A method for building a digital twin model of a bearing and optimizing model parameters

Abstract

The present invention provides a method for constructing a digital twin model of a bearing and optimizing model parameters, and relates to the field of digital twin technology. The present invention realizes accurate monitoring and evaluation of the operating status of the bearing by introducing a lubricant degradation evaluation index and a stress abnormality evaluation index based on historical data. The method comprehensively considers the bearing geometric parameters, lubricant status and bearing stress conditions, and dynamically adjusts the parameters of the digital twin model, including stiffness, damping coefficient and wear coefficient, to optimize the lubricant replacement cycle, thereby overcoming the shortcomings of the prior art in terms of model accuracy, adaptability and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins, and in particular to a method for constructing a digital twin model of a bearing and optimizing model parameters. Background Art

[0002] In the bearing system, the introduction of digital twin technology makes it possible to build a virtual model based on real data. By real-time monitoring of the working state parameters of the bearing, such as temperature, vibration, stress and wear, the actual operating state of the bearing can be dynamically reflected and its future behavior can be predicted. Traditional digital twin models mainly rely on static physical models, which make it difficult to effectively consider the dynamic changes of multi-dimensional input parameters (such as stress state and lubricating oil state) of the bearing under the actual operating environment. This limitation means that there is still a lot of room for improvement in the accuracy and real-time performance of the existing digital twin models.

[0003] In the prior art, the publication number is CN117709190A, and the name is a method for constructing a digital twin model of a rolling bearing and optimizing model parameters, which includes the following steps: S1: In combination with the radial force, the outer ring, the inner ring, and the dynamic relationship between the rolling elements, a four-degree-of-freedom digital twin model of the bearing is established, and the geometric parameters and operating parameters used to measure the bearing are input; S2: Based on the locust algorithm, the key parameters that cannot be measured inside the bearing are initialized and optimized, and the four-degree-of-freedom digital twin model of the bearing outputs a virtual twin signal; S3: The root mean square difference between the virtual twin signal and the actual vibration signal is calculated, and the locust algorithm searches for key parameters in the direction of reducing the difference; S4: After the locust algorithm iteration ends, the optimal twin signal is the one with the smallest error between the output record and the measured vibration signal, and the bearing under the corresponding parameters is the optimal digital twin, which provides rich data for training anomaly detection, fault classification and remaining service life prediction models, and monitors the health status of the bearing in real time.

[0004] The shortcomings of the bearing digital twin model in terms of model parameter optimization are mainly reflected in the incomplete consideration of key factors such as lubricant degradation and bearing stress abnormality. Lubricant is a key medium in bearing operation, and its state change directly affects the operating performance and wear state of the bearing. However, the existing methods lack effective quantitative means to evaluate the degree of lubricant degradation, and cannot dynamically reflect the impact of lubricant degradation on bearing wear and performance. The existing bearing digital twin model construction method has limited flexibility and reliability when facing actual complex working conditions, and lacks methods for dynamically optimizing model parameters and lubricant replacement cycles, which affects its ability to manage the entire life cycle of the bearing system.

[0005] The above information disclosed in the above Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a method for constructing a bearing digital twin model and optimizing model parameters to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for constructing a bearing digital twin model and optimizing model parameters, the specific steps comprising:

[0009] Step S1: constructing a digital twin model based on the geometric parameters of the bearing and historical operating status data including multi-dimensional input parameters, wherein the multi-dimensional input parameters include force state data acting on the rolling element of the bearing and a first state parameter of the lubricating oil, and the lubricating oil affects the operating state of the rolling element by forming an oil film;

[0010] Step S2: obtaining a first state parameter of the lubricating oil in the historical operating state data, determining a degradation warning threshold of the first state parameter according to the historical operating state data, combining and analyzing the first state parameter and the corresponding degradation warning threshold, and generating a lubricating oil degradation evaluation index for evaluating the degree of lubricating oil degradation;

[0011] Step S3: Collect the first state parameter corresponding to the bearing in the current operating state, and input the first state parameter into step S2 to obtain the lubricant degradation evaluation index, combine the lubricant degradation evaluation index with the digital twin model signal, analyze the influence of lubricant degradation on bearing wear and operating state, and obtain the bearing wear analysis result;

[0012] Step S4: receiving the bearing wear analysis result obtained in step S3, and using the bearing wear analysis result to preliminarily optimize the model parameters of the digital twin model, where the model parameters include bearing stiffness, damping coefficient, and wear coefficient;

[0013] Step S5: collecting stress state data of the rolling element in the bearing under the current operating state, determining an abnormal warning threshold of the stress state data according to the historical operating state data, combining and analyzing the stress state data and its corresponding abnormal warning threshold, and generating a stress abnormality evaluation index for evaluating the degree of stress abnormality of the rolling element;

[0014] Step S6: Based on the digital twin model initially optimized in step S4, a comprehensive optimization model is constructed in combination with the lubricant degradation evaluation index and stress abnormality evaluation index of the bearing under the current operating state. The comprehensive optimization model is used to dynamically adjust the model parameters of the digital twin model and the lubricant replacement cycle.

