Bearing simulation correction system and method based on actually measured stress of hollow roller

Abstract

The invention relates to the technical field of bearings, in particular to a bearing simulation correction system and method based on actual measurement stress of a hollow roller. The control module is used for comparing the simulation stress value, calculated by the bearing digital simulation module, at the corresponding position of the hollow roller with the actually-measured stress value, output by the signal processing and stress calculation module, of the roller, performing parameter inversion and correction, and reversely adjusting key uncertain parameters in a simulation model based on a comparison result, so that a simulation result continuously approaches to actually-measured data; the simulation correction and feedback module is used for receiving and updating the corrected parameters to obtain a high-confidence simulation model verified by actually measured stress, and the high-confidence simulation model is used for deep stress analysis, life prediction and / or health state monitoring of the bearing under other working conditions. Measurement is direct, and the source is accurate; closed-loop correction is achieved, and the model is credible.

Description

Technical Field

[0001] This invention relates to the field of bearing technology, specifically to a bearing simulation correction system and method based on the measured stress of hollow rollers. Background Technology

[0002] The stress state of rolling bearings is a core basis for their life prediction, reliability analysis, and fault diagnosis. Currently, the main methods for obtaining bearing stress distribution rely on computer simulation (finite element analysis) and theoretical formulas (Hertz contact theory). However, these methods have inherent limitations, leading to deviations between their calculation results and actual operating conditions.

[0003] The existing technology mainly has the following problems:

[0004] The simulation model's input parameters are inaccurate; the simulation results are highly dependent on input parameters such as boundary conditions, material properties, and contact conditions. These parameters differ under actual operating conditions, leading to significant errors even with seemingly minor ones.

[0005] Model simplification introduces errors: In order to improve computational efficiency, simulation models often simplify certain complex features of bearings, and these simplifications introduce errors that cannot be ignored.

[0006] Lack of effective verification methods: Traditional verification methods mostly rely on indirect measurements of the bearing's external structure, which cannot obtain the true stress state of the bearing's internal structure and newly stressed components. Therefore, it is difficult to directly and effectively verify and correct the accuracy of the simulation model. Summary of the Invention

[0007] In view of the shortcomings of the prior art, the present invention provides a bearing simulation correction system and method based on the measured stress of hollow rollers. It can directly obtain the real stress data of rolling elements and use this as a basis to accurately feed back and correct the digital simulation model, thereby improving the confidence of bearing stress calculation.

[0008] To achieve the above objectives, the present invention provides a bearing simulation correction system based on the measured stress of hollow rollers. This system includes a sensing roller module, a testing machine, a signal processing and stress calculation module, a bearing digital simulation module, and a simulation correction and feedback module. The sensing roller module includes at least one hollow roller, a strain gauge disposed on the inner wall of the hollow roller, and an electrical signal output device connected to the strain gauge. The testing machine includes a load loading device that applies a load to the hollow roller along a direction perpendicular to the generatrix of the hollow roller. The signal processing and stress calculation module receives the strain signal from the strain gauge, amplifies, filters, and performs module conversion, and calculates the load based on a pre-established strain signal and stress value. The calibration relationship model is used to calculate the measured stress value of the roller. The bearing digital simulation module is used to establish a parametric finite element model of the bearing that includes the geometry and material properties of the hollow roller, and to perform stress simulation calculations to obtain the simulated stress value. The simulation correction and feedback module is used to compare the simulated stress value at the corresponding position of the hollow roller calculated by the bearing digital simulation module with the measured stress value of the roller output by the signal processing and stress calculation module, perform parameter inversion and correction, and adjust the key uncertain parameters in the simulation model in reverse based on the comparison results, so that the simulation results continuously approach the measured data. The simulation correction and feedback module is used to receive and update the corrected parameters to obtain a high-confidence simulation model verified by measured stress.

[0009] Furthermore, the strain gauges are arranged in three groups along the axial direction of the hollow roller on the inner wall of the hollow roller, and each group of strain gauges includes four strain gauges evenly distributed along the circumferential direction.

