Automobile magnesium alloy pre-twin crystal deformation data analysis system

Abstract

The invention provides an automobile magnesium alloy pre-twin crystal deformation data analysis system, and relates to the technical field of data processing, and the system comprises a twin crystal correction module which is provided with a first target monitoring point and a second target monitoring point in a pre-twin crystal area, the first target monitoring point is located at a crystal boundary intersection, and the second target monitoring point is located at the center of a twin crystal zone; acquiring orientation difference angle and strain energy density difference data of the two monitoring points in real time, calculating a covariance matrix of change characteristics of the two monitoring points, and correcting a proliferation rate parameter and an orientation rotation parameter in the dynamic evolution matrix according to a correlation coefficient of the covariance matrix to obtain a corrected dynamic evolution matrix; and the optimization module is used for adjusting a hot extrusion process parameter group according to the twin crystal density gradient distribution output by the corrected dynamic evolution matrix, the parameter group comprises a mold temperature, an extrusion speed and a strain path, and the twin crystal density is dynamically controlled in a target interval. According to the invention, the complex stress state of the automobile part under the actual working condition can be accurately simulated.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a data analysis system for pre-twinning deformation of automotive magnesium alloys. Background Art

[0002] In actual working conditions, automotive components need to withstand multi-axial composite loads such as vertical force, braking force, and lateral force, resulting in differences in the twinning evolution path compared to uniaxial loading. For example, when a wheel hub turns, it simultaneously bears radial load and tangential force. Such a complex stress state will activate multiple twinning systems (such as {10-12} tensile twins and {10-11} compressive twins), leading to uneven distribution of the twinning density gradient. For example, traditional experimental equipment (such as an MTS testing machine) can only apply uniaxial load and cannot reproduce the three-dimensional stress tensor in actual working conditions. For example, a certain research shows that the fatigue life of AZ31 magnesium alloy is shortened by 30% under uniaxial ratcheting-fatigue interaction, and the life loss may be higher under multi-axial loading. For example, existing twinning evolution models (such as the dislocation slip theory based on the Schmid factor) do not consider the microscopic stress differences at the intersection of grain boundaries and the center of twinning bands. For example, a certain university found through in-situ EBSD observation that the local stress concentration near grain boundaries will increase the twinning nucleation probability by more than 50%. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a data analysis system for pre-twinning deformation of automotive magnesium alloys, which can accurately simulate the complex stress state under the actual working conditions of automotive components.

[0004] To solve the above technical problem, the technical solution of the present invention is as follows: A data analysis system for pre-twinning deformation of automotive magnesium alloys, comprising: A multi-axial stress loading module, configured to establish a composite stress field model including vertical force, braking force, and lateral force, apply multi-axial proportional load through a six-axis loading device, and generate a three-dimensional stress tensor data set; A twinning evolution analysis module, configured to calculate the nucleation probability density function of twins according to the three-dimensional stress tensor data set, and construct a dynamic evolution matrix including nucleation energy barrier, proliferation rate, and orientation rotation; A twinning correction module, which sets a first target monitoring point and a second target monitoring point in the pre-twinning region. The first target monitoring point is located at the intersection of grain boundaries, and the second target monitoring point is located at the center of the twinning band. It acquires the orientation difference angle and strain energy density difference data of the two monitoring points in real time, calculates the covariance matrix of the change characteristics of the two monitoring points, and corrects the proliferation rate parameter and orientation rotation parameter in the dynamic evolution matrix according to the correlation coefficient of the covariance matrix to obtain a corrected dynamic evolution matrix; Optimization module, according to the twin density gradient distribution output by the corrected dynamic evolution matrix, adjusts the hot extrusion process parameter set, and the parameter set includes die temperature, extrusion speed, and strain path, and dynamically controls the twin density within the target range; Regulation module, according to the twin orientation distribution data and fatigue strength detection data collected in real time, generates a closed-loop control signal including the grain boundary slip compensation amount and the dynamic recrystallization threshold, wherein the dynamic recrystallization threshold is dynamically adjusted according to the orientation rotation parameter in the corrected dynamic evolution matrix.

[0005] The above solution of the present invention has at least the following beneficial effects: By using a six-axis loading device to construct a composite stress field model including vertical force (Fz), braking force (Fx), and lateral force (Fy), it is possible to simulate the multi-axial proportional load of automotive parts under complex road conditions (such as the coupling state of Fx = 8000N during emergency braking, Fy = 5000N during turning, and vertical load Fz = 12000N). Compared with the traditional uniaxial tensile test (only simulating the stress in the Fz direction), this module can completely capture the cooperative activation effect of the multi-axial stress tensor on the twin system, reducing the prediction error of twin nucleation position from ±50μm to ±15μm. It supports the input of multi-type load spectra such as sine wave and triangular wave, and can simulate periodic variable amplitude loads (such as stress amplitude fluctuations when switching between highway and urban roads). For example, when simulating 100,000 cycles of loading of the wheel hub, the system accurately reproduced the alternating activation phenomenon of {10-12} twins in the actual working condition by adjusting the triaxial load ratio (σx:σy:σz = 4:2:1) in real time.

[0006] Real-time monitor the strain energy density difference (ΔU). When ΔU exceeds 0.6MJ / m3, the system automatically triggers the correction of the proliferation rate parameter, improving the coincidence degree between the calculated value of twin nucleation probability and the in-situ TEM observation result from 68% to 91%. By tracking the dynamic change of the orientation difference angle (θ) (5°~25°), the dynamic response speed of the corrected orientation rotation parameter (α) is increased by 10 times (from 500ms to 50ms), effectively capturing the instantaneous behavior of twin boundary migration.

[0007] Based on the covariance analysis of the data at two monitoring points (such as when the correlation coefficient ρ between θ and ΔU > 0.8), the system can automatically adjust the cross-influence factor in the dynamic evolution matrix (such as the coupling coefficient β between the proliferation rate and the orientation rotation is corrected from 0.3 to 0.7), reducing the simulation error of the twin density gradient distribution from ±12% to ±4%. For example, in the hot extrusion simulation of AZ91 magnesium alloy, the corrected model accurately predicted the twin density difference (Δρ = 15%) between the extrusion direction (ED) and the transverse direction (TD), while the traditional model did not consider this difference.

[0008] When the corrected dynamic evolution matrix predicts that the twin density will exceed the target upper limit (e.g., 30%), the system automatically raises the die temperature from 350°C to 380°C. By activating dynamic recrystallization (refining the grain size from 12 μm to 6 μm), the twin density is reduced by 25%, and at the same time, the elongation is increased from 9% to 13%. For the uneven twin density gradient (e.g., ρ = 40% in the spoke area and ρ = 20% in the rim area), the system automatically adjusts the extrusion speed (from 5 mm / s to 3 mm / s) and superimposes a periodic torsional strain (γ = ±0.02), reducing the gradient difference to ±3% and increasing the fatigue crack initiation life by 80%.

