A method, equipment, medium, and product for predicting lubrication performance across scales.

By combining first-principles molecular dynamics and hybrid lubrication numerical models, the problem of the disconnect between microscopic simulation and macroscopic prediction has been solved, enabling quantitative prediction and optimization of lubrication performance across scales and providing precise guidance for the design of lubrication systems for high-end equipment under extreme operating conditions.

CN123120139APending Publication Date: 2026-11-17LANZHOU INSTITUTE OF CHEMICAL PHYSICS CHINESE ACADEMY OF SCIENCES +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611303302.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-26
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, microscopic simulation and macroscopic lubrication performance prediction are disconnected. Macroscopic models lack support from microscopic physical properties, resulting in insufficient accuracy in lubrication performance prediction and limited applicability across all operating conditions, making it difficult to meet the lubrication performance requirements of high-end equipment under extreme operating conditions.

Method used

First-principles molecular dynamics methods are used to obtain thermodynamically stable equilibrium trajectories, calculate microstructural characteristic parameters and macroscopic thermophysical transport parameters, and perform benchmark verification through a hybrid lubrication numerical model to establish a cross-scale lubrication performance prediction method, achieving reliable transmission and coupling between microscopic and macroscopic parameters.

Benefits of technology

It enables quantitative prediction of lubrication performance across scales, from the microscopic atomic scale to the macroscopic full-condition range, providing a theoretical basis for lubrication medium modification design and friction pair system performance optimization, and ensuring prediction accuracy and applicability to all operating conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123120139A_ABST
    Figure CN123120139A_ABST
Patent Text Reader

Abstract

This application discloses a method, device, medium, and product for predicting cross-scale lubrication performance, relating to the field of lubrication material performance technology. The method includes: determining the lubrication medium and friction pair system, constructing a set of operating conditions; obtaining the equilibrium trajectory from first-principles molecular dynamics simulations, calculating microstructure and macroscopic physical property parameters, and calibrating them with experimental data; constructing a hybrid lubrication numerical model and verifying it with benchmark examples; substituting the calibrated macroscopic physical property parameters into the hybrid lubrication numerical model, solving the set of operating conditions point by point to obtain the lubrication performance under all operating conditions; determining the lubrication state based on the film thickness ratio, and establishing an empirical formula for the critical speed; analyzing the cross-scale action mechanism based on the microstructure and lubrication performance, and establishing the action chain from component modification to lubrication performance. This application can solve the problems of insufficient accuracy caused by the disconnect between micro and macroscopic aspects, reliance on empirical parameters, and lack of standardized procedures in existing predictions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of lubricating material performance technology, and in particular to a method, equipment, medium and product for predicting lubricating performance across scales. Background Technology

[0002] High-performance lubrication systems are crucial for the stable operation of core equipment such as high-end rotating machinery and precision transmission devices. Their lubrication performance directly determines the equipment's operational accuracy, service life, and reliability. As equipment develops towards extreme operating conditions (high vacuum, strong radiation, wide temperature range, heavy load and low speed), traditional lubrication materials and lubrication design methods face severe challenges.

[0003] Current methods for modifying and evaluating the performance of lubricating materials mainly rely on experimental trial and error and single-scale numerical simulations, which present two major bottlenecks: First, there is a disconnect between microscopic atomic simulations and macroscopic lubrication performance predictions. High-precision microscopic simulation methods such as first-principles molecular dynamics can only calculate the microscopic structure and intrinsic properties of materials, and cannot directly predict the macroscopic lubrication performance of friction pairs across the entire operating range, making it difficult to establish a quantitative correlation between microscopic modification and macroscopic performance. Second, macroscopic lubrication numerical models are highly dependent on empirical physical property parameters and lack parameter support at the microscopic mechanism level, which limits the accuracy of model predictions. This is especially true for novel modified lubricating media, where the lack of accurate physical property data often prevents reliable performance simulations.

[0004] Existing cross-scale lubrication research often employs simple parameter transfer methods, lacking standardized processes for microscopic parameter calibration, macroscopic model verification, and multi-condition performance mapping. This hinders the systematic revelation of the cross-scale mechanisms of lubricant modification and makes it difficult to achieve high-precision and rapid prediction of lubrication performance across a wide range of operating conditions. This situation results in long development cycles and high costs for novel lubricating materials, and a lack of precise theoretical guidance for the condition-matching design of lubrication systems, making it difficult to meet the stringent lubrication performance requirements of high-end equipment under extreme conditions. Therefore, there is an urgent need to develop a standardized, high-precision cross-scale lubrication performance prediction method that organically combines microscopic atomic-scale mechanistic analysis with quantitative prediction of macroscopic tribological properties, providing systematic technical support for lubricant modification design and friction pair system performance optimization. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, medium and product for predicting cross-scale lubrication performance, which can solve the technical problems in existing cross-scale lubrication performance prediction, such as the disconnect between microscopic simulation and macroscopic prediction, the lack of microscopic physical property support for macroscopic models and reliance on empirical parameters, and the lack of standardized prediction process, which lead to insufficient prediction accuracy and limited applicability to all working conditions.

[0006] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for predicting cross-scale lubrication performance, including: Determine the lubrication system of the lubricating medium to be predicted and the target friction pair, and construct a set of calculation conditions covering temperature, speed of movement, and applied load; Obtain the thermodynamically stable equilibrium trajectory; the thermodynamically stable equilibrium trajectory is obtained by simulating several representative temperature points in the calculation case set using first-principles molecular dynamics under isothermal and isobaric ensemble conditions. Microstructural characteristic parameters and macroscopic thermophysical transport parameters are calculated based on the equilibrium trajectory. The macroscopic thermophysical transport parameters were compared and verified with publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters. The hybrid lubrication numerical model was benchmarked using a classical sliding friction pair example, resulting in a benchmark-verified macroscopic hybrid lubrication numerical model. The hybrid lubrication numerical model is a hybrid lubrication numerical model of the friction pair lubrication system, which is constructed based on the fluid lubrication control equation, the rough surface micro-protrusion contact theory, and the load balance equation, and includes liquid film pressure, film thickness distribution, contact pressure, friction force, and friction loss. The calibrated macroscopic thermophysical transport parameters are substituted into the benchmark-verified hybrid lubrication numerical model as boundary conditions, and the numerical solutions are obtained point by point for each working point in the calculation working condition set to obtain the lubrication performance characteristic parameters of the friction pair in the entire working condition range. Based on the lubrication performance characteristic parameters, the film thickness ratio is used as the criterion for the state transition between hydrodynamic lubrication and mixed lubrication. The critical conditions for the lubrication state transition under different temperatures and different applied loads are determined, and a quantitative empirical correlation formula for the critical speed of temperature-load dual variables is established. Based on the microstructure characteristic parameters and the lubrication performance characteristic parameters, the cross-scale action mechanism of the modification of the lubricating medium to be predicted is analyzed, and a complete cross-scale action chain from component modification to macroscopic lubrication performance is established.

[0007] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the cross-scale lubrication performance prediction method described in any one of the above.

[0008] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the cross-scale lubrication performance prediction method described in any one of the above descriptions.

[0009] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the cross-scale lubrication performance prediction method described above.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, device, medium, and product for predicting lubrication performance across scales. By obtaining the stable equilibrium trajectory of thermodynamic parameters simulated using first-principles molecular dynamics, this application calculates macroscopic thermophysical transport parameters and compares these parameters with publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters. This ensures the accuracy of microscopic property calculations and provides accurate input parameters with microscopic mechanism support for macroscopic prediction. Furthermore, this application uses a classical sliding friction pair example to benchmark a hybrid lubrication numerical model constructed based on fluid lubrication control equations, rough surface micro-protrusion contact theory, and load balance equations. This yields a benchmark-verified macroscopic hybrid lubrication numerical model, ensuring the computational accuracy and reliability of the macroscopic solution method and overcoming the problem of insufficient prediction accuracy in macroscopic lubrication numerical models that rely on empirical property parameters. This application achieves reliable transfer and coupling between microscale physical properties and macroscale lubrication models by substituting calibrated macroscopic thermophysical transport parameters as boundary conditions into a benchmark-validated hybrid lubrication numerical model. Based on this, it solves point-by-point a set of computational conditions covering multiple dimensions such as temperature, velocity, and applied load to obtain lubrication performance characteristic parameters across the entire operating range, realizing quantitative prediction of lubrication performance across scales from the microscopic atomic scale to the macroscopic full operating range. Furthermore, this application uses film thickness ratio as the state transition criterion between hydrodynamic lubrication and hybrid lubrication, determines the critical conditions for lubrication state transitions under different temperatures and applied loads, and establishes a quantitative empirical correlation formula for the critical speed of the temperature-load dual-variable system. This enables rapid prediction of lubrication state boundaries over a wide operating range, providing a quantitative basis for lubrication system design under different service conditions. This application analyzes the cross-scale action mechanism of the modification of the lubricating medium to be predicted based on the microstructure characteristic parameters and lubrication performance characteristic parameters, establishes a complete cross-scale action chain from component modification to macroscopic lubrication performance, realizes a systematic mechanistic analysis from microstructure evolution to macroscopic performance change, clarifies the quantitative transmission relationship between each scale, and provides a complete theoretical basis and methodological support for the design of lubricating medium modification and the performance optimization of friction pair system. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is an application environment diagram of a cross-scale lubrication performance prediction method in one embodiment of this application; Figure 2 A flowchart illustrating a cross-scale lubrication performance prediction method provided in an embodiment of this application; Figure 3 A flowchart illustrating a cross-scale lubrication performance prediction method provided in another embodiment of this application; Figure 4 This is a schematic diagram of the microscopic geometric model of GaInSn liquid metal and Cu-doped GaInSn liquid metal provided in another embodiment of this application; Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] This application belongs to the field of lubricant material performance calculation and tribological numerical simulation technology. Specifically, it relates to a cross-scale lubrication performance prediction method that combines first-principles molecular dynamics and continuous medium lubrication model. It is applicable to the modification design of various liquid and semi-solid lubricating media and the high-precision evaluation of the lubrication performance of friction pair systems with different structures. It can realize the organic connection between microscopic mechanism analysis and macroscopic performance quantitative prediction, and provide methodological support for the optimization design of high-performance lubrication systems under extreme conditions.