[0015] Furthermore, the first state parameter is combined with its corresponding degradation warning threshold for analysis to generate a lubricant degradation evaluation index, which is used to evaluate the degree of degradation of the lubricant, and specifically includes:

[0016] The particle concentration, oil viscosity change, and oil conductivity are labeled MC, VC, and EC, respectively;

[0017] According to the historical operating status data, a critical point between normal use and before deterioration of the lubricating oil is set, and the critical point is used as a degradation warning threshold of the first status parameter;

[0018] Set the particle concentration degradation warning threshold to MC th , when MC exceeds MC th When , it indicates lubricating oil contamination;

[0019] Set the degradation warning threshold of oil viscosity change to VC th , when VC is lower than VC th When , it indicates that the viscosity of the lubricating oil is insufficient;

[0020] Set the oil conductivity degradation warning threshold to EC th , when EC exceeds EC th When , it indicates that the lubricating oil is deteriorating;

[0021] The lubricant degradation evaluation index is defined as DI, and the calculation formula is as follows:

[0022]

[0023] Among them, w MC 、w VC 、w EC They are the weight coefficients of particle concentration, oil viscosity change, and oil conductivity;

[0024] In the preprocessing stage of building the digital twin model in step S1, the standardized output values ​​of MC, VC and EC are adjusted to the range of (0, 1) by scaling and offsetting; and the value range of the degradation warning threshold is also limited to the range of (0, 1);

[0025] If DI>1.0, the system issues a lubricant deterioration warning, indicating that replacement or maintenance is required;

[0026] If 0.7<DI≤1.0, it means that the lubricating oil is in a warning state, and regular inspection suggestions are provided;

[0027] If DI≤0.7, it means the lubricating oil is in good condition and can continue to be used normally.

[0028] Furthermore, the lubricant degradation evaluation index is combined with the digital twin model signal to analyze the impact of lubricant degradation on bearing wear and operating status, and the bearing wear analysis results are obtained, including:

[0029] The bearing wear analysis results are as follows:

[0030] When DI>1.0, the lubricant is seriously deteriorated, the bearing wear is aggravated, and the bearing wear rate exceeds the normal standard value by 25%;

[0031] When 0.7<DI≤1.0, the lubricating oil state is at the boundary, and the bearing wear rate exceeds the normal standard value by 10%-25%;

[0032] When DI≤0.7, the lubricating oil is in good condition and the bearing wear rate is within the normal standard value;

[0033] In the digital twin model, the bearing stiffness, damping coefficient, and wear coefficient are marked as ZK, ZN, and MS respectively;

[0034] Preliminary optimization of model parameters, including:

[0035] When DI>1.0, the adjustment strategies for bearing stiffness, damping coefficient and wear coefficient are as follows:

[0036]

[0037] Among them, ZK′, ZN′, and MS′ represent the adjusted values ​​of bearing stiffness, damping coefficient, and wear coefficient, respectively;

[0038] for The bearing's deformation capacity decreases due to lubricant deterioration, and the bearing stiffness needs to be reduced;

[0039] for As the fluidity of the lubricating oil decreases, the attenuation performance decreases, and the damping coefficient needs to be reduced;

[0040] for Since lubricant degradation accelerates the wear process, the wear factor needs to be increased;

[0041] When 0.7<DI≤1.0, the adjustment strategies for bearing stiffness, damping coefficient and wear coefficient are as follows:

[0042]

[0043] Among them, ZK′, ZN′, and MS′ represent the adjusted values ​​of bearing stiffness, damping coefficient, and wear coefficient, respectively;

[0044] for Since the lubricating oil condition is in the warning zone, the bearing stiffness needs to be reduced;

[0045] for Since the vibration suppression ability of the system is weakened, the damping coefficient needs to be reduced;

[0046] for Since the bearing wear rate has an increasing trend, the wear coefficient needs to be increased;

[0047] When DI≤0.7, ZK′, ZN′, and MS′ remain unchanged; at this time, the lubricating oil is in good condition, the vibration suppression ability is normal, and the lubricating oil condition will not accelerate wear.

[0048] Furthermore, the stress state data is combined with the corresponding abnormal warning threshold for analysis to generate a stress abnormality evaluation index, which is used to evaluate the degree of stress abnormality of the rolling element, specifically including:

[0049] The radial force, axial force and torque after the data preprocessing in step S1 are marked as Fr, Fa and Tn respectively;

[0050] The abnormal warning threshold is set to represent the maximum allowable variation of each stress state parameter under the current working condition; the abnormal warning thresholds of radial force, axial force and torque are set to Fr th 、Fa th 、Tn th ;

[0051] The stress abnormality evaluation index is defined as FI, and the calculation formula is as follows:

[0052]

[0053] Among them, η1 is the adjustment coefficient, 0.12≤η1≤0.67;

[0054] The stress state is divided into the following three intervals:

[0055] When 0.75≤FI<1, the rolling element is subjected to abnormally severe force, which may easily lead to bearing damage and requires immediate shutdown and inspection;

[0056] When 0.53≤FI<0.75, the rolling element force tends to be abnormal, and it is necessary to pay close attention and adjust the operating conditions to prevent further deterioration;

[0057] When FI is less than 0.53, the stress state is normal and no further measures are required.