[0009] By monitoring the twin orientation distribution (e.g., when the proportion of {10-12} twins is > 60%), the system generates a grain boundary slip compensation amount (Δγ = 0.005) and adjusts the loading path, reducing the stress concentration factor (Kt) at the grain boundary from 3.5 to 2.3 and decreasing the fatigue crack growth rate (da / dN) by 55%. According to the orientation rotation parameter (α), the recrystallization temperature is dynamically adjusted (e.g., when α = 0.1 rad / s, the threshold is raised from 300°C to 330°C). While ensuring the twin strengthening effect, excessive grain coarsening is avoided (the size is controlled within 8 ± 2 μm), maintaining the tensile strength of the material above 240 MPa, which is 18% higher than the traditional process. Description of the Drawings

[0010] Figure 1 It is a schematic diagram of a pre-twin deformation data analysis system for automotive magnesium alloys provided by an embodiment of the present invention. Detailed Embodiments

[0011] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0012] As Figure 1 shown, an embodiment of the present invention provides a pre-twin deformation data analysis system for automotive magnesium alloys, including: A multi-axial stress loading module for establishing a composite stress field model including vertical force, braking force, and lateral force, applying multi-axial proportional loads through a six-axis loading device, and generating a three-dimensional stress tensor data set; A twin evolution analysis module for calculating the nucleation probability density function of twins based on the three-dimensional stress tensor data set and constructing a dynamic evolution matrix including nucleation energy barrier, proliferation rate, and orientation rotation; Twin correction module, which sets a first target monitoring point and a second target monitoring point in the pre-twin region. The first target monitoring point is located at the intersection of grain boundaries, and the second target monitoring point is located at the center of the twin band. It obtains the orientation difference angle and strain energy density difference data of the two monitoring points in real time, calculates the covariance matrix of the change characteristics of the two monitoring points, and corrects the proliferation rate parameter and orientation rotation parameter in the dynamic evolution matrix according to the correlation coefficient of the covariance matrix to obtain a corrected dynamic evolution matrix; Optimization module, which adjusts the hot extrusion process parameter group according to the twin density gradient distribution output by the corrected dynamic evolution matrix. The parameter group includes die temperature, extrusion speed, and strain path, and dynamically controls the twin density within the target range; Regulation module, which generates a closed-loop control signal including the grain boundary slip compensation amount and the dynamic recrystallization threshold according to the twin orientation distribution data and fatigue strength detection data collected in real time, where the dynamic recrystallization threshold is dynamically adjusted according to the orientation rotation parameter in the corrected dynamic evolution matrix.

[0013] In the embodiment of the present invention, a composite stress field model including a vertical force (Fz), a braking force (Fx), and a lateral force (Fy) is constructed through a six-axis loading device, which can simulate the multi-axis proportional load of automotive parts under complex road conditions (such as the coupling state of Fx = 8000N during emergency braking, Fy = 5000N during turning, and Fz = 12000N for vertical load). Compared with the traditional uniaxial tensile test (only simulating the stress in the Fz direction), this module can completely capture the cooperative activation effect of the multi-axis stress tensor on the twin system, reducing the prediction error of twin nucleation position from ±50μm to ±15μm. It supports the input of multiple types of load spectra such as sine waves and triangular waves, and can simulate periodic variable amplitude loads (such as stress amplitude fluctuations during highway-urban road switching). For example, when simulating 100,000 cycles of loading of the wheel hub, the system accurately reproduces the alternating activation phenomenon of {10-12} twins in the actual working condition by adjusting the triaxial load ratio (σx:σy:σz = 4:2:1) in real time.

[0014] Real-time monitor the strain energy density difference (ΔU). When ΔU exceeds 0.6 MJ / m 3 ³, the system automatically triggers the correction of the proliferation rate parameter, improving the coincidence degree between the calculated value of twin nucleation probability and the in-situ TEM observation result from 68% to 91%. By tracking the dynamic change of the orientation difference angle (θ) (5° to 25°), the dynamic response speed of the corrected orientation rotation parameter (α) is increased by 10 times (from 500 ms to 50 ms), effectively capturing the instantaneous behavior of twin boundary migration.

[0015] Based on the covariance analysis of data from two monitoring points (when the correlation coefficient ρ between θ and ΔU > 0.8), the system can automatically adjust the cross - influence factor in the dynamic evolution matrix (for example, the coupling coefficient β between the growth rate and the orientation rotation is corrected from 0.3 to 0.7), reducing the simulation error of the twin density gradient distribution from ±12% to ±4%. For example, in the hot extrusion simulation of AZ91 magnesium alloy, the corrected model accurately predicted the difference in twin density (Δρ = 15%) between the extrusion direction (ED) and the transverse direction (TD), while the traditional model did not consider this difference.

[0016] When the corrected dynamic evolution matrix predicts that the twin density will exceed the target upper limit (such as 30%), the system automatically raises the die temperature from 350°C to 380°C. By activating dynamic recrystallization (the grain size is refined from 12μm to 6μm), the twin density is reduced by 25%, and at the same time, the elongation is increased from 9% to 13%. For the uneven twin density gradient (such as ρ = 40% in the spoke area and ρ = 20% in the rim area), the system automatically adjusts the extrusion speed (from 5mm / s to 3mm / s) and superimposes a periodic torsional strain (γ = ±0.02), reducing the gradient difference to ±3% and increasing the fatigue crack initiation life by 80%.

[0017] By monitoring the twin orientation distribution (when the proportion of {10 - 12} twins > 60%), the system generates a grain - boundary slip compensation amount (Δγ = 0.005) and adjusts the loading path, reducing the stress concentration coefficient (Kt) at the grain boundary from 3.5 to 2.3 and reducing the fatigue crack growth rate (da / dN) by 55%. According to the orientation rotation parameter (α), the recrystallization temperature is dynamically adjusted (when α = 0.1rad / s, the threshold is raised from 300°C to 330°C). While ensuring the twin strengthening effect, excessive grain coarsening is avoided (the size is controlled within 8 ± 2μm), maintaining the tensile strength of the material above 240MPa, which is 18% higher than the traditional process.

[0018] In a preferred embodiment of the present invention, a composite stress - field model including vertical force, braking force, and lateral force is established. Through a six - axis loading device, multi - axial proportional loads are applied to generate a three - dimensional stress - tensor dataset, including: Based on the mechanical coupling relationship between the vertical force, braking force, and lateral force, mapping rules for stress components in each direction and the load ratio are defined to generate an initial parameter set for the composite stress - field model, specifically including: According to the basic principles of mechanics, the vertical force is mapped to a compressive stress in the vertical direction (acting on the upper and lower surfaces of the material, along the thickness direction), the braking force is mapped to a shear stress in the braking direction (the tangential force along the material movement direction), and the lateral force is mapped to a tensile stress in the lateral direction (the pulling force along the vehicle body transverse direction).

[0019] Analyze the proportional relationship of the three forces in the actual working conditions of the vehicle (such as when driving normally, the vertical force: Braking force: Lateral force ≈ 5:2:1, adjusted to 4:1:3 during turning), determine the initial proportionality coefficients of the stress components in each direction (such as the ratio of the vertical stress σz, braking shear stress τxy, and lateral tensile stress σy).

[0020] Based on the geometric dimensions of automotive components (such as wheel hub diameter, cross-sectional area of the chassis bracket), convert the load force into stress values (stress = load / area). For example, when a vertical force Fz = 10 kN acts on a component with an area A = 0.01 m 2 the initial vertical compressive stress σz = -1000 kPa (the negative sign indicates the compression direction).

[0021] Combined with the stress state analysis in material mechanics, determine the action planes of each stress component (such as the vertical stress acts on the XY plane, and the shear stress acts on the XZ plane), and form an initial stress component list including σx, σy, σz, τxy, τxz, and τyz.

[0022] According to the said initial parameter set, synchronously apply a vertical compressive load, a braking shear load, and a lateral tensile load through a six-axis loading device, and dynamically adjust the loading amplitude and phase difference of each axis based on the preset proportionality coefficients, so that the composite stress field is consistent with the stress distribution of the actual working conditions of the vehicle, specifically including: Through the six-axis loading device, apply loads in three directions to the specimen simultaneously: apply a compressive load in the vertical direction (simulating the road surface support force), apply a shear load in the braking direction (simulating the frictional force during braking), and apply a tensile load in the lateral direction (simulating the centrifugal force during turning); the load form supports static loads (such as steady-state stress at a constant vehicle speed) or dynamic loads (such as periodic loads in the form of sine waves and square waves, simulating the vibration stress of a bumpy road surface).

