Method for measuring wear coefficient and predicting service life of carburizing material at different positions

Abstract

This invention belongs to the field of materials science and tribology, specifically a method for measuring the wear coefficient and predicting the life of carburized materials at different locations. The method includes the following steps: Testing is performed on the cross-section of a carburized sample from the surface to the core. Combining the microstructure and hardness data at corresponding depths, a quantitative mapping relationship is established between depth, microstructure, hardness, wear characteristic parameters, and wear coefficient, thus reversing the wear coefficient function distributed along the depth. An Archard wear finite element model is established, and the wear coefficient function is introduced as a field variable varying with wear depth into the model to establish the relationship between wear depth and cycle number. Random sampling simulations are performed, calculating the wear life under each sampling condition. The simulation results are statistically analyzed to generate the probability distribution and reliability function of the wear life, and the predicted life value is output. This invention establishes a continuous model from differential wear calculation to integral life assessment, resulting in stronger predictive capabilities.

Description

Technical Field

[0001] This invention belongs to the field of materials science and tribology, specifically a method for measuring the wear coefficient and predicting the life of carburized materials at different locations. Background Technology

[0002] Carburizing is a key surface engineering technique for improving the surface hardness, wear resistance, and fatigue strength of steel parts. After carburizing, a gradient distribution of carbon concentration, microstructure (such as martensite, retained austenite, and carbides), and hardness is formed from the surface to the core of the material. This gradient characteristic determines that the friction and wear behavior of the material varies significantly at different depths. For example, the high-hardness area on the surface may exhibit abrasive wear, while the subsurface area with a higher content of retained austenite may undergo induced phase transformation during friction, affecting the wear mechanism and wear rate.

[0003] Currently, the conventional method for evaluating the wear resistance of carburized materials is to conduct a standard "hardened layer wear-through test," which involves testing the specimen under specific load and friction conditions until it wears through, recording the wear-through time or measuring the weight loss during wear. While this method is intuitive, it has significant limitations:

[0004] 1) The testing cycle is long and the cost is high;

[0005] 2) It can only obtain the overall wear resistance of the material under specific conditions, and cannot reflect the gradient information of the wear coefficient with depth;

[0006] 3) It is difficult to directly correlate with the key parameter in theoretical wear models (such as the Archard wear equation)—the wear coefficient—thus limiting its ability to be used for accurate life prediction.

[0007] In terms of lifespan prediction, traditional methods mostly employ deterministic models based on a single average wear coefficient. However, the wear process is inherently a gradual process with stochastic characteristics. Factors such as the inhomogeneity of the material's microstructure and fluctuations in friction conditions cause the wear amount and wear life to exhibit dispersion. Existing studies have attempted to introduce stochastic process theory to model the statistical characteristics of wear amount, or to combine wear simulation with fatigue life prediction based on critical distance theory, but no method has yet been found that can systematically integrate the gradient wear characteristics of the carburized layer, rapid parameter acquisition, and stochastic lifespan prediction.

[0008] Therefore, developing a method that can quickly and accurately obtain the gradient wear coefficient of carburized materials and, based on this, achieve high-precision wear life prediction is of great engineering value and scientific significance for optimizing carburizing processes, guiding part design, and predicting maintenance cycles. Summary of the Invention

[0009] The purpose of this invention is to provide an efficient and accurate integrated method for measuring the gradient wear coefficient and predicting the life of carburized materials. The core of this invention lies in combining reciprocating sliding wear testing technology, multi-scale finite element wear simulation, and a reliability prediction model based on stochastic processes. Furthermore, a differential-integral life prediction model considering the continuous variation of the wear coefficient along depth is established for carburized steel such as 16MnCrS5.

[0010] The technical solution adopted by this invention to achieve the above objectives is: a method for measuring the wear coefficient and predicting the life of carburized materials at different locations, comprising the following steps:

[0011] Step S1: Using a reciprocating sliding wear testing machine, tests are performed on the cross-section of the carburized specimen from the surface to the core. Combining the hardness and microstructure data at the corresponding depths, a quantitative mapping relationship is established between depth, microstructure, hardness, wear characteristic parameters, and wear coefficient. The wear coefficient function distributed along the depth is then derived. ;

[0012] Step S2: Considering the gradient distribution of the wear coefficient, perform Archard wear finite element simulation and life modeling, establish the Archard wear finite element model, and apply the wear coefficient function... The Arcard wear finite element model is introduced as a field variable, and the differential Arcard formula is used to calculate local wear. The relationship between wear depth and cycle number is then established. ;

[0013] Step S3: Identify the key random variables affecting wear life and set a probability distribution for each random variable; use the Monte Carlo method to perform multiple random samplings, with each sampling based on the relationship between wear depth and cycle number established in Step S2. To obtain the corresponding wear life. For all Perform statistical analysis to generate the probability distribution and reliability function of wear life, and output the life prediction value.