[0058] Furthermore, a comprehensive optimization model is constructed by combining the lubricant degradation evaluation index and the stress abnormality evaluation index. The comprehensive optimization model is used to dynamically adjust the model parameters of the digital twin model and the lubricant replacement cycle, specifically including:

[0059] Take the lubricating oil deterioration evaluation index DI and the abnormal force evaluation index FI as the inputs of the comprehensive optimization model:

[0060] Dynamically adjust ZK′, ZN′, and MS′, and take the adjusted bearing stiffness, damping coefficient, and wear coefficient as the output parameters of the comprehensive optimization model. At the same time, take the recommended lubricating oil replacement period as the output result of the comprehensive optimization model;

[0061] In the comprehensive optimization model, the basis for adjusting the model parameters is the comprehensive result of DI and FI; the following specifically explains the optimization rules for the bearing stiffness, damping coefficient, and wear coefficient;

[0062] Only when DI > 1.0 and FI ≥ 0.75, the adjustment strategies for the bearing stiffness, damping coefficient, and wear coefficient are as follows:

[0063]

[0064] Only when 0.7 < DI ≤ 1.0 and 0.53 ≤ FI, the adjustment strategies for the bearing stiffness, damping coefficient, and wear coefficient are as follows:

[0065]

[0066] Among them, ZK″, ZN″, and MS″ respectively represent the values of ZK′, ZN′, and MS′ after dynamic adjustment; w DI and w FI are the weight coefficients of the corresponding indices,

[0067] Adjust the lubricating oil replacement period as follows:

[0068] When DI > 1.0 or FI ≥ 0.75, it is necessary to immediately replace the lubricating oil;

[0069] When 0.7 < DI ≤ 1.0 or 0.53 ≤ FI, regularly check the lubricating oil status and formulate a maintenance plan; and it is necessary to ensure that at least two of the particle concentration, oil viscosity change, and oil conductivity meet the requirements in the following formula;

[0070]

[0071] When DI ≤ 0.7 or FI < 0.53, there is no need to replace the lubricating oil, and continue to operate normally.

[0072] Compared with the prior art, the beneficial effects of the present invention are:

[0073] By introducing the lubricant degradation evaluation index and stress abnormality evaluation index based on historical data, accurate monitoring and evaluation of the bearing operating status can be achieved; this method comprehensively considers the bearing geometric parameters, lubricant status and bearing stress conditions, and dynamically adjusts the parameters of the digital twin model, including stiffness, damping coefficient and wear coefficient, to optimize the lubricant replacement cycle, thereby overcoming the shortcomings of existing technologies in model accuracy, adaptability and reliability; through this optimization method, the operating stability and life of the bearing system can be significantly improved, maintenance costs can be reduced, and intelligent full life cycle management of bearings can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION

[0075] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0076] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0077] Embodiment 1:

[0078] See also Figure 1 , the present invention provides a technical solution:

[0079] A method for constructing a digital twin model of a bearing and optimizing model parameters, the method is used to monitor and optimize the operating state of the bearing in real time through the state of lubricating oil, and the specific steps include:

[0080] Step S1: constructing a digital twin model based on the geometric parameters of the bearing and historical operating status data including multi-dimensional input parameters, wherein the multi-dimensional input parameters include force state data acting on the rolling element of the bearing and a first state parameter of the lubricating oil, and the lubricating oil affects the operating state of the rolling element by forming an oil film;

[0081] Step S2: obtaining a first state parameter of the lubricating oil in the historical operating state data, determining a degradation warning threshold of the first state parameter according to the historical operating state data, combining and analyzing the first state parameter and the corresponding degradation warning threshold, and generating a lubricating oil degradation evaluation index, wherein the lubricating oil degradation evaluation index is used to evaluate the degree of degradation of the lubricating oil;

[0082] Step S3: Collect the first state parameter corresponding to the bearing in the current operating state, and input the first state parameter into step S2 to obtain the lubricant degradation evaluation index, combine the lubricant degradation evaluation index with the digital twin model signal, analyze the influence of lubricant degradation on bearing wear and operating state, and obtain the bearing wear analysis result;

[0083] Step S4: receiving the bearing wear analysis result obtained in step S3, where the bearing wear analysis result is used to preliminarily optimize the model parameters of the digital twin model, where the model parameters include bearing stiffness, damping coefficient, and wear coefficient;

[0084] Step S5: collecting stress state data of the rolling element in the bearing under the current operating state, determining an abnormal warning threshold of the stress state data according to the historical operating state data, combining and analyzing the stress state data and the corresponding abnormal warning threshold, and generating a stress abnormality evaluation index, which is used to evaluate the degree of stress abnormality of the rolling element;

[0085] Step S6: Based on the digital twin model initially optimized in step S4, a comprehensive optimization model is constructed in combination with the lubricant degradation evaluation index and stress abnormality evaluation index of the bearing under the current operating state. The comprehensive optimization model is used to dynamically adjust the model parameters of the digital twin model and the lubricant replacement cycle.