[0015] This application addresses the problems in existing technologies, such as the disconnect between microscopic simulation and macroscopic performance prediction, the lack of standardized procedures for cross-scale calculations, and insufficient prediction accuracy and reliability. It provides a cross-scale lubrication performance prediction method that combines first-principles molecular dynamics and a continuous medium hybrid lubrication model. This application employs a technical path of "microscopic-macroscopic hierarchical decoupling calculation → dual calibration verification → cross-scale combined prediction → mechanism system analysis." Through hierarchical calculation, high-precision calculation of microscopic properties and accurate construction of the macroscopic lubrication model are achieved. After dual verification, reliable transfer of cross-scale parameters is realized, ultimately establishing a performance prediction method with a wide operating range and a microscopic-macroscopic mechanism correlation.

[0016] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] The cross-scale lubrication performance prediction method provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server.

[0018] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0019] In one exemplary embodiment, such as Figure 2 As shown, a method for predicting cross-scale lubrication performance is provided. This method is executed by a computer device, specifically a terminal or server, or both. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S1 to S8. Wherein: S1. Determine the lubrication system of the lubricating medium to be predicted and the target friction pair, and construct a set of calculation conditions covering temperature, speed, and applied load.

[0020] S2. Obtain the equilibrium trajectory with stable thermodynamic parameters; the equilibrium trajectory with stable thermodynamic parameters is obtained by simulating several representative temperature points in the calculation condition set under isothermal and isobaric ensemble using first-principles molecular dynamics.

[0021] S3. Calculate the microstructure characteristic parameters and macroscopic thermophysical transport parameters based on the equilibrium trajectory.

[0022] S4. The macroscopic thermophysical transport parameters are compared and verified with the publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters.

[0023] S5. The hybrid lubrication numerical model is benchmarked using a classical sliding friction pair example to obtain a benchmark-verified macroscopic hybrid lubrication numerical model. The hybrid lubrication numerical model is a hybrid lubrication numerical model of the friction pair lubrication system, which is constructed based on the fluid lubrication control equation, the rough surface micro-protrusion contact theory and the load balance equation, and includes liquid film pressure, film thickness distribution, contact pressure, friction force and friction loss.

[0024] S6. Substitute the calibrated macroscopic thermophysical transport parameters as boundary conditions into the benchmark-verified hybrid lubrication numerical model, and solve the numerical values ​​of each working point in the calculation working condition set to obtain the lubrication performance characteristic parameters of the friction pair in the entire working condition range.

[0025] S7. Based on the lubrication performance characteristic parameters, the film thickness ratio is used as the criterion for the state transition between hydrodynamic lubrication and mixed lubrication. The critical conditions for the lubrication state transition under different temperatures and different applied loads are determined, and a quantitative empirical correlation formula for the critical speed of temperature-load dual variables is established.

[0026] S8. Based on the microstructure characteristic parameters and the lubrication performance characteristic parameters, analyze the cross-scale action mechanism of the modification of the lubricating medium to be predicted, and establish a complete cross-scale action chain from component modification to macroscopic lubrication performance.

[0027] This embodiment obtains stable equilibrium trajectories of thermodynamic parameters simulated using first-principles molecular dynamics methods, calculates macroscopic thermophysical transport parameters, and compares these parameters with publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters. This ensures the accuracy of microscopic-scale property calculations and provides accurate input parameters with microscopic mechanism support for macroscopic predictions. This embodiment also uses a classical sliding friction pair example to benchmark a hybrid lubrication numerical model constructed based on fluid lubrication control equations, rough surface micro-protrusion contact theory, and load balance equations. This yields a benchmark-verified macroscopic hybrid lubrication numerical model, ensuring the computational accuracy and reliability of the macroscopic solution method and overcoming the problem of insufficient prediction accuracy in macroscopic lubrication numerical models due to reliance on empirical property parameters. This embodiment achieves reliable transfer and coupling between microscale physical properties and macroscale lubrication models by substituting calibrated macroscopic thermophysical transport parameters as boundary conditions into a benchmark-validated hybrid lubrication numerical model. Based on this, it solves point-by-point a set of calculation conditions covering multiple dimensions such as temperature, velocity, and applied load to obtain lubrication performance characteristic parameters across the entire operating range, realizing quantitative prediction of lubrication performance across scales from the microscopic atomic scale to the macroscopic full operating range. Furthermore, this embodiment uses film thickness ratio as the state transition criterion between hydrodynamic lubrication and hybrid lubrication, determining the critical conditions for lubrication state transition under different temperatures and applied loads, and establishing a quantitative empirical correlation formula for the critical speed of the temperature-load dual-variable system. This enables rapid prediction of lubrication state boundaries over a wide operating range, providing a quantitative basis for lubrication system design under different service conditions. This embodiment analyzes the cross-scale action mechanism of the modification of the lubricating medium to be predicted based on the microstructure characteristic parameters and lubrication performance characteristic parameters, establishes a complete cross-scale action chain from component modification to macroscopic lubrication performance, realizes a systematic mechanistic analysis from microstructure evolution to macroscopic performance change, clarifies the quantitative transmission relationship between each scale, and provides a complete theoretical basis and methodological support for lubricating medium modification design and friction pair system performance optimization.

[0028] In another exemplary embodiment of this application, step S1 mainly performs the process of constructing a cross-scale computing system and a working condition mapping framework, specifically including the following steps: S11. Define the thermophysical transfer parameters: The lubrication system of the lubricating medium material to be studied and the target friction pair is determined, and the thermophysical transfer parameters between the microscopic atomic simulation system and the macroscopic continuous medium hybrid lubrication model are clarified. The thermophysical transfer parameters include parameters such as the dynamic viscosity, density and kinematic viscosity of the lubricating medium.

[0029] S12. Define the calculation set of operating conditions: A set of calculation conditions covering multiple dimensions of variables such as temperature, speed, and applied load is constructed. The temperature, speed, and load ranges cover the entire operating range of the friction pair in actual operation, ensuring the engineering applicability of the calculation results.

[0030] In another exemplary embodiment of this application, step S2 described above may be implemented by the following steps: S21. Construct a periodic supercell atomic model of the lubricating medium to be predicted, and perform high-temperature melting and structural relaxation treatment on the periodic supercell atomic model.

[0031] S22. Using the first-principles molecular dynamics method based on density functional theory, the electron-ion interaction mode, exchange correlation energy, plane wave cutoff energy, number of k-points, temperature control method, time step and total simulation time are set, and simulations are carried out on several representative temperature points in the calculation case set under isothermal and isobaric ensemble.

[0032] S23. Discard the initial transient segment and retain the equilibrium segment trajectory with stable thermodynamic parameters.

[0033] In another exemplary embodiment of this application, when constructing the periodic supercell atomic model in step S21 above, the pure medium system is configured with each element atom according to the target component ratio; the doping modification system adopts a substitution doping method, replacing the matrix atoms with dopant atoms while keeping the total number of atoms unchanged.

[0034] After the high-temperature melting treatment and structural relaxation treatment bring the initial configuration to an isotropic and uniform state, the equilibrium trajectory is then extracted for statistical calculation of physical properties.

[0035] The microstructure characteristic parameters mentioned in step S8 above include radial distribution function, coordination number, and atomic diffusion coefficient.

[0036] The macroscopic thermophysical transport parameters mentioned in step S3 above include dynamic viscosity and density; the dynamic viscosity is based on the Green-Kubo linear response theory and is obtained by time integration of the autocorrelation function of the off-diagonal stress tensor components.

[0037] In another exemplary embodiment of this application, the macroscopic thermophysical transport parameters in step S3 above include dynamic viscosity, density, and kinematic viscosity; the dynamic viscosity is used as a fluid property input parameter for the average Reynolds equation in the hybrid lubrication numerical model; the density is used for calculating the liquid film inertial effect in the hybrid lubrication numerical model; and the kinematic viscosity is used for benchmark unification conversion in the step of "comparing and verifying the macroscopic thermophysical transport parameters with publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters".

[0038] In another exemplary embodiment of this application, steps S2-S4 above mainly perform first-principles molecular dynamics simulations and quantification of physical property parameters at the microscale, and mainly include the following steps: A1. Constructing a microscopic geometric model: A periodic supercell atomic model of the lubricating medium was constructed. In the pure medium system, the atoms of each element were configured according to the target composition ratio. In the doping modification system, a substitution doping method was adopted, replacing the matrix atoms with dopant atoms while keeping the total number of atoms unchanged, thus avoiding the introduction of local structural distortions. All initial configurations were subjected to high-temperature melting and sufficient structural relaxation to eliminate artificially applied atomic order. Statistical sampling was carried out after the system relaxed to an isotropic homogeneous state.

[0039] A2. First-principles molecular dynamics simulations: Simulations were performed using first-principles molecular dynamics based on density functional theory, with calculation parameters including electron-ion interaction modes, exchange correlation energies, plane wave cutoff energies, and the number of k-points. All molecular dynamics simulations were conducted under isothermal and isobaric ensembles, with operating parameters such as temperature control, time step, and total simulation duration set. Simulations were performed at multiple representative temperature points within the target temperature range. After discarding the initial transient phase, the thermodynamically stable equilibrium phase was retained for property calculations.