[0023] According to the actual working condition data (such as the real-time load ratio collected by in-vehicle sensors), dynamically modify the loading amplitude of each axis. For example, when simulating an emergency brake, increase the amplitude of the shear load corresponding to the braking force to 80% of the vertical load, and at the same time reduce the lateral load to 10% of the vertical load. For dynamic loads, set the phase difference of the loading of each axis changing with time (such as the waveform phase difference between the vertical load and the braking load is π / 2, simulating the time sequence of the vertical impact first and then the tangential frictional force during braking), so that the change time sequence of the composite stress field is consistent with the actual motion state of the vehicle.

[0024] Real-time collect the spatial stress distribution data output by the six-axis loading device, and fuse the stress components in each direction through tensor superposition operation to generate a three-dimensional stress tensor data set including the principal stress direction, shear stress amplitude, and equivalent stress density.

[0025] In a preferred embodiment of the present invention, spatial stress distribution data output by a six-axis loading device is collected in real time, and stress components in each direction are fused through tensor superposition operation to generate a three-dimensional stress tensor dataset including the principal stress direction, shear stress amplitude, and equivalent stress density, including: Raw spatial stress distribution data in the vertical direction, braking direction, and lateral direction is synchronously collected through strain sensors and force sensors built in the six-axis loading device, specifically including: raw stress data in three directions is simultaneously collected through strain sensors (such as resistance strain gauges) and force sensors (such as piezoelectric force sensors) built in the six-axis loading device: Vertical direction: Measure the compressive stress borne by the material in the thickness direction (such as the stress caused by the vertical support force of the road surface). Braking direction: Capture the shear stress along the moving direction of the material (such as the tangential stress generated by friction when braking). Lateral direction: Record the tensile stress in the lateral direction of the vehicle body (such as the pulling stress caused by centrifugal force when turning).

[0026] Data collection covers the entire area of the specimen (such as key parts like the spokes and rims of the wheel hub) to ensure the integrity of the spatial stress distribution.

[0027] The raw data is subjected to filtering and noise reduction processing, and based on a pre-calibrated sensor coordinate system transformation matrix, stress components in each direction are uniformly transformed into a global three-dimensional reference coordinate system, specifically including: using a moving average filter or wavelet noise reduction algorithm to remove high-frequency interference signals (such as noise caused by equipment vibration) to ensure the authenticity of the stress data. For example, when it is detected that the data fluctuation in a certain channel exceeds ±15% of the mean value, adaptive filter correction is triggered. Through the pre-calibrated sensor coordinate system transformation matrix, local stress data in each direction (such as σ'x and τ'xy in the sensor's own coordinate system) is uniformly mapped to the global three-dimensional reference coordinate system (with the vehicle coordinate system as the reference, the X-axis is the braking direction, the Y-axis is the lateral direction, and the Z-axis is the vertical direction) to solve the error caused by inconsistent coordinate systems of multi-sensor data (such as the stress synthesis error in the traditional non-unified coordinate system can reach 20%).

[0028] According to the stress components in the converted global coordinate system, tensor superposition operation is performed on the vertical compressive stress component, braking shear stress component, and lateral tensile stress component in time series to generate an instantaneous three-dimensional stress tensor, specifically including: stress components in three directions (vertical compressive stress σz, braking shear stress τxy, lateral tensile stress σy, etc.) are tensor-superposed in time series to form a three-dimensional stress tensor describing the internal stress state of the material at a certain instant. For example: The vertical compressive stress component acts on the XY plane and is represented as σz; The coupled shear stress component in the braking direction and the lateral direction is represented as τxy; The lateral tensile stress component acts on the XZ plane and is represented as σy.

[0029] The superimposed stress tensor contains six independent components (σx, σy, σz, τxy, τxz, τyz), which completely describe the spatial stress state of the material point.

[0030] Perform eigenvalue decomposition on the instantaneous three-dimensional stress tensor, determine the principal stress direction through the eigenvector corresponding to the maximum eigenvalue, and record its dynamic trajectory changing with the loading time to obtain the eigenvalue decomposition result, specifically including: performing eigenvalue decomposition on the instantaneous three-dimensional stress tensor, and identifying three principal stresses (the maximum tensile stress σ1, the intermediate stress σ2, and the maximum compressive stress σ3) and their directions: Determination of the principal stress direction: The eigenvector corresponding to the maximum eigenvalue points to the direction of the maximum tensile stress inside the material (principal stress axis 1), and the minimum eigenvalue corresponds to the direction of the maximum compressive stress (principal stress axis 3), and both are perpendicular to the intermediate principal stress axis (σ2). For example, in the wheel hub turning condition, the principal stress axis 1 may point to the outside of the rim (tensile), and the principal stress axis 3 points to the inside of the spoke (compressive).

[0031] Recording the dynamic trajectory: Track the change of the principal stress direction with the loading time (such as during the process from straight driving to turning, the principal stress axis gradually deflects laterally from the vertical direction), providing a basis for analyzing the activation sequence of the twin system (such as the principal stress direction determines the preferential initiation of {10-12} twins or {10-11} twins).

[0032] Based on the eigenvalue decomposition result, extract the difference between the intermediate eigenvalue and the minimum eigenvalue as the maximum shear stress amplitude, and synchronously calculate the distribution state of each shear stress plane, specifically including: Based on the eigenvalue decomposition result, calculate the maximum shear stress amplitude (equal to half of the difference between the maximum principal stress and the minimum principal stress), and determine its acting plane (the plane at a 45° angle to the principal stress axes 1 and 3). For example: When σ1 = 80 MPa (tensile) and σ3 = -30 MPa (compressive), the maximum shear stress amplitude is 55 MPa, acting on the plane at a 45° angle to both σ1 and σ3, and this plane is a high-probability region for twin nucleation. At the same time, analyze the stress distribution of each shear plane (such as the degree of shear stress concentration at the grain boundary) to identify potential twin initiation sites.

[0033] Substitute the second invariant of the instantaneous three-dimensional stress tensor into the von Mises equivalent criterion to calculate the equivalent stress density, and generate a continuously distributed equivalent stress density field in the three-dimensional space grid by Gaussian interpolation method; Correlate the principal stress direction, shear stress amplitude, and equivalent stress density with the time stamp and spatial coordinates to construct a three-dimensional stress tensor data set with a hierarchical index, specifically including: Correlate parameters such as the principal stress direction, shear stress amplitude, and equivalent stress density with the time stamp (such as the stress state at the 100th millisecond of loading) and spatial coordinates (such as the specimen surface coordinates (x, y, z)) to establish a hierarchical index (such as a three-level index of "time-position-stress component"). For example: At the time stamp t = 500 ms, the principal stress direction at the rim edge point (x = 100 mm, y = 0, z = 50 mm) is 30° with respect to the X-axis, the maximum shear stress amplitude is 45 MPa, and the equivalent stress density is 110 MPa.

[0034] In the embodiments of the present invention, through multi-sensor synchronous sampling (microsecond-level synchronous accuracy), the time sequence misalignment of stress components caused by traditional time-sharing acquisition is avoided (such as when measuring shear stress after uniaxial loading, the stress state has changed), reducing the error of the synthesized three-dimensional stress tensor from ±18% to ±5%. The global coordinate system transformation solves the problem of "each sensor speaking its own language" for multi-sensor data (such as the deviation of the stress component direction caused by the installation angle of different sensors), increasing the coincidence degree between the stress distribution cloud map and the actual stress state of the material by more than 90%. Through the real-time change trajectory of the principal stress direction, the switching process of the twin system under complex loads can be intuitively observed (such as switching from {10-11} compressive twinning to {10-12} tensile twinning), providing dynamic boundary conditions for calculating the twin nucleation probability (the error of the traditional static principal stress assumption can reach 40%). Accurately identify high shear stress regions (such as the shear stress amplitude at the grain boundary intersection is 60% higher than that of the matrix), and early warn of twin nucleation hotspots, reducing the prediction deviation of the twin evolution model for the nucleation position from ±100 μm to ±30 μm. The continuous equivalent stress density distribution can be directly input into the fatigue damage model (such as Miner's linear cumulative damage theory) to quantify the comprehensive effect of multiaxial stress on the material life. In a certain case, the fatigue life prediction error based on this data is reduced from ±50% to ±15%. The hierarchical index data set supports rapid retrieval of stress characteristics under specific working conditions (such as the stress peak during emergency braking). Combining historical process parameters, the optimal loading path can be automatically recommended through machine learning algorithms (such as adjusting the triaxial load ratio to increase the equivalent stress uniformity by 30%), shortening the process debugging cycle by more than 50%. The stress tensor data can be directly correlated with the microscopic monitoring data of the subsequent twin correction module (such as the strain energy density at the grain boundary and the orientation difference angle of twin bands) to establish a cross-scale mapping relationship of "macroscopic stress state - microscopic twin behavior" (such as when the angle between the principal stress direction and the twin band expansion direction is less than 15°, the twin proliferation rate increases by 50%).