[0014] Step S1 specifically includes:

[0015] Step S1-1: Prepare the metallographic section of the carburized sample and measure the Vickers hardness gradient from the surface to the core;

[0016] Step S1-2: Analyze the metallographic images of different locations in the carburized layer using a deep learning image recognition algorithm, and quantitatively calculate the volume fraction of martensite, ferrite, and retained austenite phases at each location.

[0017] Step S1-3: Pre-set the load and time, take the cross section of the carburized sample as the test surface, and use a spherical friction pair made of the same material to conduct a reciprocating sliding wear test at different positions along the depth direction from the carburized surface, and the sliding stroke at each position is fixed;

[0018] Steps S1-4: Characterize the wear marks using a super depth-of-field surface profilometer, white light interferometer, or contact profilometer. Select multiple points from the surface to the core and measure the wear mark depth and cross-sectional area.

[0019] Among them, the selection of multiple points from the surface to the core can be the surface, 0.2 times the layer depth, 0.5 times the layer depth, the layer depth boundary, or the core;

[0020] Steps S1-5: By analyzing the wear track data, extract the characteristic parameters of each depth point and correlate them with the microstructure and microhardness of each depth point;

[0021] Steps S1-6: Based on the deformation and energy wear theory of the Archard wear equation, establish a correlation model between these wear track characteristic parameters and the wear coefficient K, and fit the distribution function of the wear coefficient along the depth z. .

[0022] In steps S1-6, the fitted wear coefficient distribution function K(z) along depth z adopts the following exponential decay model:

[0023]

[0024] in, and These are the wear coefficients for the surface layer and the core, respectively. Let z be the attenuation coefficient, and z be the depth from the surface. It is an exponential function.

[0025] In step S2, the specific steps for implementing wear simulation using the finite element method include:

[0026] Step S2-1: Establish a three-dimensional finite element model of the target carburized part, and use the fitted wear coefficient function as a material property field that varies with depth to assign different regions of the model;

[0027] Step S2-2: Define boundary conditions, contact pairs, and load spectra that match the actual working conditions. Each incremental step corresponds to a micro-sliding distance ds. The Coulomb friction model is used in the contact definition, where the friction coefficient μ affects the tangential stress at the contact interface, thereby changing the normal contact pressure. Distribution;

[0028] Step S2-3: Simulation based on modified differential Arcard theory: In the finite element software, in each incremental step corresponding to the micro-sliding distance ds, the wear coefficient is determined according to the depth z at the node or integration point. Then, based on the contact pressure The differential Arcard formula is used to calculate local wear, and the wear increment of the infinitesimal element at depth z is obtained. ;

[0029] Step S2-4: Based on the calculations... The coordinates of the contact node are updated to change the geometry of the contact surface. At the same time, the mesh is refined in the region that exceeds the wear threshold to ensure calculation accuracy.

[0030] Step S2-5: Repeat steps S2-3 to S2-4 until N load cycles are completed, and record the cumulative wear depth distribution under each cycle or time step;

[0031] Step S2-6: Simulate wear evolution using mesh adaptive technology and calculate the critical wear depth. The required number of load cycles N is used to establish the relationship between wear depth and number of cycles. ;

[0032] Step S2-7: ... The relationship is fitted in a parametric analytical form, that is: , where the coefficient , It is expressed as a function of wear coefficient parameters, load, and friction coefficient.

[0033] In step S2-3, the differential Arcard formula is used to calculate local wear, and the wear amount of the micro-element at depth z is obtained. ,Right now:

[0034]

[0035] in, Let z be the wear amount of the infinitesimal element at depth z. To contact pressure, This represents the micro-sliding distance.

[0036] Steps S2-6 are specifically as follows:

[0037] The failure criterion for a component is set as the wear depth at a critical location reaching an allowable value. Establish the relationship curve δ(N) between cumulative wear depth δ and load cycle number N, and solve the problem by interpolation or fitting. Wear life This refers to the predicted lifetime under deterministic conditions.