[0040] A3. Calculation Results Statistics and Data Processing: Based on the equilibrium trajectory, the microstructure characteristic parameters and macroscopic thermophysical transport parameters are calculated: the microstructure parameters include the radial distribution function, coordination number, and atomic diffusion coefficient, which are used to characterize the short-range order and atomic dynamics of the system; the macroscopic thermophysical transport parameters include dynamic viscosity and density, and the dynamic viscosity is calculated by integrating the stress tensor autocorrelation function using linear response theory.

[0041] A4. Verification of calculation results: The simulated macroscopic transport parameters were compared and analyzed with the experimental fitting data of kinematic viscosity of GaInSn eutectic liquid metal in published literature (the physical property benchmark value obtained by fitting the experimental viscosity test results to the temperature). After unifying the comparison benchmark, the relative error at different temperatures was quantified to verify the calculation accuracy of the microscopic simulation method and the reliability of the physical property parameters, and to provide high-precision input parameters for subsequent macroscopic models.

[0042] In another exemplary embodiment of this application, step S5 above may be implemented by the following steps: The macroscopic lubrication performance parameters are obtained by numerically solving the hybrid lubrication numerical model. The macroscopic lubrication performance parameters are compared with the benchmark results of the classical sliding friction pair example to verify the calculation accuracy and reliability of the hybrid lubrication numerical model, thus obtaining the benchmark-verified macroscopic hybrid lubrication numerical model.

[0043] The hybrid lubrication numerical model includes a liquid film pressure equation, a film thickness distribution equation, a micro-protrusion contact pressure equation, a load balance equation, and a friction force equation. The liquid film pressure equation adopts the Patir-Cheng average Reynolds equation, which considers the surface roughness effect. The micro-protrusion contact pressure equation adopts the Greenwood-Tripp equation and introduces a flow factor and a shear stress factor related to surface roughness. It is assumed that the surface micro-protrusion height follows a normal distribution and the surface morphology is isotropic.

[0044] Specifically, this embodiment mainly performs the process of constructing a macroscopic-scale hybrid lubrication numerical model and benchmark verification, which mainly includes the following steps: S51. Establishment of a macroscopic-scale hybrid lubrication model: Based on the fluid lubrication control equation, the rough surface micro-protrusion contact theory, and the load balance equation, a hybrid lubrication numerical model of the friction pair system is constructed. A complete set of control equations including liquid film pressure, film thickness distribution, contact pressure, friction force, and friction loss is established. Flow factor and shear stress factor related to surface roughness are introduced. It is assumed that the height of the surface micro-protrusions follows a normal distribution and the surface morphology is isotropic.

[0045] S52, Numerical calculation of the mixed lubrication model: The governing equations are spatially discretized using numerical discretization methods, including the finite difference method and the finite element method. Iterative acceleration algorithms are employed to improve solution efficiency, and strict load balance and liquid film pressure convergence criteria are set. Macroscopic lubrication parameters are calculated. These parameters include liquid film thickness, liquid film pressure, liquid film bearing capacity, micro-protrusion bearing capacity, frictional force generated by the liquid film, frictional force generated by the micro-protrusion, total frictional force, coefficient of friction, and frictional loss.

[0046] S53. Verification of numerical calculation results: A classic sliding friction pair example recognized in the field was selected as the benchmark for model verification. Under completely identical geometric parameters, material properties and working conditions, the calculated liquid film pressure field and film thickness field were compared with the benchmark results to verify the calculation accuracy and reliability of the macroscopic model.

[0047] In another exemplary embodiment of this application, step S6 above may be implemented by the following steps: The governing equations of the benchmark-validated hybrid lubrication numerical model are spatially discretized using the finite difference method. A successive over-relaxation iterative algorithm is then employed for solution, with dual convergence criteria: the iterative convergence tolerance of the liquid film pressure field is a first preset value, and the convergence accuracy of the load balance condition is no less than a second preset value. The first preset value can be 10. -6The second preset value can be 10. -4 For each set of operating parameters including temperature, speed, and applied load, the process of calculating liquid film thickness, iteratively solving the pressure field, and checking the load balance is iterated repeatedly until the pressure field convergence criterion and the load balance convergence criterion are simultaneously satisfied, and the corresponding lubrication performance characteristic parameters are output. The lubrication performance characteristic parameters include liquid film thickness, liquid film pressure, liquid film bearing capacity, micro-protrusion bearing capacity, friction force generated by the liquid film, friction force generated by the micro-protrusion, total friction force, friction coefficient, friction loss work, eccentricity, offset angle, maximum liquid film pressure, and minimum liquid film thickness.

[0048] In another exemplary embodiment of this application, the film thickness ratio in step S7 is defined as the ratio of the minimum lubricating medium thickness to the overall surface roughness. When the film thickness ratio is greater than 4, it is determined to be a hydrodynamic lubrication state; otherwise, it is determined to be a mixed lubrication state.

[0049] When establishing the quantitative empirical correlation formula for the critical speed of the temperature-load bivariate system, the critical speed and load are first linearly fitted under each isothermal condition, and then the quantitative empirical correlation formula for the change of critical speed with temperature and applied load is obtained by quadratic polynomial interpolation in the temperature range.

[0050] In step S8 above, when analyzing the cross-scale action mechanism, a cross-scale action chain of "component modification - micro-network reconstruction - viscosity-temperature characteristic regulation - macro-lubrication performance improvement" is established from three dimensions: microstructure relaxation, atomic dynamics constraint and macro-load-bearing performance evolution, and the quantitative transmission relationship between each scale is determined.

[0051] In another exemplary embodiment of this application, steps S6-S7 above mainly perform the cross-scale parameter coupling and global solution process for lubrication performance, specifically including the following steps: Thermophysical parameter transfer in cross-scale computation: The experimentally calibrated microscopic simulated physical property parameters are used as core boundary conditions and substituted into the benchmark-validated macroscopic hybrid lubrication numerical model to achieve a reliable combination of microscopic parameters and macroscopic model.

[0052] Macroscopic lubrication performance calculation: Numerical solutions are performed point-by-point for each operating point in the calculation condition set to obtain the lubrication performance characteristic parameters of the entire flow field of the friction pair, including maximum liquid film pressure, minimum liquid film thickness, friction coefficient, friction force, eccentricity, and offset angle. The influence of operating condition variables such as speed, load, and temperature on lubrication performance is systematically analyzed.

[0053] Determining the lubrication condition: Based on the results of multi-condition solutions, the film thickness ratio is used as the criterion for the state transition between hydrodynamic lubrication and mixed lubrication. The critical speed for the lubrication state transition under different temperatures and loads is determined, and the working condition boundaries between the hydrodynamic lubrication zone and the mixed lubrication zone are delineated.

[0054] For the dual variables of temperature and applied load, a quantitative empirical correlation formula for temperature-load dual variables at critical speed is established by combining isothermal fitting and temperature interval interpolation. This ensures that the fitting accuracy and prediction error are within a reasonable range, enabling rapid and accurate prediction of lubrication state boundaries over a wide range of operating conditions.

[0055] In another exemplary embodiment of this application, step S8 mainly performs the cross-scale interaction mechanism analysis and prediction of system output, specifically including the following steps: Analysis of cross-scale mechanisms of action: Combining the evolution of microstructure with the characteristics of changes in macroscopic lubrication performance, this study systematically analyzes the cross-scale mechanism of lubricant modification from multiple dimensions, including microstructure relaxation, atomic dynamics constraints, and macroscopic load-bearing performance evolution. It establishes a complete cross-scale action chain of "component modification - micro-network reconstruction - viscosity-temperature characteristic regulation - macroscopic lubrication performance improvement" and clarifies the quantitative transmission relationship between each scale.

[0056] Prediction system output: Ultimately, a standardized cross-scale lubrication performance prediction method system was formed, encompassing the entire process of microscopic simulation, parameter calibration, macroscopic solution, boundary prediction, and mechanism analysis, providing systematic methodological support for lubrication material modification design and friction pair system condition matching optimization.

[0057] In simple terms, this embodiment mainly includes the following steps: B1. Construct a cross-scale calculation system and working condition mapping framework: Determine the lubrication system of the lubricating medium to be studied and the target friction pair, and clarify the thermophysical transfer parameters between the microscopic atomic simulation system and the macroscopic continuous medium hybrid lubrication model; Construct a set of calculation working conditions covering multiple dimensions such as temperature, speed, and applied load, and the working condition range covers the entire working condition range of the actual operation of the friction pair.

[0058] B2. Microscopic First-Principles Molecular Dynamics Simulation and Quantification of Physical Properties: A periodic supercell atomic model of the lubricating medium is constructed. In the pure medium system, the atoms of each element are configured according to the target composition ratio. In the doped and modified system, a substitution doping method is used to replace the matrix atoms with dopant atoms while keeping the total number of atoms unchanged. After all initial configurations are melted at high temperature and fully relaxed to an isotropic and homogeneous state, first-principles molecular dynamics based on density functional theory is used to simulate multiple representative temperature points under isothermal and isobaric ensembles. The equilibrium trajectory of thermodynamically stable parameters is intercepted to calculate the microscopic structural characteristic parameters and macroscopic thermophysical transport parameters. The macroscopic transport parameters obtained by simulation are compared and verified with publicly available experimental fitting data to confirm the accuracy and reliability of the calculated physical property parameters.

[0059] B3. Construction and Benchmark Verification of Macroscale Hybrid Lubrication Numerical Model: Based on the fluid lubrication control equations, the contact theory of rough surface micro-protrusions, and the load balance equations, a hybrid lubrication numerical model of the friction pair system is constructed, establishing a complete set of control equations including liquid film pressure, film thickness distribution, contact pressure, friction force, and friction loss. The control equations are solved using numerical discretization methods and iterative acceleration algorithms to obtain macroscopic lubrication performance parameters. A classic sliding friction pair example recognized in the field is selected as a benchmark to verify the computational accuracy and reliability of the macroscopic model.