[0086] To further explain, the construction of the digital twin model includes:

[0087] The geometric parameters of the bearing include inner diameter, outer diameter, width, rolling element type, number of rolling elements and their arrangement;

[0088] The multidimensional input parameters also include radial forces acting on the bearing, the rotational characteristics of the inner and outer rings;

[0089] Historical operating status data also includes temperature, speed, workload status and lubrication conditions;

[0090] For stress state data, sensors are used to monitor in real time the radial force, axial force and torque acting on the rolling element under different working conditions;

[0091] Description of radial force;

[0092] Decompose each radial force into two orthogonal components;

[0093] Vector addition:

[0094] Add the components of all radial forces in the horizontal and vertical directions respectively to obtain the total horizontal component and the total vertical component;

[0095] Calculate the resultant radial force using the vector synthesis formula;

[0096] The first state parameters of the lubricating oil include particle concentration, oil viscosity change and oil conductivity;

[0097] Create a 3D model of the bearing according to the design parameters of the bearing, and export the completed 3D model to STL and STEP formats;

[0098] Select MATLAB modeling software to build the basic framework of the digital twin model;

[0099] Use MATLAB's Simulink tool or write code directly in MATLAB to build a digital twin model;

[0100] Define the system equations, taking into account the mechanical behavior and lubrication characteristics of the bearing;

[0101] Integrate real-time monitoring data and set model inputs for subsequent simulation;

[0102] Integrate the collected geometric parameters, 3D model of the bearing and historical operating status data and input them into the basic framework of the digital twin model;

[0103] Set up multiple degrees of freedom of the model, including rotational characteristics, lubrication conditions, and workload conditions to simulate real operating environments;

[0104] Set simulation parameters: define simulation time, step size, and initial conditions to prepare for subsequent dynamic simulation;

[0105] Multi-dimensional input parameters: ensure that the model can receive data from different sensors and perform real-time or near real-time dynamic simulation;

[0106] Use the input data to perform preliminary dynamic simulation, observe the model's response, and record the operating status data;

[0107] Based on the preliminary simulation results, the degrees of freedom and input parameters in the model are gradually adjusted to optimize the model performance;

[0108] Compare the optimized model output with the actual collected operation data to verify the accuracy of the model;

[0109] The digital twin model performs preliminary dynamic simulation through multi-dimensional input parameters, and simulates the operating status of the bearing under different workloads and lubrication conditions by adjusting multiple degrees of freedom in the model.

[0110] Further explanation: the first state parameter is combined with its corresponding degradation warning threshold for analysis to generate a lubricant degradation evaluation index, which is used to evaluate the degree of degradation of the lubricant, and specifically includes:

[0111] Particle concentration: The concentration of wear particles in the lubricating oil is detected by a particle counting sensor;

[0112] Oil viscosity change: Use online viscosity sensors to monitor real-time changes in oil viscosity and obtain direct reflection parameters of lubricant aging;

[0113] Oil conductivity: Use conductivity sensors to monitor the conductivity of lubricating oil as a basis for judging the degree of contamination and deterioration of lubricating oil;

[0114] The particle concentration, oil viscosity change, and oil conductivity are labeled MC, VC, and EC, respectively;

[0115] According to the historical operating status data, the critical point between normal use and before deterioration of the lubricating oil is set as the degradation warning threshold of the first status parameter;

[0116] Set the particle concentration degradation warning threshold to MC th , when MC exceeds MC th When , it indicates lubricating oil contamination;

[0117] Set the degradation warning threshold of oil viscosity change to VC th , when VC is lower than VC th When , it indicates that the viscosity of the lubricating oil is insufficient;

[0118] Set the oil conductivity degradation warning threshold to EC th , when EC exceeds EC th When , it indicates that the lubricant is deteriorating;

[0119] The lubricant degradation evaluation index is defined as DI, and the calculation formula is as follows:

[0120]

[0121] Among them, w MC 、w VC 、w EC They are the weight coefficients of particle concentration, oil viscosity change, and oil conductivity; w MC 、w VC 、w EC The specific value is determined by the expert group through experimental data;

[0122] In the preprocessing stage of building the digital twin model in step S1, the standardized output values ​​of MC, VC and EC are adjusted to the range of (0, 1) by scaling and offsetting; and the value range of the degradation warning threshold is also limited to the range of (0, 1);

[0123] If DI>1.0, the system issues a lubricant deterioration warning, indicating that replacement or maintenance is required;

[0124] If 0.7<DI≤1.0, it means that the lubricating oil is in a warning state, and regular inspection suggestions are provided;

[0125] If DI≤0.7, it means the lubricating oil is in good condition and can continue to be used normally.

[0126] It is further explained that the lubricant degradation evaluation index is combined with the digital twin model signal to analyze the impact of lubricant degradation on bearing wear and operating status, and the analysis results are obtained, including:

[0127] Using the digital twin model, the DI value is coupled with the wear coefficient and operating status of the bearing for analysis. The specific impact of DI changes on bearing performance is evaluated. Based on historical operating status data, the normal standard value of the bearing wear rate is determined. The bearing wear analysis results are as follows:

[0128] When DI>1.0, the lubricant is seriously deteriorated, the bearing wear is aggravated, and the bearing wear rate exceeds the normal standard value by 25%;

[0129] When 0.7<DI≤1.0, the lubricating oil state is at the boundary, and the bearing wear rate exceeds the normal standard value by 10%-25%;

[0130] When DI≤0.7, the lubricating oil is in good condition and the bearing wear rate is within the normal standard value;

[0131] The lubricant degradation evaluation index DI obtained in step S3 and the combined bearing wear analysis results are transmitted to the optimization module of the digital twin model; the optimization module is responsible for preliminary optimization of the model parameters;

[0132] In the digital twin model, the bearing stiffness, damping coefficient, and wear coefficient are marked as ZK, ZN, and MS respectively;