[0038] Step S3 specifically includes:

[0039] Step S3-1: Obtain the deterministic wear model δ(N) and its parameterized form output from step S2; the parameterized form is a power-law model or a response surface model;

[0040] Step S3-2: Identify key random variables affecting wear life, including: calibration parameters of the wear coefficient function K(z). Normal load F, friction coefficient μ;

[0041] Step S3-3: Based on the experimental data, set the probability distribution type and distribution parameters for each random variable;

[0042] The probability distribution types include: normal distribution, log-normal distribution, or Weibull distribution.

[0043] Step S3-4: Construct the response surface mapping from input parameters to wear model coefficients: Use finite element simulation sample points to fit the analytical function relationship between wear model parameters and input parameters;

[0044] Step S3-5: Perform Monte Carlo simulation: Number of random samples Each time, a set of random variables is independently sampled. ); δ(N) curves for this set of samples are quickly generated using response surface mapping; and the equations are solved. The wear life of this sample was obtained. ;

[0045] Step S3-6: For all Statistical analysis of the sample: plotting the probability density histogram; calculating the reliability function R(N) = P( >N); Outputs the reliability lifetime or RN curve at the specified reliability level.

[0046] A system for measuring the wear coefficient and predicting the life of a carburized material at different locations, used to perform the method, comprising:

[0047] The gradient wear coefficient rapid measurement module includes: a reciprocating sliding wear testing machine, a deep-field surface profilometer or a white light interferometer, and a microstructure quantitative analysis unit based on a deep learning image recognition algorithm. This module is used to perform tests on the cross-section of a carburized sample using the reciprocating sliding wear testing machine, and to inversely derive the wear coefficient function along the depth distribution by combining hardness and microstructure data. ;

[0048] The finite element wear simulation and life modeling module is used to receive the distribution function of the wear coefficient along the depth. It automatically builds a three-dimensional finite element model of the target part, configures boundary conditions, contact pairs, and load spectra, and calls the finite element solver to perform wear simulation based on the differential Arcard formula, tracking the cumulative wear depth in real time. With load cycle number Relationship ;

[0049] The reliability analysis submodule is used to receive the cumulative wear depth from the simulation output. With load cycle number Relationship Based on the key random variables and their probability distribution parameters set by the user, the system automatically performs Monte Carlo simulations to calculate the wear life under each sampling. Statistical analysis generates the probability distribution and reliability function of wear life, and outputs the life prediction value under the specified reliability.

[0050] Graphical user interface for displaying including Curves, wear depth evolution cloud maps, The system includes curves and reliability-life curves, and supports user modification of load spectrum and allowable wear. Input parameters for the number of Monte Carlo simulations.

[0051] The rapid measurement module for gradient wear coefficient includes: a reciprocating sliding wear tester, a morphologist, and a tissue quantitative analysis unit based on a deep learning image recognition algorithm; the reciprocating sliding wear tester is used to perform reciprocating sliding wear tests at different positions along the depth direction on the cross section of the carburized sample, and to collect the friction coefficient and wear data at each position.

[0052] The profilometer is a super depth-of-field surface profilometer or a white light interferometer, used to measure the depth and cross-sectional area of ​​the wear marks at various locations.

[0053] The quantitative analysis unit is used to perform phase segmentation and quantitative calculation on metallographic images at different depths of the carburized layer, and output the volume fractions of martensite, retained austenite and ferrite.

[0054] The gradient wear coefficient rapid measurement module inputs the tissue volume fraction, microhardness and wear track characteristic parameters at each depth point into the wear coefficient inversion model based on the Archard wear equation, and fits the wear coefficient distribution function K(z) along the depth direction.

[0055] The present invention has the following beneficial effects and advantages:

[0056] 1. Compared with the traditional wear-through test, this invention can quickly invert the wear coefficient by testing the carburized section, thus shortening the material wear resistance assessment cycle from several days or even weeks to several hours.

[0057] 2. By introducing the differential Arcard formula and the wear coefficient function K(z), this invention accurately considers the essential influence of the carburized layer gradient characteristics on wear behavior. Compared with the homogeneous material assumption, the accuracy of wear simulation and life prediction is significantly improved.

[0058] 3. This invention establishes a continuous model from differential wear calculation to integral life assessment, with clear physical meaning and stronger predictive ability.

[0059] 4. The lifetime prediction results provided by this invention are in probabilistic form, containing uncertainty information, which is more in line with engineering practice and can provide direct input for reliability-based design (RBD) and condition-based maintenance (CBM).