[0060] B4. Cross-scale parameter coupling and global solution of lubrication performance: The experimentally calibrated microscopic simulated physical property parameters are used as core boundary conditions and substituted into the benchmark-verified macroscopic hybrid lubrication numerical model to achieve coupling between microscopic parameters and macroscopic models; the point-by-point numerical solution of each working point in the calculation working condition set is performed to obtain the lubrication performance characteristic parameters of the friction pair in the entire working condition range; the film thickness ratio is used as the state transition criterion between hydrodynamic lubrication and hybrid lubrication to determine the critical conditions for lubrication state transition under different temperatures and loads, and a quantitative empirical correlation formula for critical speed of temperature-load dual variables is established.

[0061] B5. Cross-scale action mechanism analysis and prediction system output: Combining the evolution law of microstructure and the characteristics of macroscopic lubrication performance changes, the cross-scale action mechanism of lubrication medium modification is analyzed from multiple dimensions and systems. A complete cross-scale action chain from component modification to macroscopic lubrication performance improvement is established, forming a standardized cross-scale lubrication performance prediction method system that includes microscopic simulation, parameter calibration, macroscopic solution, boundary prediction, and mechanism analysis.

[0062] Compared with the prior art, this application has at least the following advantages: 1. It pioneered a cross-scale computational strategy of micro-macro hierarchical decoupling and re-coupling, which decomposes the complex multi-scale lubrication problem into two independent modules: micro-property calculation and macro-performance simulation. Seamless connection is achieved through standardized parameter transfer, breaking through the technical bottleneck that single-scale simulation cannot take into account both the depth of mechanism and the breadth of working conditions.

[0063] 2. The method is highly versatile and scalable, not limited to specific lubricating media and friction pair structures. The microscopic model components, macroscopic model geometry and operating parameters can be flexibly adjusted according to the research object. It is applicable to the modification design and performance evaluation of various liquid lubricating media, greases and semi-solid lubricating materials.

[0064] 3. The entire process forms standardized calculation specifications, data processing methods, and mechanism analysis frameworks. All steps adopt common theories and algorithms in the field, which are highly repeatable and provide a unified methodological paradigm for cross-scale tribology research and lubrication performance prediction.

[0065] In another exemplary embodiment of this application, two specific embodiments are provided. This application provides a method for predicting cross-scale lubrication performance by combining first-principles molecular dynamics with a continuous medium hybrid lubrication model, such as... Figure 3 As shown, this method employs a technical approach of "microscopic-macroscopic hierarchical decoupling calculation → dual calibration verification → cross-scale combined prediction → mechanism system analysis." Through hierarchical decoupling and precise coupling of high-precision microscopic property calculations and macroscopic lubrication performance simulations, it achieves cross-scale quantitative prediction and mechanism revelation of the lubrication performance of friction pairs, from the atomic-level microstructure to the macroscopic lubrication performance. The method is standardized throughout its entire process, highly reproducible, and can be widely applied to the modification design of various lubricating media and the performance optimization of friction pair systems. The specific implementation steps are as follows: Example 1: Cross-scale performance prediction of GaInSn liquid metal lubricated sliding bearings.

[0066] This embodiment uses eutectic GaInSn liquid metal as the lubrication medium and a radial sliding bearing as the target friction pair. The cross-scale lubrication performance prediction method described in this application is employed for the entire calculation process. All parameter settings and calculation procedures strictly follow the framework of the method described in this application. The specific implementation process is as follows: Step 1: Construct a cross-scale computing system and a working condition mapping framework.

[0067] 1.1 Define the thermophysical transfer parameters.

[0068] The target friction pair is a radial sliding bearing. The core thermophysical transfer parameters between the microscopic first-principles molecular dynamics simulation system and the macroscopic continuous medium hybrid lubrication model are defined as three items: the dynamic viscosity, density, and kinematic viscosity of the lubricating medium. The dynamic viscosity is directly used as the fluid property input parameter for the macroscopic average Reynolds equation; the density is used for calculating the liquid film inertial effect and converting it to kinematic viscosity; and the kinematic viscosity is used for experimental verification of the microscopic calculation results.

[0069] 1.2 Define the calculation condition set.

[0070] Based on the typical service conditions of liquid metal bearings, a complete set of calculation conditions covering three dimensions—temperature, speed, and load—is constructed: Temperature dimension: Three representative temperature points, 300K, 330K and 360K, are selected to cover the service range from room temperature to high temperature, corresponding to the common operating temperature range above the melting point of the lubricating medium; Speed ​​dimension: set to 100~3000r / min, covering the full speed range from low-speed start-up, rated operation to high-speed conditions, including the critical speed range of the transition from mixed lubrication to hydrodynamic lubrication; Load dimension: set to 50~1500N, covering the full load range from light load, rated load to heavy load instability, including the critical load range of the transition from hydrodynamic lubrication to hybrid lubrication.

[0071] All operating points comprehensively cover typical and extreme operating conditions of the actual operation of the friction pair, ensuring that the calculation results are applicable to engineering.

[0072] Step 2: First-principles molecular dynamics simulation and quantification of physical property parameters at the microscale.

[0073] 2.1 Constructing a microscopic geometric model.

[0074] like Figure 4 As shown in 'a', a cubic supercell atomic model under periodic boundary conditions is constructed, with a total of 216 atoms per unit cell, balancing statistical representativeness and first-principles calculation efficiency. For the pure GaInSn system in this embodiment, the atoms of each element are configured according to the eutectic composition ratio: 140 Ga atoms, 50 In atoms, and 26 Sn atoms.

[0075] All initial configurations are first subjected to high-temperature melting treatment, and then gradually annealed and cooled to the target temperatures. Sufficient structural relaxation is carried out at the target temperatures to completely eliminate the artificial atomic order of the initial configurations. The thermodynamic parameters such as potential energy, temperature, and pressure of the system are monitored in real time. After the system relaxes to an isotropic homogeneous liquid state, the equilibrium trajectory is intercepted for subsequent statistical calculations of physical properties.

[0076] 2.2 First-principles molecular dynamics simulation.

[0077] The simulation was conducted using first-principles molecular dynamics based on density functional theory, with the core calculation parameters set as follows: Electron-ion interactions are described using the projected fused wave method; the exchange correlation energy is treated using a PBE functional under the generalized gradient approximation; the plane wave energy cutoff is set to 400 eV; and the electronic self-consistent convergence criterion is set to 1 × 10⁻⁶ eV. -5 eV; The electron occupation broadening adopts the first-order Methfessel-Paxton scheme, with a broadening width of 0.2 eV; Due to the large size of the supercell and the lack of long-range order in liquid metal, Brillouin zone sampling uses a single Γ point to control the computational load while ensuring the convergence of total energy and stress tensor.

[0078] All molecular dynamics simulations were conducted under isothermal-isobaric (NPT) ensembles, with the system temperature controlled using a Langevin thermostat. The molecular dynamics integration time step was set to 1.0 fs. Independent simulations were performed at three temperature points: 300 K, 330 K, and 360 K. The total simulation time for each trajectory was 100 ps, ​​and the simulations included 10... 5 One integration step. During the simulation, the first 70 ps of the structural relaxation transient segment is discarded, and only the last 30 ps of the equilibrium segment with stable thermodynamic parameters is retained for property calculation. A total of 30,000 frames of sampling are used in the equilibrium segment to ensure the reliability of the statistical results.

[0079] 2.3 Calculation Results Statistics and Data Processing.

[0080] Based on the equilibrium simulation trajectory, the microstructural characteristic parameters and macroscopic thermophysical transport parameters are calculated respectively.

[0081] Microstructural parameters include the total radial distribution function, the partial radial distribution function, the coordination number, and the diffusion coefficient. The specific calculation methods are as follows: The radial distribution function is calculated using formula (1) and obtained by statistically analyzing the probability distribution of the spacing between atomic pairs within the system. This function is used to characterize the probability density of other atoms appearing at a distance r from the reference atom, reflecting the short-range order of the system.

[0082] in, V For supercell volume, N atom The number of atoms is represented by <·>, which indicates the ensemble average of the equilibrium trajectory over 30 ps. d (·) represents Dirac d function, r i , r j For the first i , No.j The spatial position vector of an atom.

[0083] The coordination number is calculated using formula (2a), which is obtained by integrating the partial radial distribution function of the corresponding atom pair within the first nearest neighbor peak interval, thus quantitatively describing the number of specific types of nearest neighbor atoms around the central atom.

[0084] in, r cut Let be the starting and ending radii of the first peak of the radial distribution function. r Atomic number density, g ( r ) is the radial distribution function.

[0085] Based on Einstein's diffusion relations, the atomic diffusion coefficient is obtained by long-term linear fitting of the mean square displacement curves of atoms of each element. D , is used to characterize the long-range migration ability of atoms, and the specific calculation method is shown in formulas (3a) and (3b).

[0086] Among them, MS( t )express t Mean square displacement of atoms at time t. r ( t )for t Atom position vector at time, r (0) is the atomic position vector at time 0.

[0087] Macroscopic thermophysical transport parameters include the equilibrium density and dynamic viscosity of the system, and the specific calculation methods are as follows: Based on the Green-Kubo linear response theory, by analyzing the off-diagonal stress tensor components... t Calculate dynamic viscosity by time integration of the autocorrelation function. m The specific calculation method is shown in formula (4).

[0088] in, k B Boltzmann's constant, T The system's thermodynamic temperature is... This represents the off-diagonal stress tensor components in the xy direction at time t=0. This represents the off-diagonal stress tensor components in the xy direction at time t.

[0089] The equilibrium density is obtained by statistically averaging the volume of the system in equilibrium state and the total atomic mass under the NPT ensemble, as shown in formula (2b).