[0010] Furthermore, the first group is located near the large end of the hollow roller, the second group is located near the small end of the hollow roller, and the third group is located in the middle of the hollow roller.

[0011] Furthermore, the key uncertain parameters include actual load, distribution, material properties, and / or constraints.

[0012] The bearing simulation correction method based on the measured stress of hollow rollers includes the following specific steps:

[0013] Step S100, System Construction Step: Prepare a hollow roller and attach strain gauges to the inner wall of the hollow roller. Connect the strain gauges to an electrical signal extraction device and assemble them into a sensing roller module. Connect the electrical signal extraction device to the testing machine.

[0014] Step S200, load calibration step: Install the hollow roller on the testing machine and apply a known load F to the hollow roller. calibration Simultaneously, the surface strain value ε output by the strain gauge is recorded. measured Repeatedly apply and record multiple load points, plot calibration curves, and analyze the surface strain value ε.measured With load F calibration The relationship between them:

[0015] F calibration =k×ε measured +C,

[0016] Fit k and C, where k is the strain load coefficient and C is the constant deviation;

[0017] According to the physical conversion formula between strain and stress values: σ=E×ε,

[0018] Where σ is the stress value, and the unit is MPa;

[0019] E is the elastic modulus of the roller material, expressed in GPa.

[0020] ε is the strain value, which is dimensionless; therefore, a model for the calibration relationship between strain signal and stress value is established.

[0021] The measured data acquisition steps involve installing the hollow roller in the bearing to be tested, running it under the target working condition, collecting the strain signal of the strain gauge in real time, processing the strain signal, and calculating the measured stress value of the roller based on the strain signal and stress value calibration relationship model.

[0022] Step S300, Simulation and Comparison Step: Initialize a bearing parametric finite element model in the bearing digital simulation module, input the theoretical parameters of the current working condition, run the simulation to obtain the simulated stress value of the roller; compare the simulated stress value with the measured stress value of the roller under the same working condition to obtain the comparison error;

[0023] Step S400, iterative correction step: based on the comparison error, the uncertain parameters in the simulation model are inverted and corrected through an optimization algorithm;

[0024] The revised simulation model was used as the high-confidence simulation model.

[0025] Furthermore, the comparison error includes;

[0026] For each load i and strain gauge position j:

[0027] Mean absolute error:

[0028] Root mean square error:

[0029] Correlation coefficient:

[0030] The error threshold is defined as: MAE j =5%; RMSE j =8%;

[0031] Where σ sim The simulated stress value; σ mean This represents the measured stress value of the roller.

[0032] If all indicators meet the threshold, the system is deemed qualified; if any indicator exceeds the tolerance, the parameter correction process is triggered.

[0033] Furthermore, the load calibration of the sensing roller module on the testing machine includes using a load loading device to apply a load to the hollow roller along a direction perpendicular to the generatrix of the hollow roller.

[0034] Further, the iterative correction step in step S400, based on the comparison error, involves optimizing the algorithm to invert and correct the uncertain parameters in the simulation model, including:

[0035] Step S410: Identify the key uncertain parameters that have the greatest impact on the simulation stress results, determine their priority order for adjustment, and form a set of highly sensitive parameters;

[0036] Step S420: Based on the comparison error, construct an optimization problem and automatically find a set of optimal parameters P* through an algorithm to minimize the error index between simulation and actual measurement.

[0037] Step S430: Verify the effectiveness and generalization ability of the optimal parameters P* obtained by inversion;

[0038] Step S440: Automated iteration until the model reaches a stable and reliable state.