[0060] 5. This invention is not only applicable to carburizing steel such as 16MnCrS5, but can also be extended to other surface hardening materials such as nitriding, carbonitriding, and spray coatings after appropriate model adjustments. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the overall technical solution of the present invention;

[0062] Figure 2 This is a schematic diagram of the cross-section of the carburized sample of the present invention;

[0063] Figure 3 This is a schematic diagram illustrating the reciprocating sliding wear test principle of the present invention;

[0064] Figure 4 This is a schematic diagram of the wear mark morphology according to an embodiment of the present invention;

[0065] Figure 5 This is a graph showing the distribution of wear mark depth along the sliding direction according to an embodiment of the present invention.

[0066] Figure 6 This is a schematic diagram illustrating the wear coefficient inversion principle of an embodiment of the present invention;

[0067] Figure 7 This is a flowchart of the differential Archard finite element wear simulation of the present invention, considering the gradient wear coefficient K(z);

[0068] Figure 8 This is a flowchart of the Monte Carlo simulation for predicting reliable wear life according to the present invention;

[0069] Figure 9 This is a schematic diagram of the wear reliability-life curve (RN curve) of the mechanical transmission key in an embodiment of the present invention;

[0070] Figure 10 This is a probability density function distribution diagram of the wear life of the mechanical transmission key in an embodiment of the present invention. Detailed Implementation

[0071] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0072] The overall technical solution flow of the present invention is as follows: Figure 1 The diagram shown is a flowchart of the overall technical solution of the present invention, which mainly includes three core modules: a gradient wear coefficient rapid measurement module, a finite element wear simulation and life modeling module considering gradient characteristics, and a wear reliability life prediction module.

[0073] The three modules work together to implement a method for measuring the wear coefficient and predicting the life of a carburized material at different locations, including the following steps:

[0074] Step S1: Using a reciprocating sliding wear testing machine, tests are performed on the cross-section of the carburized specimen from the surface to the core. Combining the hardness and microstructure data at the corresponding depths, a quantitative mapping relationship is established between depth, microstructure, hardness, wear characteristic parameters, and wear coefficient. The wear coefficient function distributed along the depth is then derived. ;

[0075] Step S2: Considering the gradient distribution of the wear coefficient, perform Archard wear finite element simulation and life modeling, establish the Archard wear finite element model, and apply the wear coefficient function... The Arcard wear finite element model is introduced as a field variable, and the differential Arcard formula is used to calculate local wear. The relationship between wear depth and cycle number is then established. ;

[0076] Step S3: Identify key random variables, based on the established... The model and its response surface approximation are used to perform Monte Carlo random sampling and calculate the wear life under each sampling. We obtain the probability distribution of wear life and the reliability function.

[0077] 1. Gradient Wear Coefficient Rapid Measurement Module

[0078] This module executes step S1, and its core equipment includes: a reciprocating sliding wear tester, a deep-field surface topology instrument, a white light interferometer, or a contact topology instrument. The implementation steps are as follows:

[0079] 1) Prepare the metallographic section of the carburized sample and accurately measure the Vickers hardness gradient from the surface to the core, such as... Figure 2 As shown.

[0080] 2) Analyze metallographic images of different locations in the carburized layer using deep learning image recognition algorithms (such as the UNet model) to quantitatively calculate the volume fraction of residual austenite, martensite, and other phases at each location.

[0081] 3) Pre-set the normal load and time. Determine 3 to 5 loads based on actual service conditions. The time should be sufficient to accurately measure the wear track morphology. Use the cross-section of the carburized sample as the test surface and a spherical friction pair (e.g., 6 mm in diameter) made of the same material as the friction pair. Conduct reciprocating sliding wear tests at different locations along the depth direction from the carburized surface (e.g., surface, 0.2 times the depth of the layer, 0.5 times the depth of the layer, the boundary between the layer depths, and the core). The sliding stroke at each location point is fixed.

[0082] 4) Characterize the wear tracks using a super depth-of-field surface profilometer, white light interferometer, or contact profilometer. Select multiple points from the surface to the core (e.g., surface, 0.2 times layer depth, 0.5 times layer depth, layer-depth boundary, core) to measure the wear track depth and cross-sectional area. Figures 3-6 As shown, the depth and cross-sectional area at different points on the wear track are correlated with the wear rate at that depth of the sample.