[0133] Bearing stiffness, in N / m, reflects the deformation of the bearing under load;

[0134] Damping coefficient, in Ns / m, reflects the vibration attenuation capacity of the system;

[0135] Wear coefficient, in mm / h, reflects the wear rate of the bearing;

[0136] The initial optimization of the model parameters is:

[0137] When DI>1.0, the adjustment strategies for bearing stiffness, damping coefficient and wear coefficient are as follows:

[0138]

[0139] Among them, ZK′, ZN′, and MS′ represent the adjusted values ​​of bearing stiffness, damping coefficient, and wear coefficient, respectively;

[0140] for As the bearing deformation capacity decreases due to lubricant deterioration, corresponding measures need to be taken to reduce the bearing stiffness; and the upper limit of the reduction of bearing stiffness is 10% of the current value. If the calculated value exceeds 10%, 10% is used as the reduction value; when the lubricant deteriorates, the oil film becomes thinner, resulting in an increase in the contact stress on the bearing surface, thereby reducing the elastic deformation capacity of the bearing; the reduction in stiffness can reduce the contact pressure on the bearing, thereby extending its service life;

[0141] Corresponding measures:

[0142] 1. Adjust the load distribution: Re-adjust the installation position and load distribution of the bearing to reduce the contact pressure per unit area and avoid excessive local stiffness;

[0143] 2. Use flexible support structure: introduce flexible support elements, such as using elastic gaskets or adding elastic elements, to reduce the overall stiffness of the bearing to disperse the load and slow down deformation;

[0144] 3. Optimize lubrication method: increase the supply of lubricant or use oil-gas lubrication system to ensure the supply of lubricant in high-load areas and reduce damage caused by decreased bearing stiffness;

[0145] for As the fluidity of the lubricating oil decreases, the attenuation performance decreases, and corresponding measures need to be taken to reduce the damping coefficient; and the upper limit of the reduction of the damping coefficient is 5% of the current value. If the calculated value exceeds 5%, 5% is used as the reduction value; when the fluidity of the lubricating oil decreases, the attenuation capacity of the oil film decreases, resulting in increased system vibration; therefore, it is necessary to increase the sensitivity of the vibration response by reducing the damping coefficient of the bearing, so as to better control the vibration of the system;

[0146] Corresponding measures:

[0147] 1. Replace lubricating oil: Check the lubricating oil status regularly. If the fluidity decreases, replace it with a lubricating oil with a higher viscosity index immediately to ensure the stability of the oil film;

[0148] 2. Install vibration suppression devices: Introduce vibration suppression devices into the system, such as installing viscous dampers or rubber shock absorbers, to reduce the overall vibration response of the system and compensate for the vibration impact caused by the reduction of the damping coefficient;

[0149] 3. Adjust the running speed: reduce the running speed of the bearing to reduce the influence of the separation ability of the oil film on the vibration attenuation of the system, and ensure that the lubricating oil still has good attenuation performance under low speed;

[0150] for Since lubricant degradation accelerates the wear process, corresponding measures need to be taken to increase the wear coefficient; and the upper limit of the increase in the wear coefficient is 20% of the current value. If the calculated value exceeds 20%, 20% is used as the increase value; lubricant degradation aggravates the friction of the bearing, causing the wear rate between the friction pairs to increase; therefore, it is necessary to reflect the aggravation of wear by increasing the wear coefficient, and at the same time take measures to reduce the actual wear rate;

[0151] Corresponding measures:

[0152] 1. Improve lubricant formula: Use lubricating oil containing anti-wear additives to form an anti-wear protective layer on the friction surface and reduce wear caused by friction;

[0153] 2. Optimize the lubricant replacement cycle: According to the lubricant deterioration evaluation index, replace the lubricant in time before it reaches the critical deterioration state to avoid accelerated wear caused by deteriorated oil;

[0154] 3. Surface strengthening treatment: Surface treatment of the working surface of the bearing, such as carbonitriding, laser quenching or coating technology, can enhance the wear resistance of the bearing material and fundamentally reduce the wear rate of the friction pair;

[0155] When 0.7<DI≤1.0, the adjustment strategies for bearing stiffness, damping coefficient and wear coefficient are as follows:

[0156]

[0157] Among them, ZK′, ZN′, and MS′ represent the adjusted values ​​of bearing stiffness, damping coefficient, and wear coefficient, respectively;

[0158] for Since the lubricating oil condition is in the warning zone, corresponding measures are taken to reduce the bearing stiffness; and the upper limit of the reduction of the bearing stiffness is 5% of the current value. If the calculated value exceeds 5%, 5% is used as the reduction value; when the lubricating oil deterioration is in the warning zone, although it has not yet reached the replacement standard, it has already affected the bearing performance; at this time, by reducing the bearing stiffness, the negative impact of the lubricating oil performance decline on the bearing can be mitigated;

[0159] Corresponding measures:

[0160] 1. Optimize the bearing preload: adjust the bearing installation preload to make the preload moderate, avoid excessive rigidity leading to local stress concentration, and reduce the deformation burden of the bearing;

[0161] 2. Use elastic support elements: For example, when installing the bearing, introduce appropriate elastic materials (such as springs or rubber parts) as support to reduce the overall stiffness of the bearing system and reduce the impact of lubricant deterioration on bearing stiffness;