[0039] Furthermore, the verification of the effectiveness and generalization ability of the optimal parameters P* obtained by the inversion in step S430 includes:

[0040] Step S431: Use P* to run and verify the operating conditions in the bearing digital simulation module;

[0041] Step S432: Recalculate the error indices, including MAE, RMSE, and R. 2 ;

[0042] Step S433: Determine whether all error indicators have reached the preset threshold;

[0043] If achieved, proceed to the next step;

[0044] If the desired result is not achieved, you need to return to step S420 to adjust and optimize the settings or return to step S410 to re-examine the parameter set.

[0045] The beneficial effects of this invention are as follows: This invention establishes a closed-loop feedback mechanism of "measurement-simulation-comparison-correction". By systematically comparing the measured stress data with the simulation output, the key uncertain parameters in the model are identified, and dynamic adjustments are made based on the inversion algorithm. This process can be iterated multiple times, so that the simulation results continuously approach the real physical behavior, thereby significantly improving the predictive consistency and adaptability of the model under complex working conditions. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the strain gauge distribution on the hollow roller of a bearing simulation correction system based on measured stress of hollow rollers in one embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram showing the distribution of the same set of strain gauges on the hollow rollers in a bearing simulation correction system based on measured stress of hollow rollers, according to an embodiment of the present invention.

[0048] Figure 3 This is a schematic diagram and dimensions of a hollow roller in a bearing simulation correction system based on measured stress of a hollow roller, according to an embodiment of the present invention. The units are mm.

[0049] Figure 4 This is a schematic diagram of roller load loading in one embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of roller stress distribution in one embodiment of the present invention;

[0051] Figure 6 This is a flowchart illustrating a bearing simulation correction method based on measured stress of hollow rollers in one embodiment of the present invention.

[0052] In the picture:

[0053] 100. Sensing roller module; 110. Hollow roller; 120. Strain gauge.

[0054] 200. Signal Processing and Stress Calculation Module

[0055] 300. Bearing digital simulation module.

[0056] 400. Simulation correction and feedback module.

[0057] 500. Loading device; 510. Simulated outer raceway; 520. Simulated inner raceway. Detailed Implementation

[0058] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] A bearing simulation correction system based on the measured stress of hollow rollers includes a sensing roller module 100, a testing machine, a signal processing and stress calculation module 200, a bearing digital simulation module 300, and a simulation correction and feedback module 400. The sensing roller module 100 includes at least one hollow roller 110, a strain gauge 120 disposed on the inner wall of the hollow roller 110, and an electrical signal output device connected to the strain gauge 120. The testing machine includes a load loading device 500, which applies a load to the hollow roller 110 along a direction perpendicular to its generatrix. The signal processing and stress calculation module 200 receives the strain signal from the strain gauge 120, amplifies, filters, and performs module conversion, and calculates the measured stress value of the roller based on a pre-established strain signal and stress value calibration relationship model. The bearing digital simulation module 300 is used to establish a parametric finite element model of the bearing that includes the geometric and material properties of the hollow roller, for stress simulation calculation, to obtain the simulated stress value. The simulation correction and feedback module 400 is used to compare the simulated stress value at the corresponding position of the hollow roller calculated by the bearing digital simulation module with the measured stress value of the roller output by the signal processing and stress calculation module, perform parameter inversion and correction, and adjust the key uncertain parameters in the simulation model in reverse based on the comparison results, so that the simulation results continuously approach the measured data. The simulation correction and feedback module 400 is used to receive and update the corrected parameters to obtain a high-confidence simulation model verified by measured stress. The high-confidence simulation model is used for in-depth stress analysis, life prediction and / or health status monitoring of bearings under other operating conditions.

[0060] Furthermore, three sets of strain gauges are arranged along the axial direction of the hollow roller 110 on the inner wall of the hollow roller 110, and each set of strain gauges includes four strain gauges 120 evenly distributed along the circumferential direction.

[0061] Furthermore, the first group is located near the large end of the hollow roller 110, the second group is located near the small end of the hollow roller 110, and the third group is located in the middle of the hollow roller 110.

[0062] Furthermore, key uncertainties include actual loads, distributions, material properties, and / or constraints.