[0090] To reduce the impact of long-term statistical noise, the 30 ps equilibrium trajectory was divided into 5 independent sub-blocks. Bootstrap sampling was used to calculate the 95% confidence interval, and the viscosity was quantified to calculate the uncertainty.

[0091] The core parameter results calculated in this embodiment are as follows: Microstructure: The first nearest neighbor characteristic peak of the total radial distribution function at 360 K is located at 3.00 Å with a peak intensity of 3.01; the total coordination number is approximately 12.83; the atomic diffusion coefficients of Ga, In, and Sn are distributed within the range of 4.4 × 10⁻⁶. -9 ~4.7×10 -9 m 2 / s range; the activation energy of the system for viscous flow is 2.3 kJ / mol.

[0092] Macroscopic properties: Dynamic viscosity at 300 K is 2.71 × 10⁻⁶. -3 Pa·s, density 6.34 g / cm³ 3 The dynamic viscosity at 330 K is 2.54 × 10⁻⁶. -3 Pa·s, density 6.30 g / cm³ 3 The dynamic viscosity at 360 K is 2.31 × 10⁻⁶. -3 Pa·s, density 6.25 g / cm³ 3 Within the 300–360 K range, the dynamic viscosity decreased by approximately 14.8%, and the density decreased by approximately 1.42%.

[0093] 2.4 Verification of calculation results.

[0094] The simulated dynamic viscosity and equilibrium density were converted into kinematic viscosity using formulas, and then compared with experimental fitting data from published literature at the same temperature points to quantify the relative error at each temperature. At 300 K, the simulated kinematic viscosity is 0.427 mm. 2 / s, experimental fitted value is 0.477 mm 2 / s, with a relative error of 10.3%; At 330 K, the simulated value is 0.419 mm. 2 / s, experimental value is 0.415 mm 2 / s, with a relative error of less than 1%; At 360 K, the simulated value is 0.369 mm. 2 / s, experimental value is 0.361 mm 2 / s, with a relative error of less than 3%.

[0095] The maximum relative error across the entire temperature range is 10.3%, and the errors at medium and high temperature points are all controlled within 3%, which is within an acceptable range. This indicates that the accuracy of the microscopic simulation method meets the requirements, and its output physical property parameters can be used as reliable inputs for the macroscopic model.

[0096] Step 3: Construction and benchmark verification of a macroscopic-scale hybrid lubrication numerical model.

[0097] 3.1 Establishment of a macroscopic-scale hybrid lubrication model.

[0098] Based on the fluid lubrication control equations and the contact theory of micro-protrusions on rough surfaces, a hybrid lubrication numerical model for radial sliding bearings is constructed, establishing a complete set of control equations, specifically including: the average Reynolds equation considering the surface roughness effect, the liquid film thickness geometric equation, the Greenwood-Tripp micro-protrusion contact pressure equation, the load balance equation, and the total friction calculation equation. The model introduces a flow factor and shear stress factor related to surface roughness to characterize the influence of rough surfaces on fluid flow and shear. It assumes that the height of the surface micro-protrusions follows a zero-mean Gaussian normal distribution and that the surface morphology is isotropic. The total external load is borne jointly by the liquid film dynamic pressure bearing capacity and the micro-protrusion contact bearing capacity. The total friction force consists of two parts: liquid viscous shear friction and micro-protrusion contact friction, as detailed in the following formulas.

[0099] The core governing equation for hydrodynamic pressure is the Patir-Cheng rough surface average Reynolds equation, as shown in equation (5).

[0100] in, p lm For the pressure of the lubricating medium, h For the thickness of the lubricating medium, U The relative velocity of the friction surfaces. s To comprehensively assess surface roughness, x and z Here are the azimuth coordinates, and t is the time. f x , f z , f c and f s This represents the flow factor.

[0101] Based on the geometric positional relationship, the geometric equation for the thickness of the radial bearing liquid film can be obtained, as shown in formula (6).

[0102] in, c For bearing clearance, e For eccentricity, It is the offset angle. i The bearing axial angle is where e and c The ratio is the bearing's eccentricity.

[0103] Micro-protrusion contact pressure p asp Determined by the Greenwood-Tripp equation, specifically:

[0104] in, E To measure the overall elastic modulus, E 1. n 1. E 2 and n 2 represents the elastic modulus and Poisson's ratio of the journal and bearing materials, respectively. A c This represents the contact area of ​​the rough peak. A 0 represents the nominal contact area. F ( l ) indicates different l The distribution function of the ratio, l The film thickness ratio is the ratio of the minimum lubricating medium thickness to the overall surface roughness. F 2.5 ( l The expression denoted by ) represents the statistical function of the micro-convexity load obtained by the 2.5th order integral of the Gaussian distribution of the micro-convexity height. F 2.0 ( l Let represent the statistical function of the contact area of ​​the micro-protrusion obtained by the 2.0th order integral of the Gaussian distribution of the micro-protrusion height. When l >4 is defined as hydrodynamic lubrication; otherwise, it is mixed lubrication. K Represents the elastic coefficient. or For micro-convexity density, β Let be the radius of curvature of the micro-convex body.

[0105] To ensure the integrity of the governing equations during the numerical solution process, a load balance equation is introduced. This equation shows that the total load-bearing capacity generated by the liquid metal film and surface roughness is related to the external load. P Phase equilibrium, as shown in equation (8):

[0106] in, P lm The load generated by the liquid metal film, P asp The load generated by the surface micro-protrusions.

[0107] The total friction between the journal and the bearing bushing is caused by the viscous friction of the liquid metal.f lm Friction force in contact with rough surfaces f asp The composition is as follows:

[0108] in, f f , f fs , f fp It is the average shear flow factor. m f The coefficient of friction, t 0为材料 The ultimate shear stress.

[0109] The bearing geometry and material parameters set in this embodiment are as follows: bearing length 38.000 mm, bearing radius 38.000 mm, bearing radial clearance 30.000 μm, overall surface roughness 1.610 μm, and overall elastic modulus 183.800 GPa.

[0110] 3.2 Numerical calculation of the mixed lubrication model.

[0111] The finite difference method is used to discretize the governing equations across the entire domain, and a successive over-relaxation iterative algorithm is employed to improve solution efficiency. A dual convergence criterion is set: the iterative convergence tolerance for the liquid film pressure field is 1 × 10⁻⁶. -6 The convergence accuracy of the load balance condition is not less than 1×10 -4 The specific steps are as follows: (1) First, the initial eccentricity and offset angle are given as 0.50 and 0.00° respectively. The global liquid film thickness distribution is calculated by formula (6).

[0112] (2) The first-order central difference method is used to discretize the entire domain of formula (5) in the x and z directions, where there are 360 ​​nodes in the x direction and 80 nodes in the z direction, and the discretized partial differential equation is transformed into a system of linear algebraic equations.

[0113] (3) The liquid film pressure field is solved by the successive over-relaxation (SOR) iterative algorithm until the pressure field residual satisfies the convergence tolerance of less than 1×10. -6 At that time, a convergent liquid film pressure field is obtained.

[0114] (4) Substitute the global liquid film thickness distribution calculated in step (1) into formula (7a) to calculate the micro-convex body contact pressure field.

[0115] (5) Substitute the liquid film pressure field obtained in step (3) and the micro-protrusion contact pressure field obtained in step (4) into formula (8) to calculate the bearing load.

[0116] (6) Compare the bearing capacity calculated in step (5) with the external load. The relative error between the two is less than 1×10. -4 If the calculation is successful, a convergent result will be obtained; otherwise, the eccentricity and offset angle will be adjusted and the calculation will be recalculated until a convergent result is obtained.

[0117] Complete flow field lubrication performance parameters are obtained through iterative solutions, including liquid film thickness field, liquid film pressure field, liquid film bearing capacity, micro-protrusion bearing capacity, liquid film viscous friction, micro-protrusion contact friction, total friction, friction coefficient, friction loss work, eccentricity, and offset angle.

[0118] 3.3 Verification of numerical calculation results.

[0119] A classic radial sliding bearing example, widely accepted in the field of tribology, was selected as the benchmark example. Under the same geometric parameters, material properties, and operating conditions as the benchmark literature, the model constructed in this application was used for calculation. The comparison results are as follows: At a speed of 160 r / min, the calculated peak liquid film pressure was 1.160 MPa, which was completely consistent with the literature reference value, with a relative error of 0%; the minimum liquid film thickness was 79.588 μm, while the literature reference value was 80.000 μm, with a relative error of 0.515%. At 190 r / min, the calculated peak liquid film pressure was 1.084 MPa, compared to a reference value of 1.090 MPa, with a relative error of 0.370%; the minimum liquid film thickness was 93.573 μm, compared to a reference value of 94.000 μm, with a relative error of 0.454%.

[0120] The relative errors of the two core indicators are both less than 1%, and the overall distribution of pressure and film thickness is highly consistent with the benchmark results, indicating that the calculation accuracy and reliability of the macroscopic mixed lubrication model meet the requirements of engineering applications.

[0121] Step 4: Cross-scale parameter coupling and global solution of lubrication performance.

[0122] 4.1 Transfer of thermophysical parameters in cross-scale calculations.

[0123] The dynamic viscosity and density parameters of GaInSn at 300 K, 330 K, and 360 K, which have been experimentally calibrated and verified by microscopic simulation, are substituted as core input boundary conditions into the benchmark-verified macroscopic mixed lubrication numerical model, replacing the original empirical viscosity parameters in the model. This achieves reliable coupling between the microscopic atomic-scale calculation results and the macroscopic continuous medium model, ensuring that the macroscopic performance prediction is supported by microscopic mechanisms.

[0124] 4.2 Calculation of macroscopic lubrication performance.