[0035] In a preferred embodiment of the present invention, the second invariant of the instantaneous three-dimensional stress tensor is substituted into the von Mises equivalence criterion, the equivalent stress density is calculated, and a continuously distributed equivalent stress density field is generated in the three-dimensional space grid by Gaussian interpolation method, including: Based on the components of the instantaneous three-dimensional stress tensor, the second invariant is solved by the square sum operation of the stress deviator tensor, and the time-space coordinates in the loading process are associated, specifically including: extracting each stress component (such as vertical compressive stress σz, braking shear stress τxy, lateral tensile stress σy, etc.) from the instantaneous three-dimensional stress tensor, separating the stress deviator tensor (the deviation part after removing the average stress, reflecting the shape change stress of the material) through operation; performing square sum operation on each component of the deviator tensor to obtain the second invariant value, which represents the distortion energy density of the material under the multi-axial stress state. The larger the value, the stronger the tendency of the material to undergo plastic deformation (such as twinning). The current calculation result is associated with the loading time (such as t=200ms) and the space coordinate (such as the sample surface point (x, y, z)), providing a mark for subsequent time-space analysis.

[0036] Substitute the second invariant into the von Mises equivalence criterion, calculate the three-dimensional equivalent stress density scalar value at the current loading moment, and mark its corresponding spatial position, specifically including: First, three principal stresses (maximum tensile stress, intermediate stress, and maximum compressive stress) are extracted from the instantaneous three-dimensional stress tensor, and their average value is calculated. The average value represents the average pressure (or tension) of the material under the current stress state, which mainly causes the volume change of the material (such as compression causing volume reduction and stretching causing volume expansion).

[0037] Subsequently, the mean stress is subtracted from each principal stress to obtain a new set of stress values, called "deviatoric stress components". Deviatoric stress components reflect only the shape distortion (such as twisting, shear deformation) of the material and are directly related to plastic deformation behaviors such as twinning. For example, in a working condition with simultaneous tension and shear, the deviatoric stress component will highlight the effect of shear deformation and exclude the contribution of the mean tensile stress to the volume change.

[0038] Further operations are performed on the deviatoric stress components: each deviatoric stress component is multiplied by itself (squared), and the cross-actions between deviatoric stress components in different directions (such as the coupling effects of tension and shear) are considered. By summing and adjusting with a specific proportional coefficient, a scalar value is obtained, which is the second invariant of the deviatoric stress tensor.

[0039] The physical meaning of this invariant is to measure the distortion energy density of the material under multiaxial stress state. The larger the value, the lower the energy required for the material to change shape (such as twinning), and the easier it is to activate the plastic deformation mechanism. For example, in pure shear conditions, the second invariant will increase significantly, reflecting the efficient driving effect of shear stress on twin nucleation.

[0040] Substitute the second invariant obtained in the second step into the operation logic of the von Mises equivalent criterion: By performing operations such as taking the square root or scaling the second invariant (the specific operation logic is based on the energy equivalence principle in mechanics of materials), a new scalar value, namely the "equivalent stress density", is obtained.

[0041] The core idea of this process is: Assume that the plastic deformation energy of the material under multiaxial stress is equal to the energy during uniaxial tension (or compression), so the complex stress state is equivalent to a "virtual" uniaxial stress state. For example, when the material is simultaneously subjected to a tensile stress of 100 MPa and a shear stress of 50 MPa, the equivalent stress density may be equivalent to a uniaxial tensile stress of 150 MPa, indicating that the plastic deformation tendency of the material at this time is equivalent to that under uniaxial tension of 150 MPa.

[0042] Assign corresponding spatial coordinates (such as the three-dimensional geometric coordinates of the specimen, for example, the center of the spoke is (x1, y1, z1) and the rim edge is (x2, y2, z2)) to each calculated equivalent stress density value to form discrete "stress-position" data pairs.

[0043] These data points cover key parts of the material (such as stress concentration areas and geometric discontinuities). For example, in a wheel hub model, positions prone to fatigue cracks such as the junction of the spoke and the rim and around the bolt holes will be key marked. In this way, the discrete data points can intuitively reflect the stress differences at different spatial positions.

[0044] Take the example of an automotive wheel hub under a turning condition: Actual stress state: The outer side of the rim is subjected to a tensile stress of 150 MPa (from centrifugal force) and a shear stress of 80 MPa (from road surface friction), and the inner side is subjected to a compressive stress of 50 MPa (structural support reaction force).

[0045] Average stress calculation: (150 MPa + (-50 MPa) + 0) / 3 ≈ 33.3 MPa (average tensile stress, mainly causing volume expansion).

[0046] Deviatoric stress components: Outer tensile deviatoric stress = 150 MPa - 33.3 MPa = 116.7 MPa, inner compressive deviatoric stress = -50 MPa - 33.3 MPa = -83.3 MPa, and the shear deviatoric stress remains 80 MPa unchanged.

[0047] Second invariant operation: Calculate through the sum of squares and cross terms of the deviatoric stress components to obtain a value reflecting the distortion energy.

[0048] Equivalent stress density: After equivalent calculation, an equivalent stress value of about 180 MPa is obtained, which is equivalent to the material in this area bearing a uniaxial tensile stress of 180 MPa, significantly exceeding the twin threshold (such as 120 MPa), indicating that intensive twinning will occur in this area.

[0049] In the three-dimensional space grid, taking the marked spatial position as the interpolation node, according to the preset interpolation radius and weight function, the discrete equivalent stress density scalar values are spatially smoothed and interpolated through the Gaussian kernel function to generate a continuously distributed equivalent stress density field. Compare the equivalent stress density field with the preset twin nucleation energy threshold. If the equivalent stress density in a local area exceeds the threshold range, readjust the weight function parameters of the Gaussian interpolation until the density field matches the nucleation energy barrier parameters of the dynamic evolution matrix to obtain a corrected equivalent stress density field, specifically including: comparing the generated equivalent stress density field with the preset twin nucleation energy threshold (such as the minimum equivalent stress value required for material twinning) to identify whether the local area exceeds the threshold.

[0050] If the equivalent stress density in a certain area is significantly higher than the threshold but actual twinning does not occur (possibly due to interpolation error resulting in a falsely high value), then adjust the weight function parameters of the Gaussian interpolation (such as reducing the interpolation radius or changing the width of the Gaussian kernel), reducing the interpolation weight of this area to make the corrected stress field closer to the actual twinning behavior.

[0051] Conversely, if an abnormally high twin density appears in the area within the threshold, increase the weight to enhance the stress representation in this area to ensure that the equivalent stress density field is precisely matched with the nucleation energy barrier parameters (such as the minimum energy required for nucleation) in the dynamic evolution matrix.

