to claim 3 is characterized in that: The process of calculating the time-varying uncertain sliding bearing oil film distribution load fluctuation range based on the autocovariance matrix and using the concentrated load identification method is as follows: Based on Mercer's theorem, the autocovariance matrix of the axis trajectory is spectrally decomposed to extract the eigenvalues ​​and eigenvectors of the autocovariance matrix of the axis trajectory; An equivalent eigenvector of the axis trajectory response is calculated based on the eigenvalue and the eigenvector, and an equivalent eigenvector of the load is identified based on a concentrated dynamic load identification method; Calculating an autocovariance matrix of a time history function based on the load equivalent eigenvector, and calculating a radius function of the time history function based on the autocovariance matrix of the time history function; The upper and lower boundaries of the time history function of the oil film distribution load are obtained based on the median function of the time history function and the radius function of the time history function, and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load is calculated.

5. The method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors according to claim 1 is characterized in that: The upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are calculated based on the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and the process of calculating the time-varying uncertain sliding bearing oil film distribution load is as follows: The time-varying uncertain sliding bearing oil film distribution load is analyzed, and a functional relationship between the time-varying uncertain sliding bearing oil film distribution load and the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying sliding bearing oil film distribution load is constructed; Based on the functional relationship, the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are calculated; Based on the sliding bearing oil film distribution load characteristics, the time-varying uncertain sliding bearing oil film distribution load is obtained according to the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load.

6. The method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors according to claim 5 is characterized in that: The process of carrying out an analysis on the time-varying uncertain sliding bearing oil film distribution load and constructing a functional relationship between the sliding bearing oil film distribution load and the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load is as follows: The interval orthogonal series expansion method is used to expand the time-varying uncertain oil film distribution load of sliding bearings. The expanded time-varying uncertain sliding bearing oil film distribution load is substituted into the wind turbine gear transmission shaft system dynamics differential equation, and the functional relationship between the sliding bearing oil film distribution load and the median function of the time-varying uncertain sliding bearing oil film distribution load and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load is calculated.

7. The method for identifying oil film distribution load of sliding bearing under time-varying uncertain factors according to claim 5 is characterized in that: The upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load are: ； in, is the lower boundary of the oil film distribution load fluctuation range, is the upper boundary of the fluctuation range of the oil film distribution load, f(z) is the spatial distribution function of the time-varying uncertain sliding bearing oil film distribution load, is the median function of the oil film distribution load time history function, is the radius function of the oil film distribution load time history function, is the median function of the time-varying uncertain sliding bearing oil film distribution load, is the load fluctuation range of the time-varying uncertain sliding bearing oil film distribution.

8. A system for identifying oil film distribution load of a sliding bearing under time-varying uncertain factors, implemented by using the method for identifying oil film distribution load of a sliding bearing under time-varying uncertain factors as claimed in any one of claims 1 to 7, characterized in that: include: A wind turbine gearbox transmission shaft system operating parameter acquisition module (1), an axis trajectory analysis module (2), and a time-varying uncertain sliding bearing oil film distribution load identification module (3); The wind turbine gearbox transmission shaft system operating parameter acquisition module (1) is used to extract the wind turbine gearbox transmission shaft system operating parameters through finite element modeling and analysis; The axis trajectory analysis module (2) is used to collect the axis trajectory of the wind turbine gearbox transmission shaft system, and calculate the median function and autocovariance matrix of the axis trajectory based on the operating parameters; The time-varying uncertain sliding bearing oil film distribution load identification module (3) is used to calculate the time-varying uncertain sliding bearing oil film distribution load according to the median function of the axis trajectory and the autocovariance matrix.

9. The system for identifying oil film distribution load of sliding bearings under time-varying uncertainties according to claim 8 is characterized in that: in, The time-varying uncertain sliding bearing oil film distribution load identification module (3) comprises a time-varying uncertain sliding bearing oil film distribution load median function calculation module (31), a time-varying uncertain sliding bearing oil film distribution load fluctuation range calculation module (32) and a sliding bearing oil film distribution load identification module (33); The time-varying uncertain sliding bearing oil film distributed load median function calculation module (31) is used to calculate the time-varying uncertain sliding bearing oil film distributed load median function based on the median function of the axis trajectory and using a distributed load identification method; The time-varying uncertain sliding bearing oil film distribution load fluctuation range calculation module (32) is used to calculate the time-varying uncertain sliding bearing oil film distribution load fluctuation range based on the autocovariance matrix and using a concentrated load identification method; The sliding bearing oil film distribution load identification module (33) is used to calculate the upper and lower boundaries of the time-varying uncertain sliding bearing oil film distribution load according to the time-varying uncertain sliding bearing oil film distribution load median function and the fluctuation range of the time-varying uncertain sliding bearing oil film distribution load, and calculate the sliding bearing oil film distribution load.

Images