[0125] For the preset multi-dimensional calculation condition set, numerical solutions are performed for each condition point one by one, specifically using an iterative approach: for each set of temperature, speed, and applied load parameters, the calibrated microscopic simulated viscosity and density parameters at the corresponding temperature are first substituted into the macroscopic mixed lubrication model. Then, following the process in step 3.2, the numerical flow of "liquid film thickness calculation → pressure field iterative solution → load balance check → journal position correction" is iterated repeatedly until both the pressure field convergence criterion and the load balance convergence criterion are satisfied. Finally, the complete set of lubrication performance characteristic parameters under this condition is output. After completing the calculation for all condition points, the lubrication performance characteristic parameters of the friction pair are obtained across the entire operating range. Taking a typical operating condition of 300 K ambient temperature and 100 N applied load as an example, the core performance results are as follows: At a low speed of 100 r / min, the minimum liquid film thickness is 5.936 μm, the maximum liquid film pressure is 0.120 MPa, the total friction force is 10.018 N, the friction coefficient is 0.100, the eccentricity is 0.802, and the offset angle is 70.19°. At this time, the bearing is in a mixed lubrication state and there is obvious micro-protrusion contact. When the rotational speed increases to 200 r / min, the maximum liquid film pressure reaches a peak of 0.209 MPa, the total frictional force drops to 0.277 N, the coefficient of friction drops to 0.00277, and the lubrication state changes significantly. At a high speed of 3000 r / min, the minimum liquid film thickness increased to 25.130 μm, the eccentricity decreased to 0.162, the offset angle decreased to 10.31°, the journal was close to the center position, and the bearing operation stability was significantly improved.

[0126] Regarding load characteristics, at 300 K and 1000 r / min, as the applied load increases from 50 N to 1500 N: The minimum liquid film thickness decreased monotonically from 22.989 μm to 4.880 μm, and the maximum liquid film pressure increased from 0.066 MPa to 1.533 MPa. Under a load of 500 N, the total frictional force is 1.385 N and the coefficient of friction is 0.0028; under a heavy load of 1500 N, the total frictional force increases to 178.874 N and the coefficient of friction increases to 0.119, resulting in a sharp increase in frictional loss.

[0127] 4.3 Establishing the correlation between the determination of lubrication status and experience.

[0128] The film thickness ratio λ=4 (corresponding to 4 times the minimum liquid film thickness threshold of 6.440 μm) is used as the criterion for the state transition between hydrodynamic lubrication and mixed lubrication. When the minimum liquid film thickness is greater than or equal to this threshold, the friction pair is determined to enter the state of complete hydrodynamic lubrication, and the friction interface is completely isolated by the liquid film; otherwise, it is in the state of mixed lubrication, and there is micro-protrusion contact.

[0129] Based on the full-condition calculation results, the critical condition boundary is determined: Under loads of 300 K and 100 N, the critical speed for hydrodynamic lubrication is 183 r / min. When the speed exceeds this value, the bearing enters a state of complete hydrodynamic lubrication. At 300 K and 1000 r / min, the critical load for hydrodynamic lubrication is 557 N. After the load exceeds this value, the bearing enters a mixed lubrication state.

[0130] For the dual variables of temperature and applied load, a linear fit was first performed on the critical speed and load under each isothermal condition. Then, a quantitative empirical correlation formula for the temperature-load dual variables of the critical speed was established by using quadratic polynomial interpolation over the temperature range. The determination coefficients of the isothermal linear fits were all higher than 0.99. The maximum relative error between the predicted value and the numerical calculation value of the empirical formula was 1.43% over the entire operating range, occurring at the 300 K and 900 N operating points, meeting the accuracy requirements for rapid and accurate prediction of lubrication state boundaries over a wide operating range.

[0131] Step 5: Analysis of cross-scale action mechanism and prediction system output.

[0132] 5.1 Analysis of cross-scale action mechanism.

[0133] Combining the structural evolution laws and atomic dynamics behavior at the microscale with the characteristics of lubrication performance changes at the macroscale, the lubrication performance mechanism of the pure GaInSn system is systematically analyzed from three dimensions: At the microstructure level: pure GaInSn forms a short-range ordered liquid atomic network with Ga atoms as the framework. The atomic packing density is moderate, and the short-range order decreases monotonically with increasing temperature. At the atomic dynamics level: the diffusion rate of atoms of each element is relatively fast, the activation energy of viscosity flow is only 2.3 kJ / mol, the energy barrier for relative slip between atomic layers is low, so the intrinsic viscosity is low, and the viscosity decreases monotonically with increasing temperature. At the macroscopic performance level: the low intrinsic viscosity results in limited hydrodynamic load-bearing capacity of the fluid. Under low-speed and heavy-load conditions, the liquid film thickness is insufficient, which can easily lead to liquid film rupture and mixed lubrication state, resulting in increased micro-protrusion contact and a significant increase in friction and wear.

[0134] This establishes a complete cross-scale interaction chain of "component structure - atomic dynamics - viscosity-temperature characteristics - macroscopic lubrication performance", clarifying the quantitative transmission relationship between physical quantities at each scale.

[0135] 5.2 Prediction system output.

[0136] Ultimately, a standardized cross-scale lubrication performance prediction system is formed, encompassing the entire process of microscopic simulation, parameter calibration, macroscopic solution, boundary prediction, and mechanism analysis. The outputs include: a full-condition lubrication performance parameter database, lubrication state zoning diagrams at three temperatures, empirical formulas for temperature-load dual-variable critical speeds, and conclusions on microscopic action mechanism analysis. This provides quantitative theoretical basis and data support for the condition matching design, performance optimization, and failure prevention of GaInSn liquid metal lubricated bearings.

[0137] This embodiment discloses a cross-scale lubrication performance prediction method combining first-principles molecular dynamics and a continuous medium hybrid lubrication model, belonging to the field of lubrication material performance calculation and tribological numerical simulation technology. Addressing the problems of disconnect between microscopic simulation and macroscopic performance prediction, lack of standardized procedures for cross-scale calculations, and insufficient prediction accuracy and reliability in existing technologies, it adopts a technical path of "microscopic-macroscopic hierarchical decoupled calculation → dual calibration verification → cross-scale combined prediction → mechanism system analysis": constructing a multi-dimensional set of calculation conditions, obtaining the microscopic structural characteristics and thermophysical properties of the lubricating medium through first-principles molecular dynamics simulation, coupling the property parameters to a benchmark-verified macroscopic hybrid lubrication numerical model after experimental calibration and verification, solving the global lubrication performance for each condition, establishing a critical empirical correlation between temperature and load bivariate lubrication states, and finally systematically analyzing the cross-scale mechanism of lubricating medium modification. This application achieves high-precision cross-scale lubrication performance prediction from the atomic scale to the macroscopic friction pair. The method is standardized and highly versatile, and can be widely applied to the design of various lubricating medium modifications and the performance evaluation and optimization of friction pair systems, providing theoretical and methodological support for the design of high-end equipment lubrication systems.

[0138] Example 2: Cross-scale performance prediction of Cu-doped GaInSn liquid metal lubricated radial sliding bearing.

[0139] This embodiment uses Cu-doped modified GaInSn liquid metal as the lubrication medium and a radial sliding bearing as the target friction pair. The cross-scale lubrication performance prediction method described in this application is used for full-process calculations, and the results are compared with the pure GaInSn system in Example 1. All calculation schemes and accuracy standards are consistent with those in Example 1 to ensure the reliability of the comparison results. The specific implementation process is as follows: Step 1: Construct a cross-scale computing system and a working condition mapping framework 1.1. Define the thermophysical transfer parameters.

[0140] The target friction pair is a radial sliding bearing. The core thermophysical transfer parameters between the microscopic first-principles molecular dynamics simulation system and the macroscopic continuous medium hybrid lubrication model are defined as three items: the dynamic viscosity, density, and kinematic viscosity of the Cu-doped GaInSn lubricating medium. The dynamic viscosity is directly used as the fluid property input parameter for the macroscopic average Reynolds equation; the density is used for calculating the liquid film inertial effect and converting it to kinematic viscosity; and the kinematic viscosity is used for reliability benchmarking of the microscopic calculation results.

[0141] 1.2. Define the calculation condition set.

[0142] Maintaining the same operating range as Example 1, a three-dimensional calculation set of temperature, speed, and load was constructed to ensure a unified benchmark for performance comparison between the two systems: Temperature dimension: Three representative temperature points, 300 K, 330 K and 360 K, were selected to cover the service range from room temperature to medium temperature. Speed ​​range: set to 100~3000 r / min, covering the entire range from low-speed start-up to high-speed stable operation; Load range: set to 50~1500 N, covering all working conditions from light load to heavy load instability.

[0143] Step 2: First-principles molecular dynamics simulation and quantification of physical property parameters at the microscale.

[0144] 2.1 Constructing a microscopic geometric model.

[0145] like Figure 4 As shown in b, a cubic supercell atomic model under periodic boundary conditions was constructed, with the total number of atoms in each cell remaining at 216, consistent with the total number of atoms in the pure system of Example 1. For the Cu-doped system in this example, a substitution doping method was adopted, replacing one Ga atom with one Cu atom. The final atomic configuration was: 139 Ga atoms, 50 In atoms, 26 Sn atoms, and 1 Cu atom, with a doping ratio of 0.46 at%. This doping method avoids local structural distortion and internal stress introduced by interstitial atoms, ensuring the stability of the local atomic bonding environment.

[0146] All initial configurations are first subjected to high-temperature melting treatment, and then gradually annealed and cooled to the target temperatures. Sufficient structural relaxation is carried out at the target temperatures to completely eliminate the artificial atomic order of the initial configurations. The thermodynamic parameters such as potential energy, temperature, and pressure of the system are monitored in real time. After the system relaxes to an isotropic homogeneous liquid state, the equilibrium trajectory is intercepted for subsequent statistical calculations of physical properties.

[0147] 2.2 First-principles molecular dynamics simulation.