[0052] In the embodiment of the present invention, the three-dimensional stress tensor is transformed into a single equivalent stress density value through the von Mises criterion, solving the problem that it is difficult to intuitively compare multiple parameters in traditional multi-axial stress analysis. For example, the equivalent stress values under different working conditions can be directly compared (such as 100 MPa during normal driving vs. 180 MPa during emergency braking), quickly evaluating the fatigue damage risk level. The equivalent stress density is directly related to the energy requirement for material twin nucleation (such as when the twin threshold of a certain magnesium alloy is 120 MPa, the area where the equivalent stress exceeds this value can be directly determined as a high-risk twin area). Compared with discrete stress points, the continuous stress field can reveal fine stress concentrations that cannot be covered by traditional sensors (such as the stress gradient within a range of 0.5 mm at the grain boundary). In a certain case, it is found through Gaussian interpolation that the equivalent stress at the edge of the spoke threaded hole reaches 150 MPa (only 120 MPa was measured at discrete points); the weight attenuation characteristic of the Gaussian kernel function simulates the distance effect of stress propagation (such as the farther away from the load application point, the smaller the stress influence), avoiding the non-physical mutations that may be introduced by linear interpolation and making the stress field distribution more in line with the theory of elasticity.

[0053] Adjust the interpolation parameter through threshold comparison to form a closed-loop feedback between the equivalent stress field and the twinning evolution model. For example, when the corrected stress field shows that the equivalent stress in a certain area reaches 130 MPa (exceeding the threshold of 120 MPa), and the dynamic evolution matrix predicts that the twinning nucleation probability in this area is 45%, the consistency between the two can reduce the twinning density prediction error from ±20% to ±8%. The corrected equivalent stress density field can be directly input into the twinning evolution analysis module as the calculation basis for the nucleation probability density function, and at the same time provide real-time stress distribution feedback for the optimization module (such as the temperature of the hot extrusion die needs to be dynamically adjusted according to the position of the high stress area), improving the synergy of the whole system.

[0054] In a preferred embodiment of the present invention, in a three-dimensional space grid, taking the marked spatial position as an interpolation node, according to a preset interpolation radius and weight function, perform spatial smoothing interpolation on the discrete equivalent stress density scalar values through a Gaussian kernel function to generate a continuously distributed equivalent stress density field, including: Taking the marked spatial position as an interpolation node, dynamically set the interpolation radius of each node according to the geometric characteristics and stress gradient distribution characteristics of the magnesium alloy specimen. The interpolation radius in the high stress gradient area is smaller than that in the low gradient area, specifically including: identifying the geometric mutation areas of the magnesium alloy specimen (such as the spoke-rim junction of the wheel hub, the edge of the bolt hole), where stress concentration usually exists and belongs to the high stress gradient area; relatively smooth areas (such as the middle of the rim) are low stress gradient areas; for high stress gradient areas (such as stress change rate > 50 MPa / mm), set a smaller interpolation radius (such as 0.2 mm) to retain local stress mutation details; for low stress gradient areas (such as stress change rate < 10 MPa / mm), use a larger radius (such as 1.0 mm) to reduce the influence of discrete point noise on the smoothing effect. For example, use a radius of 0.3 mm at the spoke sharp corner and a radius of 0.8 mm in the rim plane area.

[0055] Based on the non-uniformity of the local stress distribution and the material anisotropy parameters, select the Gaussian kernel function weight coefficient that matches the stress state at the current interpolation node, specifically including: determining the stress propagation characteristics of the material in different directions according to the crystal orientation of the magnesium alloy (such as the degree of basal texture). For example, the stress diffusion along the extrusion direction (ED) may be faster. Therefore, in the Gaussian kernel function, the weight coefficient along the ED direction is set to 0.6, and the weight coefficient perpendicular to the ED direction is set to 0.4.

[0056] Stress state adaptation: For nodes under unidirectional tension, the weight function uses a circularly symmetric Gaussian kernel (with the same weight in all directions); for multi-axial stress coupling nodes (such as those simultaneously subjected to tension and shear), an elliptical Gaussian kernel is used, where the major axis direction corresponds to the principal stress direction and the minor axis corresponds to the secondary stress direction, and the weight coefficients are distributed according to the ratio of the principal stress amplitudes (e.g., when the principal stresses σ1:σ2 = 3:1, the weight coefficient of the major axis is 0.7 and that of the minor axis is 0.3).

[0057] Centering on the interpolation node, traverse the adjacent grid cells in the three-dimensional space grid, and calculate the Gaussian weighted average stress density value in each cell according to the weight coefficient and the interpolation radius to complete the spatial smooth transition of the discrete scalar value. Specifically, it includes: Centering on the current interpolation node, search for all adjacent grid cells (such as a cubic neighborhood) in the three-dimensional grid whose distance is less than or equal to the interpolation radius. For example, when the node coordinates are (x0, y0, z0) and the radius is 0.5 mm, the neighborhood includes all cells within the range of (x0 ± 0.5 mm, y0 ± 0.5 mm, z0 ± 0.5 mm).

[0058] For each neighborhood cell, calculate the Gaussian weight value according to its distance d from the node (the closer the distance, the higher the weight, showing a normal distribution decay). For example, the weight of a cell at a distance of 0.1 mm from the node is 0.9, the weight at a distance of 0.4 mm is 0.2, and the weight of a cell whose distance exceeds the radius is 0. Sum the products of all discrete stress values in the neighborhood multiplied by their corresponding weights, and then divide by the total weight sum to obtain the Gaussian weighted average stress density value of the cell. Starting from one end of the specimen, traverse all three-dimensional grid cells row by row, column by column, and layer by layer, and repeat the above weighted average calculation for each cell until the entire specimen space is smoothly transitioned.

[0059] Map the Gaussian weighted average stress density value to the vertices of the three-dimensional grid, and connect adjacent vertices through the bicubic spline interpolation algorithm to generate an equivalent stress density field with spatial continuity and differential continuity. Specifically, it includes: Map the Gaussian weighted average stress value of each grid cell to the eight vertices of the cell to ensure that the vertex data contains the comprehensive stress information of the neighborhood cells. For example, a certain vertex is shared by four adjacent cells, and each cell contributes 25% of the stress value of this vertex.

[0060] In each coordinate axis direction (X, Y, Z axes) of the three-dimensional grid, use the bicubic spline interpolation algorithm to connect the stress values of adjacent vertices. This algorithm generates a smooth curve by fitting the function values, first-order derivatives, and second-order derivatives of adjacent points to ensure that the stress values at the boundaries of adjacent cells are continuous and the derivatives are continuous (without sudden jumps).

[0061] Through the above interpolation, the entire specimen space is transformed into a continuous stress density field, and the stress value at any position can be directly calculated through the field function. For example, the stress value at a certain point (x = 50 mm, y = 0, z = 30 mm) at the center of the spoke can be obtained by interpolating the spline function of the surrounding vertices without relying on discrete data points.

[0062] Compare the equivalent stress density field after interpolation with the stress distribution data measured by the sensor. If the interpolation error in a local area exceeds the preset tolerance, the interpolation radius is adaptively reduced until the density field meets the accuracy requirements of the dynamic evolution matrix, and a continuously distributed equivalent stress density field is generated, which specifically includes: comparing the generated stress field with the measured data of the strain sensor pasted on the specimen surface (such as the measured equivalent stress of 140 MPa at a certain point on the rim edge), and calculating the interpolation error in the local area (such as |calculated value - measured value| / measured value).

[0063] Adaptive reduction of the radius: If the error in a certain area exceeds the preset tolerance (such as 5%), the interpolation radius of this area is reduced by 20% (such as from 0.5 mm to 0.4 mm), and the weighted average stress values of this area and its neighborhood are recalculated, and the spline interpolation result is updated. For example, if the calculated value at the root of the spoke is 160 MPa and the measured value is 145 MPa, with an error of 10%, the radius is reduced to 0.3 mm and then re - interpolated until the error is reduced to less than 4%.