[0162] 3. Optimize load path design: By improving the mechanical design of the system, the load can be more evenly distributed on multiple bearings to avoid excessive load on a single bearing, thereby reducing stiffness;

[0163] for As the vibration suppression capability of the system is weakened, corresponding measures need to be taken to reduce the damping coefficient; and the upper limit of the reduction of the damping coefficient is 2% of the current value. If the calculated value exceeds 2%, the reduction value is 2%; the deterioration of the lubricating oil condition leads to a decrease in the vibration suppression capability of the system, and the bearing damping coefficient needs to be reduced to maintain the balance and vibration control of the system;

[0164] Corresponding measures:

[0165] 1. Optimize the viscosity of lubricating oil: select lubricating oil with appropriate viscosity to ensure that the lubricating oil can fully exert its damping effect during operation and reduce system vibration;

[0166] 2. Add external damping elements: Introduce additional mechanical damping elements into the system, such as friction dampers, hydraulic dampers, etc., to compensate for the adverse effects of reduced system damping coefficient;

[0167] 3. Adjust system operating conditions: reduce system operating speed or load to reduce the need for vibration suppression, thereby alleviating the risk of system vibration caused by a decrease in the damping coefficient;

[0168] for As the wear rate of the bearing tends to increase, corresponding measures need to be taken to increase the wear coefficient; and the upper limit of the increase in the wear coefficient is 10% of the current value. If the calculated value exceeds 10%, 10% is used as the increase value; if the wear rate of the bearing increases due to the deterioration of the lubricating oil or other factors, measures need to be taken to further increase the wear coefficient to reflect the actual wear condition and reduce the actual wear of the bearing;

[0169] Corresponding measures:

[0170] 1. Use high-performance lubricants: Choose lubricants with high anti-wear properties, especially those with added solid lubricants, such as molybdenum disulfide, which can form a solid lubricating film and effectively reduce the wear on the bearing surface;

[0171] 2. Regularly monitor bearing wear: Use the online monitoring system to monitor bearing wear in real time, use data analysis to discover wear trends, and take lubrication maintenance or bearing replacement in advance;

[0172] 3. Implement protective measures: Improve the working environment around the bearing, reduce the damage to the bearing caused by particles and dust in the environment, and reduce the increase in speed caused by wear;

[0173] When DI≤0.7, ZK′, ZN′, and MS′ remain unchanged; at this time, the lubricating oil is in good condition, the vibration suppression ability is normal, and the lubricating oil condition will not accelerate wear.

[0174] Further explanation: the stress state data and its corresponding abnormal warning threshold are combined and analyzed to generate a stress abnormality evaluation index, which is used to evaluate the degree of stress abnormality of the rolling element, specifically including:

[0175] A high-precision sensor installed near the rolling element of the bearing is used to collect the force status data of the rolling element in real time under the current operating state. The collected force status data includes the following three key parameters:

[0176] The radial force, axial force and torque after the data preprocessing in step S1 are marked as Fr, Fa and Tn respectively;

[0177] Radial force is the radial force acting on the rolling element, the unit is N;

[0178] The axial force is the axial force acting on the rolling element, and its unit is N;

[0179] Torque is the torque acting on the rolling element, measured in Nm;

[0180] These data are directly recorded by sensors, collected once per second, and the collected data are transmitted to the data processing unit in real time through the communication module;

[0181] Based on the historical operating status data, the normal range of each stress parameter is determined by statistical analysis of the long-term operating data of the bearing under different working conditions, and an abnormal warning threshold is generated; the abnormal warning threshold is set to represent the maximum allowable variation range of each stress state parameter under the current working condition; the abnormal warning thresholds of radial force, axial force and torque are set to Fr th 、Fa th 、Tn th , in order to assess the degree of abnormality of the current state;

[0182] The stress abnormality evaluation index is defined as FI, and the calculation formula is as follows:

[0183]

[0184] Among them, η1 is an adjustment coefficient, and 0.12 ≤ η1 ≤ 0.67;

[0185] The stress state is divided into the following three intervals:

[0186] When 0.75 ≤ FI < 1, the force on the rolling elements is extremely severe, which is likely to cause bearing damage and requires immediate shutdown for inspection; at this time, the force has exceeded the maximum allowable range that the system can withstand, and the system will automatically generate an alarm;

[0187] When 0.53 ≤ FI < 0.75, the force on the rolling elements approaches abnormality, and it is necessary to closely monitor and adjust the operating conditions to prevent further deterioration; although the stress state at this time has not exceeded the allowable range, it has approached the upper limit and needs to be monitored;

[0188] When FI < 0.53, the stress state is normal and no further measures are required. This interval indicates that the system is operating within the normal stress range.