[0063] The bearing simulation correction method based on the measured stress of hollow rollers includes the following specific steps:

[0064] Step S100, see Figures 1-3 The system construction steps are as follows: a hollow roller 110 is prepared and a strain gauge 120 is attached to the inner wall of the hollow roller 110. The strain gauge 120 is connected to an electrical signal extraction device and assembled into a sensing roller module 100. The electrical signal extraction device is connected to the testing machine.

[0065] Step S200, load calibration steps, see [link / reference] Figure 4 Hollow roller 110 is installed between simulated outer raceway 510 and simulated inner raceway 520 of the testing machine, and a known load F is applied to hollow roller 110 in the laboratory using load loading device 500. calibration Simultaneously, the surface strain value ε output by strain gauge 120 is recorded. measured Repeatedly apply and record multiple load points, plot calibration curves, and analyze the surface strain value ε. measured With load F calibration The relationship between them:

[0066] F calibration =k×ε measured +C,

[0067] Fit k and C, where k is the strain load coefficient and C is a constant deviation, such as preload or temperature compensation term.

[0068] According to the physical conversion formula between strain and stress values: σ=E×ε,

[0069] Where σ is the stress value (MPa); E is the elastic modulus of the roller material (Gpa); and ε is the strain value (dimensionless). Furthermore, a "load-strain-stress" relationship model is established, which includes a model for the calibration relationship between strain signals and stress values.

[0070] See Figure 4 The measured data acquisition steps involve installing the hollow roller 110 in the bearing to be tested and running it under the target working condition. When the hollow roller 110 is loaded, the inner wall of the hollow roller 110 deforms, causing the output of the strain gauge 120 bridge circuit to change. The strain signal of the strain gauge 120 is collected in real time. Then, based on the current strain signal and the "load-strain-stress" relationship model, the load borne by the hollow roller 110 is calculated, and the strain signal is processed. The measured stress value of the roller is calculated according to the strain signal and stress value calibration relationship model.

[0071] Step S300, Simulation and Comparison Step: Initialize a bearing parametric finite element model in the bearing digital simulation module 300, input the theoretical parameters of the current working condition, run the simulation to obtain the simulated stress value of the hollow roller; compare the simulated stress value with the measured stress value of the roller under the same working condition to obtain the comparison error, the comparison error includes;

[0072] For each load i and strain gauge position j:

[0073] Mean absolute error:

[0074] Root mean square error:

[0075] Correlation coefficient:

[0076] The error threshold is defined as: MAE j =5%; RMSE j =8%;

[0077] Where σ sim The simulated stress value; σ mean This represents the measured stress value of the roller.

[0078] If all indicators meet the threshold, the result is considered "qualified"; if any indicator exceeds the tolerance, the parameter correction process is triggered.

[0079] Step S400, iterative correction steps are as follows Figure 5 The specific process is as follows: based on the comparison error, parameter sensitivity is sorted, inversion optimization is performed, verification and update are performed, and then iterative convergence is achieved.

[0080] Detailed steps include:

[0081] S410, Parameter Sensitivity Analysis: Identify the key uncertain parameters that have the greatest impact on the simulation stress results and determine their priority order for adjustment in order to narrow the search space of the optimization problem and improve the inversion efficiency;

[0082] The initial bearing parameterized finite element model is used to determine a set of uncertain parameters P = [p1, p2, ..., pn] (material elastic modulus E, Poisson's ratio ν, friction coefficient μ between raceway and roller, load distribution coefficient α, boundary condition stiffness K, etc.) and a set of reference working conditions (from measured data points of S200 / S300).

[0083] Benchmark simulation: Run the simulation using the nominal value P0 of the parameter to obtain the benchmark simulation stress field σ_sim(P0);

[0084] Perturbation analysis: For each parameter pk, select a small perturbation Δpk (e.g., ±1% or ±5%) near its nominal value; Forward perturbation simulation: σ_sim(pk+Δpk, P0_{-k}),

[0085] Negative perturbation simulation: σ_sim(pk-Δpk,P0_{-k}), where P0_{-k} indicates that other parameters remain at their nominal values.