[0064] Iteration termination condition: When the interpolation errors of all key monitoring points are lower than the tolerance, and the spatial distribution trend of the stress field is consistent with the theory of material mechanics (such as the stress concentration area coincides with the geometric mutation position), stop the adjustment and output the final continuous stress field.

[0065] In the embodiments of the present invention, small-radius interpolation is used to avoid over-smoothing and masking of stress peaks in high-stress concentration areas (such as grain boundaries and notches). For example, in the stress concentration area at the edge of a bolt hole, small-radius interpolation can retain a peak stress of 180 MPa, while large-radius interpolation may smooth it to 150 MPa, resulting in an error in the prediction of twinning nucleation. Large-radius interpolation can filter the sensor measurement noise in low-stress change areas (such as ±3 MPa fluctuations), making the stress field more conform to the macroscopic stress trend of the material. For example, it can reduce the stress fluctuation in the rim plane from ±5 MPa to ±1 MPa. By matching the elliptical Gaussian kernel with the principal stress direction, the stress conduction differences along different crystal orientations in magnesium alloys can be simulated. For example, the stress diffusion is faster along the basal plane (0001) direction. The major axis of the Gaussian kernel is set along this direction, making the interpolation result closer to the actual stress propagation path. Compared with isotropic interpolation, the direction error is reduced by 30%. By adjusting the weight coefficient in combination with material texture parameters (such as the basal texture strength factor), the dependence of twinning activation on crystal orientation can be reflected. For example, in the area with strong basal texture, the stress weight in the direction perpendicular to the basal plane is higher, making it easier to trigger {10-12} tensile twinning. Bicubic spline interpolation ensures the continuity of the first derivative (stress gradient) and the second derivative (curvature) of the stress field, avoiding the "stress mutation" artifact caused by traditional linear interpolation, which is crucial for calculating the twinning multiplication rate based on the stress gradient (such as a gradient mutation may lead to a deviation of ±40% in the prediction of the multiplication rate). By calibrating the interpolation result in real time with sensor data, the model deviation caused by pure algorithm dependence can be avoided. For example, during the hot extrusion process, the change in die temperature may cause a change in the elastic modulus of the material, which in turn affects the stress distribution. Adaptive adjustment can compensate for such unexpected changes in real time, keeping the interpolation error within 5% all the time.

[0066] Taking an automotive magnesium alloy wheel hub as an example: Interpolation effect in high-stress areas: At the acute angle junction of the spoke and the rim (stress gradient > 80 MPa / mm), using a radius of 0.2 mm and an elliptical Gaussian kernel (major axis along the spoke extension direction), the stress field after interpolation clearly shows a peak stress of 190 MPa, which is in agreement with the in-situ X-ray diffraction measured value of 185 MPa, successfully capturing the local stress concentration missed by traditional discrete point measurements.

[0067] Smoothing effect in low-stress areas: In the hub center plane (stress gradient < 10 MPa / mm), using a radius of 1.0 mm and a circular Gaussian kernel, the stress fluctuation of discrete points (110 ± 8 MPa) is smoothed to 112 ± 2 MPa, providing stable basic data for fatigue life prediction.

[0068] Adaptive adjustment example: After initial interpolation, it is found that the error in a certain area on the inner side of the rim reaches 7% (calculated value 135 MPa, measured value 126 MPa). By reducing the radius to 0.4 mm and adjusting the weight coefficient, the final error is reduced to 3%, ensuring that the calculation deviation of the twinning nucleation probability in this area is less than 5%.

[0069] Through this process, the discrete stress data is converted into a high-precision continuous field, providing stress input that highly matches the actual working conditions for the twinning evolution analysis module, thereby supporting the precise optimization of hot extrusion process parameters and the reliable prediction of fatigue performance.

[0070] In a preferred embodiment of the present invention, the nucleation probability density function of twins is calculated based on the three-dimensional stress tensor dataset, and a dynamic evolution matrix including nucleation energy barrier, proliferation rate, and orientation rotation is constructed, including: Based on the principal stress direction, shear stress amplitude, and equivalent stress density field in the three-dimensional stress tensor dataset, combined with the relationship between the crystal slip system of magnesium alloy and the critical resolved shear stress of twins, a nucleation probability density function is established to quantify the twinning nucleation tendency at different spatial positions and loading times. Specifically, it includes: obtaining relevant information on the principal stress direction, shear stress amplitude, and equivalent stress density field from the three-dimensional stress tensor dataset. The principal stress direction determines the main direction of the material's stress, the shear stress amplitude reflects the intensity of the shear action inside the material, and the equivalent stress density field comprehensively reflects the stress level of the material. Determine the characteristics of the slip system of the magnesium alloy crystal and understand the critical resolved shear stress of different slip systems. The critical resolved shear stress refers to the minimum resolved shear stress required for slip or twinning to occur on a specific slip system. Combining the principal stress direction and shear stress amplitude, determine the resolved shear stress magnitude on each slip system. According to the relationship between the resolved shear stress and the critical resolved shear stress, establish a nucleation probability density function. When the resolved shear stress exceeds the critical resolved shear stress, the possibility of twinning nucleation increases. Through this function, the twinning nucleation tendency at different spatial positions and loading times can be quantified. For example, in high stress concentration areas and specific loading times, the nucleation probability may increase significantly.

[0071] According to the dynamic difference between the equivalent stress density field and the preset twinning nucleation energy threshold, calculate the nucleation energy barrier parameter of the local area, which characterizes the degree of energy accumulation required to overcome the lattice resistance. Specifically, it includes: comparing the equivalent stress density field with the preset twinning nucleation energy threshold and calculating their dynamic difference. This difference reflects the gap between the energy required for the material to undergo twinning nucleation under the current stress state. According to the above difference, calculate the nucleation energy barrier parameter of the local area. The nucleation energy barrier parameter characterizes the degree of energy accumulation required to overcome the lattice resistance and promote twinning nucleation. The larger the difference, the more energy accumulation is required to reach the nucleation condition, and the larger the nucleation energy barrier parameter.

[0072] Calibrate the twin multiplication rate parameter based on the time series change rate of the shear stress amplitude and the stability of the principal stress direction, and correlate the spatial gradient distribution of the nucleation probability density function, specifically including: Analyze the change rate of the shear stress amplitude over time. A rapidly changing shear stress may accelerate twin multiplication because it provides more energy and driving force. For example, under impact loading, a rapid increase in the shear stress amplitude will accelerate the twin multiplication rate. Evaluate the stability of the principal stress direction. A stable principal stress direction is conducive to the continuous growth and multiplication of twins in a certain direction; while frequent changes in the principal stress direction may interfere with the growth of twins and reduce the multiplication rate. Integrate the time series change rate of the shear stress amplitude and the stability of the principal stress direction to calibrate the twin multiplication rate parameter. At the same time, correlate this parameter with the spatial gradient distribution of the nucleation probability density function. A region with a large spatial gradient of the nucleation probability density function means that the possibility of twin nucleation changes violently in space, and the twin multiplication rate in these regions may also be affected.

[0073] Construct an orientation rotation parameter matrix by matching the dynamic trajectory of the principal stress direction with the magnesium alloy crystal orientation database to describe the crystallographic reorientation behavior during twin growth, specifically including: Record the dynamic trajectory of the principal stress direction over time. The change in the principal stress direction affects the growth direction and crystallographic orientation of twins. Match the dynamic trajectory of the principal stress direction with the magnesium alloy crystal orientation database, which contains relevant information on magnesium alloy crystals under different orientations and loading conditions. Through matching, the law of crystallographic reorientation during twin growth under the action of the principal stress can be determined. According to the matching results, construct an orientation rotation parameter matrix, which describes the crystallographic reorientation behavior during twin growth, including information such as the change angle and direction of the crystal orientation.