[0083] 5) By analyzing the wear track data, extract the characteristic parameters (such as wear track depth, wear track cross-sectional area, friction coefficient, etc.) of each depth point, and correlate them with the microhardness and microstructure of each depth point.

[0084] 6) Based on the deformation and energy wear theory of the Arcard wear equation, establish a correlation model between these wear track characteristic parameters and the wear coefficient K, and fit the distribution function K(z) of the wear coefficient along the depth z.

[0085] This embodiment uses the exponential decay model as follows:

[0086]

[0087] in, and These are the wear coefficients for the surface layer and the core, respectively. Let z be the attenuation coefficient, and z be the depth from the surface. It is an exponential function.

[0088] 2. Finite element wear simulation and life modeling module considering gradient characteristics

[0089] like Figure 7 As shown, the implementation steps S2 of this module are as follows:

[0090] 1) Establish a three-dimensional finite element model of the target carburized part (such as a mechanical transmission key).

[0091] 2) The wear coefficient function K(z) output by the rapid measurement module is used as a material property field that varies with depth and is assigned to different regions of the model.

[0092] 3) Define boundary conditions, contact pairs, and load spectra that correspond to actual working conditions. The Coulomb friction model is used in the contact definition, where the friction coefficient μ affects the tangential stress at the contact interface, thereby changing the normal contact pressure. The distribution of .

[0093] 4) Simulation based on modified differential Arcard theory: Wear calculation is performed in the finite element software through a user subroutine. In each incremental step (corresponding to the micro-sliding distance ds), the wear coefficient K(z) is determined according to the depth z at the node or integration point, and then the local micro-element wear depth is calculated according to the contact pressure p(z). The differential Arcard formula is used for local wear calculation to obtain the micro-element wear amount at depth z. ,Right now:

[0094]

[0095] in, Let z be the wear amount of the infinitesimal element at depth z. To contact pressure, This represents the micro-sliding distance.

[0096] 5) Based on the calculated wear amount of the micro-element at depth z The normal coordinates of the contact nodes are updated, and the mesh is refined in areas with large wear using mesh adaptive technology.

[0097] 6) Repeat the steps until N load cycles are completed, and record the cumulative wear depth distribution under each cycle or time step;

[0098] 7) Simulate wear evolution using mesh adaptive technology and calculate the critical wear depth. The required number of load cycles N is used to establish the relationship between wear depth and number of cycles. To facilitate subsequent Monte Carlo simulations, The relationship is fitted into a parametric analytical form, for example... , where the coefficient , It is expressed as a function of wear coefficient parameters, load, and friction coefficient (the mapping relationship is established through the response surface methodology).

[0099] The failure criterion for a component is set as the wear depth at a critical location reaching an allowable value. Establish the relationship curve between cumulative wear depth δ and load cycle number N. And solve the problem by interpolation or fitting. Wear life This refers to the predicted lifetime under deterministic conditions.

[0100] 3. Wear-Reliable Life Prediction Module

[0101] like Figure 8 As shown, the implementation steps S3 of the wear-reliable life prediction module are as follows:

[0102] 1) Accept the above deterministic wear model, i.e. Relationship. Receive the deterministic wear model output from step S2. And its parameterized form (such as power-law models or response surface models constructed from a small number of finite element simulations).

[0103] 2) Identify key random variables affecting wear life, such as the calibration parameters of the wear coefficient function K(z). , , The parameters include the normal load F and the friction coefficient μ. The friction coefficient μ indirectly affects the wear rate by changing the contact pressure distribution, and therefore is included in the analysis as a random variable.

[0104] 3) Based on engineering experience or test data, determine a reasonable probability distribution type (such as normal distribution, log-normal distribution, or Weibull distribution) and distribution parameters for these random variables. For example, wear-related parameters usually exhibit a right-skewed distribution, and log-normal or Weibull distributions can better fit experimental observations.

[0105] 4) Constructing a response surface mapping from input parameters to wear model coefficients: Using a small number of finite element simulation sample points, fit the wear model parameters (such as power law coefficients A and B) to the input parameters (…). , , The analytical functional relationship between δ(N), F, and μ is established. This response surface model can quickly calculate the δ(N) curve under any combination of input parameters, avoiding the need to repeatedly perform the complete finite element simulation during the Monte Carlo process.

[0106] 5) The Monte Carlo simulation method was used to conduct... Second-rate( Simulations are conducted using random sampling (1000~10000). In each simulation: a set of input parameter samples is randomly generated; the δ(N) curve for this set of parameters is quickly generated through response surface mapping; and the solution is obtained. The wear life under this sampling was obtained. ;storage .