[0086] Sensitivity calculation: For each strain gauge position j, the sensitivity coefficient S_{j,k} of parameter pk to the stress at that location can be calculated using the finite difference method.

[0087] S_{j,k}≈[σ_sim_j(pk+Δpk)-σ_sim_j(pk-Δpk)] / (2*Δpk)

[0088] To obtain a global sensitivity index, the sensitivity of all key measurement points j and all load conditions i is combined (e.g., by taking the root mean square):

[0089]

[0090] Where M is the number of load conditions and N is the number of strain gauge positions.

[0091] Output: A list of parameters sorted from largest to smallest by G_S_k, forming a "high-sensitivity parameter set". These parameters will be adjusted first in subsequent inversion.

[0092] S420, Parameter Inversion and Optimization: Based on the calculated comparison error, an optimization problem is constructed, and an algorithm is used to automatically find a set of optimal parameters P* to minimize the error index (such as RMSE) between simulation and actual measurement.

[0093] Optimization problems include:

[0094] Decision variables: The key uncertain parameter vector X to be inverted (taken from the highly sensitive parameter set selected in step 1); Objective function F(X): The comprehensive error of the stress difference between simulation and measured values ​​at all measurement points and under all working conditions. Using weighted root mean square error:

[0095]

[0096] Among them, w ij It is the weighting coefficient.

[0097] First, a fast surrogate model of F(X) is constructed using a limited number of simulation sample points. Then, intensive optimization iterations are performed on the model. Finally, the optimal solution is verified using real simulations.

[0098] process:

[0099] (1) Set the optimization algorithm and stopping criteria (such as the maximum number of iterations, the threshold for changes in the objective function).

[0100] (2) The algorithm generates a new parameter combination X_new.

[0101] (3) Call the bearing digital simulation module, input X_new, run the simulation, and obtain σ_sim(X_new).

[0102] (4) Calculate the objective function value F(X_new).

[0103] (5) Repeat steps 2-4 until the stopping criterion is met, and output the optimal parameter set P*.

[0104] S430. Verification and Update: Verify the effectiveness and generalization ability of the optimal parameters P* obtained by inversion.

[0105] S431. Use P* to run and verify the operating conditions in the bearing digital simulation module.

[0106] S432. Recalculate error indices (MAE, RMSE, R) 2 ).

[0107] S433. Determine whether all indicators have reached the preset threshold (e.g., MAE < 5%, RMSE < 8%). 2 >0.9). If this is achieved, proceed to the next step; if not, return to S420 to adjust and optimize the settings (such as expanding the search range or increasing sample points), or return to S410 to re-examine the parameter set.

[0108] S440, Iteration and Convergence Judgment. This automates the entire correction process iteratively until the model reaches a stable and reliable state.

[0109] Iteration logic:

[0110] After completing one round of iterative calculations in steps S410-S430, the updated high-confidence model is used to re-simulate and predict the new, previously unused measured operating conditions. The prediction error under the new operating conditions is then calculated. Convergence is determined when the operating condition error used for inversion meets the threshold, and the prediction error under the new operating conditions remains within an acceptable range. This indicates that the model has good generalization ability and iterative convergence. If convergence is not achieved, and the error is still large under the new operating conditions, it suggests that the current inversion parameter set may be overfitted, or that key uncertain parameters have not been included. In this case, the new operating condition data needs to be added to the inversion dataset, returning to S410 to consider incorporating new potential uncertain parameters, and restarting a new round of the "sensitivity analysis-inversion optimization" cycle.

[0111] Final output: A highly robust and high-confidence digital simulation model of a bearing, validated through multiple iterations and under various operating conditions. This model can be used for accurate in-depth stress analysis, life prediction, and health status monitoring.