[0074] Integrate the nucleation energy barrier parameter, the multiplication rate parameter, and the orientation rotation parameter matrix in the time - space dimension to generate a dynamic evolution matrix, specifically including: Integrate the nucleation energy barrier parameter, the multiplication rate parameter, and the orientation rotation parameter matrix in the time and space dimensions. The time dimension considers the change of parameters with the loading time, and the space dimension considers the differences in parameters at different positions. By integrating the above parameters, a dynamic evolution matrix is generated, which comprehensively describes the evolution of twin nucleation, multiplication, and orientation rotation behaviors at different times and spatial positions.

[0075] In the embodiments of the present invention, by establishing a nucleation probability density function, the tendency of twin nucleation at different spatial positions and loading times can be accurately quantified. This helps to predict in advance the regions and times where twins may appear, providing an important basis for the design and processing of materials. For example, in the manufacturing process of automotive magnesium alloy components, the process parameters can be optimized according to the nucleation probability distribution to avoid excessive twins in critical parts, thereby improving the performance and reliability of the components. Calculating the nucleation energy barrier parameter can intuitively understand the degree of energy accumulation required to overcome the lattice resistance, which helps to deeply study the physical mechanism of twin nucleation and provides theoretical support for the development of new materials and processes. For example, by reducing the nucleation energy barrier, the twin nucleation ability of the material can be improved, thereby improving the plastic deformation ability of the material. Dynamically tracking the difference between the equivalent stress density field and the twin nucleation energy threshold can monitor the change of the energy state of the material during loading in real time and timely discover the key factors that may lead to twin nucleation. Calibrating the twin multiplication rate parameter, combined with the time series change rate of the shear stress amplitude and the stability of the principal stress direction, can effectively control the twin multiplication process. By adjusting the loading conditions or the microstructure of the material, the multiplication rate can be changed, thereby realizing the regulation of the material properties. For example, in the case where the material strength needs to be improved, the twin multiplication rate can be appropriately reduced to reduce the number of twins; while in the case where the material plasticity needs to be improved, the twin multiplication rate can be increased. Associating the multiplication rate parameter with the spatial gradient distribution of the nucleation probability density function can more comprehensively consider the mutual relationship between twin nucleation and multiplication and improve the prediction ability of the microstructure evolution of the material. For example, in the rolling process of magnesium alloy sheets, by controlling the orientation rotation of twins, the mechanical properties and forming properties of the sheets can be improved. Using the matching of the dynamic trajectory of the principal stress direction and the magnesium alloy crystal orientation database makes the calculation of the orientation rotation parameter more accurate and reliable, providing more precise guidance for the microstructure design of the material.

[0076] In a preferred embodiment of the present invention, a first target monitoring point and a second target monitoring point are set in the pre-twin region. The first target monitoring point is located at the intersection of grain boundaries, and the second target monitoring point is located at the center of the twin band. The orientation difference angle and the strain energy density difference data of the two monitoring points are obtained in real time, the covariance matrix of the change characteristics of the two monitoring points is calculated, and the multiplication rate parameter and the orientation rotation parameter in the dynamic evolution matrix are corrected according to the correlation coefficient of the covariance matrix to obtain a corrected dynamic evolution matrix, including: According to the grain boundary topological structure and the twin band morphology characteristics in the pre-twin region, a first target monitoring point is set at the intersection of grain boundaries, a second target monitoring point is set at the geometric center of the twin band, and the initial crystal orientations and the strain energy density baseline values of the two monitoring points are calibrated; Collect the crystal orientation difference angle data of the first target monitoring point and the second target monitoring point in real time, and obtain the local strain energy density difference data of the two monitoring points to form a time-series monitoring data set, specifically including: using high-resolution EBSD (spatial resolution ≤ 200 nm) to collect the crystal orientations of the two monitoring points in real time, calculating the orientation difference angle (θ1, reflecting the degree of grain boundary slip) between adjacent grains at the grain boundary intersection and the orientation difference angle (θ2, reflecting the degree of twin rotation) between the center of the twin band and the matrix, and setting the time interval to 50 ms (matching the loading rate of 0.1 mm / min).

[0077] Measure the local strain energy densities U1 (at the grain boundary) and U2 (at the center of the twin band) of the two monitoring points through Raman spectroscopy or electron energy loss spectroscopy (EELS), and calculate the difference value ΔU = U1 - U2. For example, when the stress concentration at the grain boundary causes U1 to increase to 0.8 MJ / m 3 while U2 at the center of the twin band is 0.4 MJ / m 3 then ΔU = 0.4 MJ / m 3 . Organize θ1, θ2, and ΔU into a three-dimensional array according to the timestamps (t = 0, 50 ms, 100 ms,...). For example, the data set at a certain moment t is [θ1(t) = 12°, θ2(t) = 15°, ΔU(t) = 0.6 MJ / m 3 , forming a dynamic sequence reflecting the deformation process.

[0078] Based on the monitoring data set, extract the covariance components of the orientation difference angle change rate and the strain energy density change rate respectively, construct a covariance matrix reflecting the deformation coordination characteristics of the two monitoring points, and calculate the correlation coefficient quantifying the correlation strength between local and global deformations; If the correlation coefficient indicates that the deformation coordination between the grain boundary intersection and the center of the twin band is lower than the preset threshold, then proportionally reduce the proliferation rate parameter in the corresponding area of the dynamic evolution matrix to inhibit the excessive growth of non-uniform twins caused by local stress concentration; dynamically adjust the orientation rotation parameter in the dynamic evolution matrix according to the variance component of the orientation difference angle change rate in the covariance matrix to enhance the compensation weight of grain boundary slip to the grain reorientation behavior of the twin band, so as to obtain a corrected dynamic evolution matrix.

[0079] In the embodiment of the present invention, through the dual-site monitoring of the grain boundary and the center of the twin band, the real-time quantification of the "grain boundary slip - twin rotation - energy accumulation" ternary coupling relationship is realized. For example, the phenomenon that grain boundary slip lags behind twin rotation, which cannot be found in traditional single-site monitoring, can be timely identified through covariance analysis, avoiding the error caused by the traditional model ignoring the deformation time sequence difference (the traditional model error can reach ±30%).

[0080] The corrected dynamic evolution matrix can directly output the twin density distribution aligned with the micro-monitoring data, providing immediate feedback for adjusting the hot extrusion process parameters. For example: When the grain boundary-twin synergy decreases, the optimization module automatically reduces the extrusion speed from 5 mm / s to 3 mm / s, prolongs the dynamic recrystallization time, and reduces the twin density from 40% to the target range of 20%-30%; The dynamic correction of the orientation rotation parameter can early warn of abnormal twin textures (such as the proportion of {10-12} twins > 70%). By adjusting the die temperature (from 350 °C to 380 °C), it promotes the balanced growth of {10-11} twins and improves the isotropic properties of the material.

[0081] In a preferred embodiment of the present invention, according to the grain boundary topological structure and twin band morphological characteristics in the pre-twin region, a first target monitoring point is set at the grain boundary intersection, and a second target monitoring point is set at the geometric center of the twin band. The initial crystal orientations and strain energy density baseline values of the two monitoring points are calibrated, including: Based on the equivalent stress density field and twin nucleation probability density function in the three-dimensional stress tensor dataset, regions with equivalent stress density higher than a preset threshold are screened as pre-twin candidate regions, and the final monitoring region is determined by combining hotspot analysis of the grain boundary topological structure. Specifically, it includes: extracting the equivalent stress density field from the three-dimensional stress tensor dataset, setting the equivalent stress threshold for twin nucleation (such as 120 MPa), and screening regions in the field with equivalent stress density ≥ threshold as pre-twin candidate regions. For example, in the simulation of an automotive wheel hub, the equivalent stress at the junction of the spoke and the rim reaches 150 MPa, far exceeding the threshold, and is marked as a candidate region.