[0107] 6) Statistical analysis of all Samples were used to obtain wear life. The probability density function and cumulative distribution function are used to evaluate the reliability under a given allowable wear level. Alternatively, predict the reliability lifetime at which the target reliability (e.g., 90%) is achieved, and plot the reliability-lifetime curve (RN curve), such as... Figures 9-10 As shown.

[0108] Example:

[0109] Wear life prediction using splines in a certain type of 16MnCrS5 mechanical transmission system as an example; to verify the effectiveness of the invention, the study was conducted using 16MnCrS5 steel mechanical transmission keys for a certain type of engineering machinery.

[0110] 1. Gradient Wear Coefficient Measurement

[0111] 1) Cut, inlay, polish and etch the transmission key sample to prepare a metallographic sample.

[0112] 2) Use a microhardness tester to measure the hardness gradient from the surface to the core along the cross section.

[0113] 3) A deep learning algorithm (a trained UNet model) was used to analyze the SEM images of the carburized layer structure and calculate the content of residual austenite at different depths.

[0114] 4) The required wear tracks were obtained using a reciprocating sliding wear tester. Five characteristic points were selected for testing: the surface (high carbon high hardness martensite region), the transition region (higher retained austenite region), 0.5 times the effective layer depth, the layer depth boundary, and the core (low carbon sorbite region).

[0115] 5) Input the hardness, retained austenite content, and wear track characteristic parameters at each point into the calibrated inversion model to obtain the wear coefficient K value at each depth. Through data fitting, obtain the distribution function of the wear coefficient along depth z, for example:

[0116]

[0117] The results showed that the surface layer had the lowest K value (approximately 1.5 × 10⁻⁻⁻⁶). 6 The highest K value was found in the heart (approximately 6.0 × 10⁻). 6 ).

[0118] 2. Finite element wear simulation and deterministic life modeling

[0119] 1) Establish a three-dimensional finite element model of the mechanical transmission key. Assign the wear coefficient distribution function K(z) obtained in the previous step to the model in the form of material properties.

[0120] Applying periodic contact loads during mechanical transmission.

[0121] 2) Based on the differential Arcard formula and adaptive mesh technology, the wear depth evolution process of the contact surface under cyclic load is simulated and calculated.

[0122] 3) The maximum wear depth of the contact surface was obtained through simulation. Curve of variation with load cycle number N .

[0123] 4) Set allowable wear amount Deterministic predicted lifetimes are obtained through curve interpolation. .

[0124] 3. Reliable life prediction

[0125] 1) Determine the key parameters in the wear coefficient function (e.g.) , , ) and maximum contact stress Let them be random variables. Assume they all follow a log-normal distribution and assign them corresponding means and coefficients of variation (e.g., 15%).

[0126] 2) Run 10,000 Monte Carlo simulations. For each simulation, recalculate K(z) based on randomly generated parameters and quickly calculate the lifetime using the same finite element wear algorithm logic. .

[0127] 3) Statistical analysis of lifespan distribution: average lifespan The standard deviation is 1.2 × 10⁻⁶. 5 cycles.

[0128] 4) The reliability function R(N) indicates the reliable life of the mechanical transmission key when the reliability requirement is 90%. 6.5×10 5 This result provides a precise basis for decision-making regarding the preventive maintenance and replacement of transmission keys.