[0112] Furthermore, the load calibration of the sensing roller module 100 on the testing machine includes applying a load to the hollow roller 110 along the generatrix direction perpendicular to the hollow roller 110 using a load loading device 500.

[0113] The bearing simulation correction system and method based on measured stress of hollow rollers proposed in this invention have the following advantages:

[0114] Direct measurement ensures accurate and reliable data. Traditional bearing simulations often rely on theoretical assumptions or indirect calculations, leading to significant uncertainties in the model input. This invention achieves accurate measurement of the roller stress state by directly deploying high-precision strain gauges 120 inside the hollow roller 110. This ensures the authenticity and accuracy of stress information from the data source, avoiding simulation deviations caused by distorted parameter assumptions.

[0115] Closed-loop feedback enables adaptive correction of the simulation model. This invention establishes a closed-loop feedback mechanism of "measurement-simulation-comparison-correction," which systematically compares measured stress data with simulation output. The measured stress data represents the measured stress value of the roller, while the simulation output represents the simulated stress value. Key uncertain parameters in the model (such as material constitutive properties, contact characteristics, boundary conditions, etc.) are identified and dynamically adjusted based on an inversion algorithm. This process can be iterated multiple times, allowing the simulation results to continuously approximate the actual physical behavior, thereby significantly improving the model's predictive consistency and adaptability under complex working conditions.

[0116] The model boasts high reliability and multi-scenario transfer capabilities. It can also be robustly extended to other untested operating conditions (such as variable loads, variable speeds, and extreme temperatures), providing a trustworthy digital twin foundation for in-depth stress analysis, fatigue life prediction, health status monitoring, and remaining life assessment of bearings. This significantly reduces the cost of full-condition physical testing while enhancing the scientific rigor of design optimization and operational decisions.

[0117] The system and methodology are highly practical and support intelligent operation and maintenance as well as design optimization. They can be embedded into the entire lifecycle of bearing R&D, testing, and service. They can be used for simulation calibration and performance verification of new product designs, as well as for condition inversion and fault early warning of in-service bearings. This enables a shift from "experience-driven" to "data-model fusion-driven" approaches, providing key technical support for the reliability and intelligent development of high-end equipment.

[0118] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0119] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0120] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0121] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0122] It should be noted that when an element is referred to as being "fixed to" or "set on" another element, it can be directly on the other element or there may be an intervening element. When an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

Claims

1. A bearing simulation correction system based on measured stress of hollow rollers, characterized in that: include The sensing roller module includes at least one hollow roller, a strain gauge disposed on the inner wall of the hollow roller, and an electrical signal output device connected to the strain gauge. The testing machine includes a load loading device that applies a load to the hollow roller along a direction perpendicular to the generatrix of the hollow roller. The signal processing and stress calculation module is used to receive the strain signal of the strain gauge, amplify, filter and convert it, and calculate the measured stress value of the roller according to the pre-established strain signal and stress value calibration relationship model. The bearing digital simulation module is used to establish a parametric finite element model of the bearing that includes the geometry and material properties of the hollow roller, and to perform stress simulation calculations to obtain the simulated stress values. The simulation correction and feedback module is used to compare the simulated stress value at the corresponding position of the hollow roller calculated by the bearing digital simulation module with the measured stress value of the roller output by the signal processing and stress calculation module, perform parameter inversion and correction, and adjust the key uncertain parameters in the simulation model in reverse based on the comparison results, so that the simulation results continuously approach the measured data; the simulation correction and feedback module is used to receive and update the corrected parameters to obtain a high-confidence simulation model verified by measured stress.

2. The bearing simulation correction system based on measured stress of hollow rollers according to claim 1, characterized in that: The strain gauges are arranged in three groups along the axial direction of the hollow roller on the inner wall of the hollow roller, and each group of strain gauges includes four strain gauges evenly distributed along the circumferential direction.

3. The bearing simulation correction system based on measured stress of hollow rollers according to claim 2, characterized in that: The first group is located near the large end of the hollow roller, the second group is located near the small end of the hollow roller, and the third group is located in the middle of the hollow roller.