[0148] The calculation scheme and parameter settings are exactly the same as in Example 1 to ensure calculation accuracy and fairness of comparison: The simulation was conducted using first-principles molecular dynamics based on density functional theory, with the core calculation parameters set as follows: Electron-ion interactions are described using the projected fused wave method; the exchange correlation energy is treated using a PBE functional under the generalized gradient approximation; the plane wave energy cutoff is set to 400 eV; and the electronic self-consistent convergence criterion is set to 1 × 10⁻⁶ eV. -5 eV; the electron occupation broadening adopts the first-order Methfessel-Paxton scheme with a broadening width of 0.2 eV; the Brillouin zone sampling adopts a single Γ point to balance the convergence of energy and stress tensors and computational efficiency.

[0149] All molecular dynamics simulations were conducted under isothermal-isobaric (NPT) ensembles, with the system temperature controlled using a Langevin thermostat. The molecular dynamics integration time step was set to 1.0 fs. Independent simulations were performed at three temperature points: 300 K, 330 K, and 360 K. The total simulation time for each trajectory was 100 ps, ​​and the simulations included 10... 5 One integration step. During the simulation, the first 70 ps of the structural relaxation transient segment is discarded, and only the equilibrium segment trajectory with stable thermodynamic parameters in the last 30 ps is retained for property calculation. The equilibrium segment has a total of 30,000 frames of sampling.

[0150] 2.3 Calculation Results Statistics and Data Processing.

[0151] Based on the equilibrium-state simulated trajectory, the microscopic structural characteristic parameters and macroscopic thermophysical transport parameters are calculated respectively: Microstructure parameters include the total radial distribution function, the partial radial distribution function, the coordination number of each atom pair, the mean square displacement of each element atom, and the diffusion coefficient. Macroscopic thermophysical transport parameters include the equilibrium density and dynamic viscosity of the system. The dynamic viscosity is calculated based on the Green-Kubo linear response theory by integrating the autocorrelation function of the off-diagonal stress tensor components. Statistical noise is handled using the same truncation scheme and bootstrap resampling method as in Example 1.

[0152] The core parameter results calculated in this embodiment are as follows: Microstructure: At 360 K, the first nearest neighbor characteristic peak of the total radial distribution function remains around 3.00 Å, with the peak intensity decreasing to 2.04, a decrease of approximately 32% compared to the pure system; the total coordination number is 0.13–0.29 lower than the pure system, indicating a slight decrease in atomic packing density; the diffusion rate of Cu atoms is significantly lower than that of the matrix element, with a Cu atomic diffusion coefficient of 1.71 × 10⁻⁶ at 300 K. -9 m 2 / s, increasing to 2.44×10 at 360 K. -9 m 2 / s, which is only 30%~50% of the diffusion rate of Ga, In and Sn elements; the viscous flow activation energy of the system is increased to 3.3kJ / mol.

[0153] Macroscopic physical properties: Dynamic viscosity at 300 K is 3.30 × 10⁻⁶. -3 Pa·s, density 6.25 g / cm³ 3 The dynamic viscosity at 330 K is 2.96 × 10⁻⁶. -3 Pa·s, density 6.21 g / cm³ 3 The dynamic viscosity at 360 K is 2.66 × 10⁻⁶. -3 Pa·s, density 6.16 g / cm³ 3 Within the 300~360 K range, the dynamic viscosity decreased by approximately 19.4%, and the density decreased by approximately 1.44%, indicating that the viscosity is more temperature-sensitive than that of the pure GaInSn system.

[0154] 2.4 Verification of calculation results.

[0155] This embodiment uses the same calculation method and statistical scheme as Example 1, and the reliability of the method has been verified by experimental data for the pure system. The density and viscosity of the Cu-doped system change with temperature in a highly consistent manner with the pure system, and the coefficient of thermal expansion is basically the same as that of the pure system, which conforms to the basic physical laws of alloy doping modification. At the same time, the viscosity change trend is consistent with the modification laws reported in previous experiments, indicating that the microscopic simulation results have sufficient accuracy and can be used as reliable input parameters for macroscopic models.

[0156] Step 3: Construction and benchmark verification of a macroscopic-scale hybrid lubrication numerical model.

[0157] 3.1 Establishment of a macroscopic-scale hybrid lubrication model.

[0158] The governing equations and theoretical assumptions of the macroscopic mixed lubrication model are completely consistent with those of Example 1, including the average Reynolds equation, the liquid film thickness geometric equation, the Greenwood-Tripp micro-protrusion contact pressure equation, the load balance equation, and the total friction calculation equation. The same surface roughness flow factor and shear stress factor are introduced, and it is assumed that the surface micro-protrusions follow a zero-mean Gaussian normal distribution and that the surface morphology is isotropic.

[0159] The bearing geometry and material parameters in this embodiment are completely consistent with those in Embodiment 1: bearing length 38.000 mm, bearing radius 38.000 mm, bearing radial clearance 30.000 μm, overall surface roughness 1.610 μm, and overall elastic modulus 183.800 GPa.

[0160] 3.2 Numerical calculation of the mixed lubrication model.

[0161] The same numerical discretization and iterative solution scheme as in Example 1 is adopted: global spatial discretization using the finite difference method, and solution by successive over-relaxation iterative algorithm; the convergence criterion remains consistent: the convergence tolerance of the liquid film pressure iteration is 1×10⁻⁶. -6 The load balance convergence accuracy is not less than 1×10 -4 .

[0162] 3.3 Verification of numerical calculation results.

[0163] The macroscopic model has been verified for accuracy in Example 1 through a classic benchmark calculation. The relative errors between the peak liquid film pressure and the minimum liquid film thickness are both less than 1%, and the calculation accuracy and reliability meet the requirements. This example directly uses the verified model to carry out subsequent calculations.

[0164] Step 4: Cross-scale parameter coupling and global solution of lubrication performance.

[0165] 4.1 Transfer of thermophysical parameters in cross-scale calculations.

[0166] The dynamic viscosity and density parameters of Cu-doped GaInSn at 300 K, 330 K, and 360 K, which have been verified by reliability and obtained from microscopic simulation, are substituted into the macroscopic mixed lubrication numerical model as core input boundary conditions to complete the coupling and transfer of microscopic atomic-scale calculation results with macroscopic continuous medium model.

[0167] 4.2 Calculation of macroscopic lubrication performance.

[0168] For point-by-point numerical solutions of the full operating condition set, taking a typical operating condition of 300 K ambient temperature and 100 N applied load as an example, the core performance results are as follows, and compared with the pure system: At a low speed of 100 r / min, the minimum liquid film thickness is 6.027 μm, which is greater than the 5.936 μm of the pure system; the maximum liquid film pressure is slightly higher than that of the pure system; although it is still in a mixed lubrication state, the thicker liquid film effectively reduces the contact area of ​​micro-protrusions, and the contact friction is significantly reduced.

[0169] The critical speed for hydrodynamic lubrication is 151 r / min, which is 17.5% lower than the 183 r / min of the pure system, indicating that the hydrodynamic lubrication range extends towards lower speeds.

[0170] At a high speed of 3000 r / min, the minimum liquid film thickness increased to 25.954 μm, and the eccentricity decreased to 0.135. Compared with the pure system, the journal alignment was higher, and the bearing operation stability was better. Due to the increased viscosity, the friction coefficient was slightly higher than that of the pure system under this condition.

[0171] Regarding load characteristics, at 300 K and 1000 r / min, as the applied load increases from 50 N to 1500 N, the minimum liquid film thickness decreases monotonically, while the maximum liquid film pressure increases monotonically. The critical load for hydrodynamic lubrication is approximately 657 N, an increase of about 31% compared to the pure system, significantly enhancing the liquid film load-bearing capacity under heavy load conditions. Entering the mixed lubrication range, the liquid film thickness of the Cu-doped system remains consistently higher than that of the pure system, effectively suppressing the sharp increase in the friction coefficient under heavy loads and reducing the risk of wear.

[0172] 4.3 Establishing the correlation between the determination of lubrication status and experience.

[0173] Using a film thickness ratio λ=4 (corresponding to 4 times the minimum liquid film thickness threshold of 6.440 μm) as the lubrication state transition criterion, the critical operating condition boundary was determined based on the full-condition calculation results: At 300 K and 100 N load, the critical speed for hydrodynamic lubrication is 151 r / min; At 300 K and 1000 r / min, the critical load for hydrodynamic lubrication is 657 N.

[0174] For the dual variables of temperature and applied load, the fitting method consistent with that in Example 1 was adopted. First, a linear fit was performed on the critical speed and load under each isothermal condition. Then, a quantitative empirical correlation formula for the temperature-load dual variables of the critical speed of the Cu-doped system was established through quadratic polynomial interpolation over the temperature range. The determination coefficients of the isothermal linear fits were all higher than 0.99. The maximum relative error between the predicted value and the numerical calculation value of the empirical formula was 2.91% over the entire operating range, occurring at the 360 ​​K and 50 N operating points, which meets the accuracy requirements for rapid prediction over a wide range of operating conditions.

[0175] Step 5: Analysis of cross-scale action mechanism and prediction system output.

[0176] 5.1 Analysis of cross-scale action mechanism.

[0177] By combining microstructure evolution, atomic dynamics behavior, and macroscopic lubrication performance changes, the cross-scale mechanism of Cu doping modification is systematically analyzed from three dimensions: At the microscopic level: trace amounts of Cu atoms enter the Ga-based atomic network by substitution, without causing local structural distortion, but reconstructing the original atomic network, resulting in a slight decrease in coordination number, slight structural relaxation of the system, and a slight decrease in macroscopic density. At the atomic dynamics level: Cu atoms interact strongly with surrounding Ga atoms, resulting in extremely low diffusion rates. As "dynamic anchors," Cu atoms extend the binding time of surrounding matrix atoms, hindering the dynamic rearrangement of the atomic cage structure and increasing the viscous flow activation energy (from 2.3 kJ / mol to 3.3 kJ / mol). The contribution of this dynamic constraint effect is greater than the viscosity reduction effect brought about by structural relaxation, ultimately resulting in a significant increase in viscosity on a macroscopic scale. At the macroscopic performance level: the increase in viscosity directly enhances the hydrodynamic bearing capacity, enabling the bearing to maintain complete liquid film isolation at lower speeds and higher loads, significantly widening the operating range of hydrodynamic lubrication and reducing the risk of friction and wear under low speed and heavy load.