[0129] In summary, this invention proposes a method and system for measuring the wear coefficient and predicting the life of carburized materials at different locations. This method rapidly inverts the gradient wear coefficient function distributed along depth through reciprocating sliding wear tests. By introducing this as a field variable into finite element wear simulation based on differential Arcard formulas, a deterministic relationship between wear depth and cycle number is established. Furthermore, by combining Monte Carlo simulation and stochastic process models, a quantitative assessment of wear life uncertainty is achieved, outputting a reliability function and a predicted reliable life value. This invention overcomes the limitations of traditional wear-through tests, such as long testing cycles and the inability to obtain gradient information. It establishes an integrated closed-loop technical framework from material gradient characteristic testing and finite element simulation to probabilistic life prediction, significantly improving the accuracy and efficiency of wear life prediction for carburized parts, and providing a direct basis for reliability-based design and condition-based maintenance. This invention is not only applicable to carburized steels such as 16MnCrS5, but can also be extended to other surface-hardening materials and gradient materials such as nitriding, carbonitriding, and spray coatings. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0130] Those skilled in the art will understand that the above description is merely a preferred embodiment of the present invention, and the features described in the various embodiments of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. This is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for measuring the wear coefficient and predicting the life of a carburized material at different locations, characterized in that, Includes the following steps: Step S1: Using a reciprocating sliding wear testing machine, tests are performed on the cross-section of the carburized specimen from the surface to the core. Combining the hardness and microstructure data at the corresponding depths, a quantitative mapping relationship is established between depth, microstructure, hardness, wear characteristic parameters, and wear coefficient. The wear coefficient function distributed along the depth is then derived. ; Step S2: Considering the gradient distribution of the wear coefficient, perform Archard wear finite element simulation and life modeling, establish the Archard wear finite element model, and apply the wear coefficient function... The Arcard wear finite element model is introduced as a field variable, and the differential Arcard formula is used to calculate local wear. The relationship between wear depth and cycle number is then established. ; Step S3: Identify the key random variables affecting wear life and set a probability distribution for each random variable; use the Monte Carlo method to perform multiple random samplings, with each sampling based on the relationship between wear depth and cycle number established in Step S2. To obtain the corresponding wear life. For all Perform statistical analysis to generate the probability distribution and reliability function of wear life, and output the life prediction value.

2. The method for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 1, characterized in that, Step S1 specifically includes: Step S1-1: Prepare the metallographic section of the carburized sample and measure the Vickers hardness gradient from the surface to the core; Step S1-2: Analyze the metallographic images of different locations in the carburized layer using a deep learning image recognition algorithm, and quantitatively calculate the volume fraction of martensite, ferrite, and retained austenite phases at each location. Step S1-3: Pre-set the load and time, take the cross section of the carburized sample as the test surface, and use a spherical friction pair made of the same material to conduct a reciprocating sliding wear test at different positions along the depth direction from the carburized surface, and the sliding stroke at each position is fixed; Steps S1-4: Characterize the wear marks using a super depth-of-field surface profilometer, white light interferometer, or contact profilometer. Select multiple points from the surface to the core and measure the wear mark depth and cross-sectional area. Among them, the selection of multiple points from the surface to the core can be the surface, 0.2 times the layer depth, 0.5 times the layer depth, the layer depth boundary, or the core; Steps S1-5: By analyzing the wear track data, extract the characteristic parameters of each depth point and correlate them with the microstructure and microhardness of each depth point; Steps S1-6: Based on the deformation and energy wear theory of the Archard wear equation, establish a correlation model between these wear track characteristic parameters and the wear coefficient K, and fit the distribution function of the wear coefficient along the depth z. .

3. The method for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 2, characterized in that, In steps S1-6, the fitted wear coefficient distribution function K(z) along depth z adopts the following exponential decay model: ； in, and These are the wear coefficients for the surface layer and the core, respectively. Let z be the attenuation coefficient, and z be the depth from the surface. It is an exponential function.

4. The method for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 1, characterized in that, The specific steps for achieving wear simulation using the finite element method in step S2 include: Step S2-1: Establish a three-dimensional finite element model of the target carburized part, and use the fitted wear coefficient function as a material property field that varies with depth to assign different regions of the model; Step S2-2: Define boundary conditions, contact pairs, and load spectra that match the actual working conditions. Each incremental step corresponds to a micro-sliding distance ds. The Coulomb friction model is used in the contact definition, where the friction coefficient μ affects the tangential stress at the contact interface, thereby changing the normal contact pressure. Distribution; Step S2-3: Simulation based on modified differential Arcard theory: In the finite element software, in each incremental step corresponding to the micro-sliding distance ds, the wear coefficient is determined according to the depth z at the node or integration point. Then, based on the contact pressure The differential Arcard formula is used to calculate local wear, and the wear increment of the infinitesimal element at depth z is obtained. ; Step S2-4: Based on the calculations... The coordinates of the contact node are updated to change the geometry of the contact surface. At the same time, the mesh is refined in the area exceeding the wear threshold using mesh adaptive technology to ensure calculation accuracy. Step S2-5: Repeat steps S2-3 to S2-4 until N load cycles are completed, and record the cumulative wear depth distribution under each cycle or time step; Step S2-6: Simulate wear evolution using mesh adaptive technology and calculate the critical wear depth. The required number of load cycles N is used to establish the relationship between wear depth and number of cycles. ; Step S2-7: ... The relationship is fitted in a parametric analytical form, that is: , where the coefficient , It is expressed as a function of wear coefficient parameters, load, and friction coefficient.