4. The bearing simulation correction system based on measured stress of hollow rollers according to claim 1, characterized in that: The key uncertain parameters include actual load, distribution, material properties, and / or constraints.

5. A bearing simulation correction method based on measured stress of hollow rollers, characterized in that: The specific steps include Step S100, System Construction Step: Prepare a hollow roller and attach strain gauges to the inner wall of the hollow roller. Connect the strain gauges to an electrical signal extraction device and assemble them into a sensing roller module. Connect the electrical signal extraction device to the testing machine. Step S200, load calibration step: Install the hollow roller on the testing machine and apply a known load to the hollow roller. Simultaneously record the surface strain values ​​output by the strain gauge. Repeatedly apply and record multiple load points, plot calibration curves, and analyze the surface strain values. With load The relationship between them: ， Fit k and C, where k is the strain load coefficient and C is the constant deviation; According to the physical conversion formula between strain and stress values: , in, This is the stress value, in MPa. E is the elastic modulus of the roller material, expressed in GPa. The strain value is dimensionless; therefore, a model for the calibration relationship between the strain signal and the stress value is established. The measured data acquisition steps involve installing the hollow roller in the bearing to be tested, running it under the target working condition, collecting the strain signal of the strain gauge in real time, processing the strain signal, and calculating the measured stress value of the roller based on the strain signal and stress value calibration relationship model. Step S300, Simulation and Comparison Step: Initialize a bearing parametric finite element model in the bearing digital simulation module, input the theoretical parameters of the current working condition, run the simulation to obtain the simulated stress value of the roller; compare the simulated stress value with the measured stress value of the roller under the same working condition to obtain the comparison error; Step S400, iterative correction step: based on the comparison error, the uncertain parameters in the simulation model are inverted and corrected through an optimization algorithm; The revised simulation model was used as the high-confidence simulation model.

6. The bearing simulation correction method based on measured stress of hollow rollers according to claim 5, characterized in that: The comparison error includes: For each load i and strain gauge position j: Mean absolute error: ; Root mean square error: ; Correlation coefficient: ; The error threshold is defined as: =5%; =8%; =0.9; in These are simulated stress values; This represents the measured stress value of the roller. If all indicators meet the threshold, the system is deemed qualified; if any indicator exceeds the tolerance, the parameter correction process is triggered.

7. The bearing simulation correction method based on measured stress of hollow rollers according to claim 1, characterized in that: The load calibration of the sensing roller module on the testing machine includes using a load loading device to apply a load to the hollow roller along the generatrix direction perpendicular to the hollow roller.

8. The bearing simulation correction method based on measured stress of hollow rollers according to claim 1, characterized in that: The iterative correction step in step S400, based on the comparison error, involves optimizing and correcting the uncertain parameters in the simulation model using an optimization algorithm, including: Step S410: Identify the key uncertain parameters that have the greatest impact on the simulation stress results, determine their priority order for adjustment, and form a set of highly sensitive parameters; Step S420: Based on the comparison error, construct an optimization problem and automatically find a set of optimal parameters P* through an algorithm to minimize the error index between simulation and actual measurement. Step S430: Verify the effectiveness and generalization ability of the optimal parameters P* obtained by inversion; Step S440: Automated iteration until the model reaches a stable and reliable state.

9. The bearing simulation correction method based on measured stress of hollow rollers according to claim 8, characterized in that: The verification of the effectiveness and generalization ability of the optimal parameters P* obtained by the inversion in step S430 includes: Step S431: Use P* to run and verify the operating conditions in the bearing digital simulation module; Step S432: Recalculate the error indices, including MAE, RMSE, and R². Step S433: Determine whether all error indicators have reached the preset threshold; If achieved, proceed to the next step; If the desired result is not achieved, you need to return to step S420 to adjust and optimize the settings or return to step S410 to re-examine the parameter set.

Images