[0178] This establishes a complete cross-scale interaction chain of "Cu substitution doping - microscopic atomic network reconstruction - dynamic anchoring effect - viscosity-temperature characteristic enhancement - macroscopic load-bearing capacity enhancement", clarifying the quantitative transmission relationship of physical quantities at each scale.

[0179] 5.2 Prediction system output.

[0180] The final result establishes a complete cross-scale prediction system for Cu-doped GaInSn lubricated bearings. Outputs include: a database of lubrication performance parameters under all operating conditions, lubrication state zoning diagrams at three temperatures, an empirical formula for the temperature-load bivariate critical speed, and conclusions on the cross-scale action mechanism of Cu doping modification. These findings provide quantitative theoretical basis and data support for the material design, operating condition matching, and performance optimization of doped modified liquid metal lubricated bearings.

[0181] This embodiment uses GaInSn and Cu-doped GaInSn liquid metal lubricated radial sliding bearings as an example. The cross-scale lubrication performance prediction method of this application can be widely applied to the performance evaluation and modification design of various lubricating media and friction pairs in various liquid metal lubrication systems, mineral oil-based lubrication systems, grease-lubricated transmission mechanisms, precision equipment bearings, gear transmission pairs, and friction pairs in extreme environments. Those skilled in the art can flexibly adjust the microscopic supercell configuration, calculation parameters, macroscopic model geometric parameters, and operating boundary conditions in this application according to the medium type, friction pair structure, and actual service conditions of the research object, without departing from the core concept and protection scope of this application.

[0182] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 5As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a cross-scale lubrication performance prediction method.

[0183] Figure 5 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0184] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0185] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0186] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with relevant regulations and be authorized by the owner of the corresponding device.

[0187] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0188] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0189] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0190] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method of cross-scale lubrication performance prediction, characterized in that, include: Determine the lubrication system of the lubricating medium to be predicted and the target friction pair, and construct a set of calculation conditions covering temperature, speed of movement, and applied load; Obtain the thermodynamically stable equilibrium trajectory; the thermodynamically stable equilibrium trajectory is obtained by simulating several representative temperature points in the calculation case set using first-principles molecular dynamics under isothermal and isobaric ensemble conditions; Microstructural characteristic parameters and macroscopic thermophysical transport parameters are calculated based on the equilibrium trajectory. The macroscopic thermophysical transport parameters were compared and verified with publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters. The hybrid lubrication numerical model was benchmarked using a classical sliding friction pair example, resulting in a benchmark-verified macroscopic hybrid lubrication numerical model. The hybrid lubrication numerical model is a hybrid lubrication numerical model of the friction pair lubrication system, which is constructed based on the fluid lubrication control equation, the rough surface micro-protrusion contact theory, and the load balance equation, and includes liquid film pressure, film thickness distribution, contact pressure, friction force, and friction loss. The calibrated macroscopic thermophysical transport parameters are substituted into the benchmark-verified hybrid lubrication numerical model as boundary conditions, and the numerical solutions are obtained point by point for each working point in the calculation working condition set to obtain the lubrication performance characteristic parameters of the friction pair in the entire working condition range. Based on the lubrication performance characteristic parameters, the film thickness ratio is used as the criterion for the state transition between hydrodynamic lubrication and mixed lubrication. The critical conditions for the lubrication state transition under different temperatures and different applied loads are determined, and a quantitative empirical correlation formula for the critical speed of temperature-load dual variables is established. Based on the microstructure characteristic parameters and the lubrication performance characteristic parameters, the cross-scale action mechanism of the modification of the lubricating medium to be predicted is analyzed, and a complete cross-scale action chain from component modification to macroscopic lubrication performance is established.

2. The cross-scale lubrication performance prediction method of claim 1, wherein, The macroscopic thermophysical transport parameters include dynamic viscosity, density, and kinematic viscosity; The dynamic viscosity is used as a fluid property input parameter for the average Reynolds equation in the mixed lubrication numerical model; The density is used for calculating the liquid film inertial effect in the hybrid lubrication numerical model; The kinematic viscosity is used for benchmark conversion in the step of "comparing and verifying the macroscopic thermophysical transport parameters with publicly available experimental fitting data to obtain calibrated macroscopic thermophysical transport parameters".

3. The cross-scale lubrication performance prediction method of claim 1, wherein, The acquisition of the thermodynamically stable equilibrium trajectory specifically includes: A periodic supercell atomic model of the lubricating medium to be predicted is constructed, and the periodic supercell atomic model is subjected to high-temperature melting treatment and structural relaxation treatment. Using a first-principles molecular dynamics method based on density functional theory, the electron-ion interaction mode, exchange correlation energy, plane wave cutoff energy, number of k-points, temperature control method, time step, and total simulation duration were set, and simulations were carried out at several representative temperature points in the calculation case set under isothermal and isobaric ensemble. Discard the initial transient segment and retain the trajectory of the equilibrium segment where thermodynamic parameters are stable.

4. The cross-scale lubrication performance prediction method of claim 3, wherein, When constructing the periodic supercell atomic model, the pure medium system is configured with each element atom according to the target composition ratio; the doped modification system adopts a substitution doping method, replacing the matrix atoms with dopant atoms while keeping the total number of atoms unchanged; After the high-temperature melting treatment and structural relaxation treatment bring the initial configuration to an isotropic and uniform state, the equilibrium trajectory is then extracted for statistical calculation of physical properties. The microstructural characteristic parameters include radial distribution function, coordination number, and atomic diffusion coefficient; the macroscopic thermophysical transport parameters include dynamic viscosity and density; the dynamic viscosity is based on Green-Kubo linear response theory and is obtained by time integration of the autocorrelation function of the off-diagonal stress tensor components.

5. The cross-scale lubrication performance prediction method of claim 1, wherein, The method of benchmarking the hybrid lubrication numerical model using classical sliding friction pair examples to obtain a benchmark-validated macroscopic hybrid lubrication numerical model includes: Macroscopic lubrication performance parameters are obtained by numerically solving the mixed lubrication numerical model; The macroscopic lubrication performance parameters are compared with the benchmark results of the classical sliding friction pair calculation example to verify the calculation accuracy and reliability of the hybrid lubrication numerical model, and a benchmark-verified macroscopic hybrid lubrication numerical model is obtained. The hybrid lubrication numerical model includes the liquid film pressure equation, film thickness distribution equation, micro-protrusion contact pressure equation, load balance equation, and friction force equation. The liquid film pressure equation adopts the Patir-Cheng average Reynolds equation that takes into account the surface roughness effect. The contact pressure equation for the micro-protrusions adopts the Greenwood-Tripp equation and introduces a flow factor and shear stress factor related to surface roughness. It is assumed that the height of the surface micro-protrusions follows a normal distribution and the surface morphology is isotropic.

6. The cross-scale lubrication performance prediction method of claim 1, wherein, The calibrated macroscopic thermophysical transport parameters are substituted as boundary conditions into the benchmark-validated hybrid lubrication numerical model. The numerical solutions are then applied point-by-point to each operating point in the calculation set to obtain the lubrication performance characteristic parameters of the friction pair across the entire operating range. Specifically, these parameters include: The control equations of the benchmark-verified hybrid lubrication numerical model are spatially discretized using the finite difference method, and iteratively solved using a successive over-relaxation iterative algorithm. A dual convergence criterion is set: the iterative convergence tolerance of the liquid film pressure field is a first preset value, and the convergence accuracy of the load balance condition is not lower than a second preset value. For each set of operating parameters including temperature, speed, and applied load, the process of calculating liquid film thickness, iteratively solving the pressure field, and checking the load balance is iterated repeatedly until both the pressure field convergence criterion and the load balance convergence criterion are satisfied. The corresponding lubrication performance characteristic parameters are then output. These lubrication performance characteristic parameters include liquid film thickness, liquid film pressure, liquid film bearing capacity, micro-protrusion bearing capacity, friction force generated by the liquid film, friction force generated by the micro-protrusion, total friction force, friction coefficient, friction loss work, maximum liquid film pressure, and minimum liquid film thickness.

7. The cross-scale lubrication performance prediction method of claim 1, wherein, The film thickness ratio is defined as the ratio of the minimum lubricating medium thickness to the overall surface roughness. When the film thickness ratio is greater than 4, it is determined to be a hydrodynamic lubrication state; otherwise, it is determined to be a mixed lubrication state. When establishing the quantitative empirical correlation formula for the critical speed of the temperature-load bivariate system, the critical speed and load are first linearly fitted under each isothermal condition, and then the quantitative empirical correlation formula for the change of critical speed with temperature and applied load is obtained by polynomial interpolation in the temperature range. When analyzing the cross-scale action mechanism, a cross-scale action chain of "component modification - micro-network reconstruction - viscosity-temperature characteristic regulation - macro-lubrication performance improvement" is established from three dimensions: microstructure relaxation, atomic dynamics constraint, and macro-load-bearing performance evolution, and the quantitative transmission relationship between each scale is determined.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the cross-scale lubrication performance prediction method according to any one of claims 1-7.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, When executed by a processor, the computer program implements the cross-scale lubrication performance prediction method according to any one of claims 1-7.

10. A computer program product comprising a computer program, characterized in that, When executed by a processor, the computer program implements the cross-scale lubrication performance prediction method according to any one of claims 1-7.