[0189] Further explanation, combining the lubricating oil deterioration evaluation index and the stress abnormality evaluation index, a comprehensive optimization model is constructed. This comprehensive optimization model is used to dynamically adjust the model parameters of the digital twin model and the replacement cycle of the lubricating oil, specifically including:

[0190] Taking the lubricating oil deterioration evaluation index DI and the stress abnormality evaluation index FI as the inputs of the comprehensive optimization model:

[0191] Dynamically adjust ZK′, ZN′, and MS′, and take the adjusted bearing stiffness, damping coefficient, and wear coefficient as the output parameters of the comprehensive optimization model. At the same time, take the lubricating oil replacement cycle recommendation as the output result of the comprehensive optimization model;

[0192] In the comprehensive optimization model, the basis for adjusting the model parameters is the comprehensive result of DI and FI; the following specifically explains the optimization rules for the bearing stiffness, damping coefficient, and wear coefficient respectively;

[0193] Only when DI > 1.0 and FI ≥ 0.75, the adjustment strategies for the bearing stiffness, damping coefficient, and wear coefficient are as follows:

[0194]

[0195] Only when 0.7 < DI ≤ 1.0 and 0.53 ≤ FI, the adjustment strategies for the bearing stiffness, damping coefficient, and wear coefficient are as follows:

[0196]

[0197] Among them, ZK″, ZN″, and MS″ respectively represent the values of ZK′, ZN′, and MS′ after dynamic adjustment; w DI and w FIis the weight coefficient of the corresponding exponent, w DI and w FI are determined according to the experimental data by the expert group;

[0198] The replacement period of the lubricating oil is adjusted as follows:

[0199] When DI > 1.0 or FI ≥ 0.75, the lubricating oil needs to be replaced immediately;

[0200] When 0.7 < DI ≤ 1.0 or 0.53 ≤ FI, regularly check the status of the lubricating oil, and formulate a maintenance plan; and at least two parameters among the particle concentration, oil viscosity change, and oil conductivity need to meet the requirements in the following formula;

[0201]

[0202] When DI ≤ 0.7 or FI < 0.53, there is no need to replace the lubricating oil, and continue to operate normally.

[0203] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0204] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0205] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0206] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A method for constructing a digital twin model of a bearing and optimizing model parameters, the method is used to monitor and optimize the operating state of the bearing in real time through the state of lubricating oil, and is characterized in that: The specific steps include: Step S1: constructing a digital twin model based on the geometric parameters of the bearing and historical operating status data of multi-dimensional input parameters, where the multi-dimensional input parameters include force state data acting on the rolling elements of the bearing and a first state parameter of the lubricating oil; Step S2: obtaining a first state parameter of the lubricating oil in the historical operating state data, determining a degradation warning threshold of the first state parameter according to the historical operating state data, combining and analyzing the first state parameter and the corresponding degradation warning threshold, and generating a lubricating oil degradation evaluation index for evaluating the degree of lubricating oil degradation; The particle concentration, oil viscosity change, and oil conductivity are labeled MC, VC, and EC, respectively; According to the historical operating status data, a critical point between normal use and before deterioration of the lubricating oil is set, and the critical point is used as a degradation warning threshold of the first status parameter; Set the particle concentration degradation warning threshold to MC th , when MC exceeds MC th When , it indicates lubricating oil contamination; Set the degradation warning threshold of oil viscosity change to VC th , when VC is lower than VC th When , it indicates that the viscosity of the lubricating oil is insufficient; Set the oil conductivity degradation warning threshold to EC th , when EC exceeds EC th When , it indicates that the lubricant is deteriorating; The lubricant degradation evaluation index is defined as DI, and the calculation formula is as follows: Among them, w MC 、w VC 、w EC They are the weight coefficients of particle concentration, oil viscosity change, and oil conductivity; Step S3: Collect the first state parameter corresponding to the bearing in the current operating state, and input the first state parameter into step S2 to obtain the lubricant degradation evaluation index, combine the lubricant degradation evaluation index with the digital twin model signal, analyze the influence of lubricant degradation on bearing wear and operating state, and obtain the bearing wear analysis result; the bearing wear analysis result is as follows: When DI>1.0, the lubricant is seriously deteriorated, the bearing wear is aggravated, and the bearing wear rate exceeds the normal standard value by 25%; When 0.7<DI≤1.0, the lubricating oil state is at the boundary, and the bearing wear rate exceeds the normal standard value by 10%-25%; When DI≤0.7, the lubricating oil is in good condition and the bearing wear rate is within the normal standard value; Step S4: using the bearing wear analysis results to preliminarily optimize the model parameters of the digital twin model, the model parameters including bearing stiffness, damping coefficient and wear coefficient; In the digital twin model, the bearing stiffness, damping coefficient, and wear coefficient are marked as ZK, ZN, and MS respectively; Preliminary optimization of model parameters, including: When DI>1.0, the adjustment strategies for bearing stiffness, damping coefficient and wear coefficient are as follows: Among them, ZK′, ZN′, and MS′ represent the adjusted values ​​of bearing stiffness, damping coefficient, and wear coefficient, respectively; for The bearing's deformation capacity decreases due to lubricant deterioration, and the bearing stiffness needs to be reduced; for As the fluidity of the lubricating oil decreases, the attenuation performance decreases, and the damping coefficient needs to be reduced; for Since lubricant degradation accelerates the wear process, the wear factor needs to be increased; When 0.7<DI≤1.0, the adjustment strategies for bearing stiffness, damping coefficient and wear coefficient are as follows: Among them, ZK′, ZN′, and MS′ represent the adjusted values ​​of bearing stiffness, damping coefficient, and wear coefficient, respectively; for Since the lubricating oil condition is in the warning zone, the bearing stiffness needs to be reduced; for Since the vibration suppression ability of the system is weakened, the damping coefficient needs to be reduced; for Since the bearing wear rate has an increasing trend, the wear coefficient needs to be increased; When DI≤0.7, ZK′, ZN′, and MS′ remain unchanged; at this time, the lubricating oil is in good condition, the vibration suppression ability is normal, and the lubricating oil condition will not accelerate wear; Step S5: collecting stress state data of the rolling element in the bearing under the current operating state, determining an abnormal warning threshold of the stress state data according to the historical operating state data, combining and analyzing the stress state data and its corresponding abnormal warning threshold, and generating a stress abnormality evaluation index for evaluating the degree of stress abnormality of the rolling element; The radial force, axial force and torque after the data preprocessing in step S1 are marked as Fr, Fa and Tn respectively; Set the abnormal warning threshold to represent the maximum allowable change range of each force state parameter under the current working condition; set the abnormal warning thresholds of the radial force, axial force, and torque to Frth, Fath, and Tnth respectively; Define the force abnormal evaluation index as FI, and the calculation formula is as follows: Among them, η1 is an adjustment coefficient, 0.12 ≤ η1 ≤ 0.67; The force state is divided into the following three intervals: When 0.75 ≤ FI < 1, the force on the rolling elements is severely abnormal, which is likely to cause bearing damage and requires immediate shutdown for inspection; When 0.53 ≤ FI < 0.75, the force on the rolling elements approaches abnormality, and it is necessary to closely monitor and adjust the operating conditions to prevent further deterioration; When FI < 0.53, the force state is normal and no further measures are required; Step S6: Based on the preliminarily optimized digital twin model, combine the lubricating oil deterioration evaluation index and the force abnormal evaluation index of the bearing under the current operating state to construct a comprehensive optimization model, which is used to dynamically adjust the model parameters of the digital twin model and the replacement cycle of the lubricating oil; Dynamically adjust ZK′, ZN′, and MS′. In the comprehensive optimization model, the basis for adjusting the model parameters is the comprehensive result of DI and FI; the following specifically explains the optimization rules for the bearing stiffness, damping coefficient, and wear coefficient; Only when DI > 1.0 and FI ≥ 0.75, the adjustment strategies for the bearing stiffness, damping coefficient, and wear coefficient are as follows: Only when 0.7 < DI ≤ 1.0 and 0.53 ≤ FI, the adjustment strategies for the bearing stiffness, damping coefficient, and wear coefficient are as follows: Among them, ZK″, ZN″, MS″ respectively represent the values ​​of ZK′, ZN′, MS′ after dynamic adjustment; w DI and w FI is the weight coefficient of the corresponding index, The adjustment of the replacement cycle of the lubricating oil is as follows: When DI > 1.0 or FI ≥ 0.75, the lubricating oil needs to be replaced immediately; When 0.7 < DI ≤ 1.0 or 0.53 ≤ FI, regularly check the lubricating oil state and formulate a maintenance plan; and it is necessary to ensure that at least two of the particle concentration, oil viscosity change, and oil conductivity meet the requirements in the following formula; When DI ≤ 0.7 or FI < 0.53, there is no need to replace the lubricating oil and continue normal operation.