5. The method for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 4, characterized in that, In step S2-3, the differential Arcard formula is used to calculate local wear, and the wear amount of the micro-element at depth z is obtained. ,Right now: ； in, Let z be the wear amount of the infinitesimal element at depth z. To contact pressure, This represents the micro-sliding distance.

6. The method for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 4, characterized in that, Steps S2-6 are specifically as follows: The failure criterion for a component is set as the wear depth at a critical location reaching an allowable value. Establish the relationship curve δ(N) between cumulative wear depth δ and load cycle number N, and solve the problem by interpolation or fitting. Wear life This refers to the predicted lifetime under deterministic conditions.

7. The method for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 1, characterized in that, Step S3 specifically includes: Step S3-1: Obtain the deterministic wear model δ(N) and its parameterized form output from step S2; the parameterized form is a power-law model or a response surface model; Step S3-2: Identify key random variables affecting wear life, including: calibration parameters of the wear coefficient function K(z). Normal load F, friction coefficient μ; Step S3-3: Based on the experimental data, set the probability distribution type and distribution parameters for each random variable; The probability distribution types include: normal distribution, log-normal distribution, or Weibull distribution; Step S3-4: Construct the response surface mapping from input parameters to wear model coefficients: Use finite element simulation sample points to fit the analytical function relationship between wear model parameters and input parameters; Step S3-5: Perform Monte Carlo simulation: Number of random samples Each time, a set of random variables is independently sampled. ); δ(N) curves for this set of samples are quickly generated using response surface mapping; and the equations are solved. The wear life of this sample was obtained. ; Steps S3-6: For all Statistical analysis of the sample: plotting the probability density histogram; calculating the reliability function R(N) = P( >N); Outputs the reliability lifetime or RN curve at the specified reliability level.

8. A system for measuring the wear coefficient and predicting the life of a carburized material at different locations, used to perform the method as described in claim 1, characterized in that, include: A rapid gradient wear coefficient measurement module includes: a reciprocating sliding wear testing machine, a deep-field surface profilometer or a white light interferometer, and a microstructure quantitative analysis unit based on a deep learning image recognition algorithm. This module is used to perform tests on the cross-section of a carburized sample using the reciprocating sliding wear testing machine, and to inversely derive the wear coefficient function along the depth distribution by combining hardness and microstructure data. ; The finite element wear simulation and life modeling module is used to receive the distribution function of the wear coefficient along the depth. It automatically builds a three-dimensional finite element model of the target part, configures boundary conditions, contact pairs, and load spectra, and calls the finite element solver to perform wear simulation based on the differential Arcard formula, tracking the cumulative wear depth in real time. With load cycle number Relationship ; The reliability analysis submodule is used to receive the cumulative wear depth from the simulation output. With load cycle number Relationship Based on the key random variables and their probability distribution parameters set by the user, the system automatically performs Monte Carlo simulations to calculate the wear life under each sampling. Statistical analysis generates the probability distribution and reliability function of wear life, and outputs the life prediction value under the specified reliability. A graphical user interface for displaying including Curves, wear depth evolution cloud maps, The system includes curves and reliability-life curves, and supports user modification of load spectrum and allowable wear. Input parameters for the number of Monte Carlo simulations.

9. The system for measuring the wear coefficient and predicting the life of a carburized material at different locations according to claim 8, characterized in that, The rapid measurement module for gradient wear coefficient includes: a reciprocating sliding wear tester, a morphologist, and a tissue quantitative analysis unit based on a deep learning image recognition algorithm; the reciprocating sliding wear tester is used to perform reciprocating sliding wear tests at different positions along the depth direction on the cross section of the carburized sample, and to collect the friction coefficient and wear data at each position. The profilometer is a super depth-of-field surface profilometer or a white light interferometer, used to measure the depth and cross-sectional area of ​​the wear marks at various locations. The quantitative analysis unit is used to perform phase segmentation and quantitative calculation on metallographic images at different depths of the carburized layer, and output the volume fractions of martensite, retained austenite and ferrite. The gradient wear coefficient rapid measurement module inputs the tissue volume fraction, microhardness and wear track characteristic parameters at each depth point into the wear coefficient inversion model based on the Archard wear equation, and fits the wear coefficient distribution function K(z) along the depth direction.

Images