2. A method for constructing a bearing digital twin model and optimizing model parameters according to claim 1, characterized in that: The construction of the digital twin model includes: For the force state data, use sensors to continuously monitor the radial force, axial force, and torque acting on the rolling elements under different working conditions; The first state parameters of the lubricating oil include particle concentration, oil viscosity change, and oil conductivity; According to the design parameters of the bearing, create a three-dimensional model of the bearing and export the completed three-dimensional model in STL and STEP formats; Select MATLAB modeling software to establish the basic framework of the digital twin model; Integrate the collected geometric parameters, the three-dimensional model of the bearing, and the historical operating state data and input them into the basic framework of the digital twin model; The digital twin model conducts preliminary dynamic simulation through multi-dimensional input parameters and simulates the operating state of the bearing under different working loads and lubrication conditions by adjusting multiple degrees of freedom in the model.

3. A method for constructing a bearing digital twin model and optimizing model parameters according to claim 2, characterized in that: Combine the first state parameters with their corresponding deterioration warning thresholds for combined analysis to generate a lubricating oil deterioration evaluation index for evaluating the degree of lubricating oil deterioration, specifically including: In the preprocessing stage of building the digital twin model in step S1, the standardized output values ​​of MC, VC and EC are adjusted to the range of (0, 1) by scaling and offsetting; and the value range of the degradation warning threshold is also limited to the range of (0, 1); If DI>1.0, the system issues a lubricant deterioration warning, indicating that replacement or maintenance is required; If 0.7<DI≤1.0, it means that the lubricating oil is in a warning state, and regular inspection suggestions are provided; If DI≤0.7, it means the lubricating oil is in good condition and can continue to be used normally.

4. A method for constructing a bearing digital twin model and optimizing model parameters according to claim 3, characterized in that: Combining the lubricant degradation evaluation index and the stress abnormality evaluation index, a comprehensive optimization model is constructed, including: The lubricant degradation evaluation index DI and stress abnormality evaluation index FI are used as inputs of the comprehensive optimization model: The adjusted bearing stiffness, damping coefficient and wear coefficient are used as output parameters of the comprehensive optimization model, and the lubricant replacement cycle recommendation is used as the output result of the comprehensive optimization model.

Images