[0082] Grain boundary topological hotspot analysis: Perform a grain boundary network scan on the candidate region (such as using the electron backscatter diffraction EBSD technique) to identify regions with a grain boundary density ≥ 3 grains / mm 2 of high complexity. Calculate the stress concentration factor (Kt) at the grain boundary through finite element simulation, and preferentially select regions with Kt ≥ 2.5 as the final monitoring regions (such as near the triple grain boundary intersection points); Scan the grain boundary network of the monitoring region, identify geometric feature points where three or more grain boundaries intersect, and determine the corresponding intersection points as the first target monitoring points according to the stress concentration degree. Specifically, it includes: in the EBSD map of the monitoring region, locate geometric feature points where three or more grain boundaries intersect through grain boundary angle analysis (such as an angle < 90°). For example, in a certain AZ91 magnesium alloy specimen, a star-shaped node formed by the intersection of 4 grain boundaries is found, and the grain boundary angles are 60°, 80°, 100°, and 120° respectively.

[0083] Stress concentration degree ranking: For each intersection point, calculate its local strain energy density (U) based on finite element method, and select the top 10% nodes with the highest U value as the first target monitoring points. For example, among 50 candidate intersection points, select 5 points where U≥0.7 MJ / m 3 as the final monitoring points, where the highest U value reaches 1.2 MJ / m 3 (located in the fine-grained region with grain size <5μm).

[0084] Within the monitoring area, determine the geometric center point of the longitudinal extension direction of the twin band according to the geometric symmetry of the twin band morphology and the distribution gradient of the strain energy density, and use it as the second target monitoring point, specifically including: observing through a transmission electron microscope (TEM) or an in-situ tensile stage, tracking the initiated twin bands (such as {10-12} tensile twins) within the monitoring area, and measuring their length (L), width (W) and extension direction (the angle θ with the principal stress axis). For example, a certain twin band has a length of 20μm, a width of 8μm, and extends along the principal stress axis direction (θ = 15°).

[0085] Geometric center calculation: Define the midpoint of the longitudinal direction (extension direction) of the twin band as the geometric center point, and the calculation formula is: longitudinal coordinate = starting coordinate + L / 2×cosθ, transverse coordinate = starting coordinate + L / 2×sinθ. Ensure that the distance from this point to the twin boundary is ≥3μm (to avoid the interference of grain boundary stress). For example, the center point of the above twin band is located 10μm from the starting point and 4μm from the twin boundaries on both sides.

[0086] Perform an initial crystal orientation scan on the two monitoring points to obtain the initial Euler angle data in the unloaded state, and verify the consistency of its crystal structure based on the magnesium alloy crystallography database, specifically including: using high-resolution EBSD (spatial resolution 200nm) to perform an orientation scan on the two monitoring points to obtain the Euler angles (φ1, Φ, φ2) in the unloaded state. For example, the Euler angles of adjacent grains A of the first target monitoring point are (30°, 45°, 60°), and those of grain B are (15°, 30°, 75°), and the orientation difference angle θ = 25°; the Euler angles at the center of the twin band where the second target monitoring point is located are (40°, 50°, 55°), and the orientation difference angle θ0 with the matrix is 8°.

[0087] Crystal structure verification: Import the Euler angle data into the magnesium alloy crystallography database (such as the ICSD database) to verify whether the monitoring points are located in the target crystal structure region (such as the basal texture dominant region). If the deviation exceeds 5° (such as the difference between the calculated value and the database standard value >5°), reselect the monitoring points.

[0088] Under zero load condition, measure the initial local strain fields of two monitoring points, and calculate the initial strain energy density value in combination with the material elastic modulus parameters, specifically including: measure the initial strain fields (εx, εy, εz) of the two monitoring points through digital image correlation method (DIC) or nano-indentation technology. For example, the initial strain of the first target monitoring point is εx = 0.001, εy = 0.002, εz = -0.003 (compressive strain); the second target monitoring point is εx = 0.0005, εy = 0.0008, εz = -0.001.

[0089] Calculation of strain energy density: In combination with the material elastic modulus parameters (such as E = 45 GPa and V = 0.35 for AZ31), calculate the initial strain energy density U0 = 0.5×E×(εx 2 + εy 2 + εz 2 + 2V(εxεy + εyεz + εzεx)) through Hooke's law. For example, U0 of the first target monitoring point is 0.6 MJ / m 3 , and U0 of the second target monitoring point is 0.3 MJ / m 3 .

[0090] In the initial loading stage, monitor the strain energy density change rate of the two monitoring points in real time to calibrate the initial crystal orientations and the strain energy density baseline values of the two monitoring points, specifically including: in the initial loading stage (such as applying 10% of the target load), monitor the strain energy density change rate dU / dt of the two monitoring points in real time. If the change rate fluctuates by more than ±10% (such as the theoretical value dU / dt = 0.05 MJ / m 3 ·s, and the measured value deviates to 0.04 or 0.06), then adjust the baseline value until the fluctuation is within ±5% to ensure the accuracy of the initial parameters.

[0091] In the embodiment of the present invention, through Kt value sorting and grain boundary topology analysis, ensure that the first target monitoring point is located in the region with the highest probability of twinning nucleation (such as the nucleation probability at the triple grain boundary is 40% higher than the average), and avoid the problem of signal drowning caused by traditional random sampling (the missed detection rate of the traditional method reaches 30%). The geometric center positioning method ensures that the second target monitoring point is free from grain boundary stress interference and accurately reflects the orientation rotation behavior inside the twin (the rotation error in the edge region is 25% higher than that in the center), providing pure data for the orientation evolution model. Exclude non-target texture regions (such as randomly oriented grains) through database verification, so that the coincidence degree between the crystal orientations of the monitoring points and the model assumptions is increased from 70% to 95%, and avoid the calculation error of the nucleation probability caused by orientation deviation (the error can reach ±20% when not verified traditionally). The real-time monitoring in the loading stage effectively compensates for the influence of residual stress in the specimen preparation process (such as the residual strain energy after hot extrusion can reach 0.2 MJ / m 3 ), and reduces the initial U0 error from ±0.15 MJ / m3 Reduced to ±0.05MJ / m 3 , improving the reliability of subsequent covariance analysis.

[0092] The high stress sensitivity at the grain boundary complements the orientation sensitivity at the twin zone center, allowing the coupling effect of "stress concentration-orientation rotation" to be directly observed (traditional single-point monitoring cannot analyze the relationship between the two), and the calculation error of the correlation coefficient is reduced by 40%. Accurate initial parameters reduce the prediction error of the model's proliferation rate from ±15% to ±8% at the initial stage of loading, especially in multi-axial load mutation scenarios (such as switching from tension to shear), and the model response lag time after baseline calibration is shortened by 60%.

[0093] In a preferred embodiment of the present invention, based on the monitoring data set, the covariance components of the orientation misalignment angle change rate and the strain energy density change rate are extracted respectively, a covariance matrix reflecting the deformation synergy characteristics of the two monitoring points is constructed, and the correlation coefficient of the correlation strength of the quantitative local and global deformation is calculated, including: The misorientation angle and strain energy density data in the monitoring data set are segmented into time windows, the change rate sequence in each time window is extracted, and the abnormal fluctuation data segments caused by sensor noise or transient interference are eliminated by statistical outlier detection algorithm, specifically including: dividing the continuous monitoring data set (such as θ1, θ2, ΔU sequence containing 1000 time points) into overlapping time windows (such as window length of 100 time points, overlap rate of 50%), each window corresponds to an analysis cycle (such as a certain stage in the loading process). For example, for a loading process with a total duration of 50s, 10 analysis windows are generated with a window length of 5s and a step length of 2.5s.

[0094] Outlier Detection: For the orientation misorientation angle change rate (dθ / dt) and strain energy density change rate (dΔU / dt) sequences in each window, the Z-score algorithm is used to detect outliers. If a data point deviates from the window mean by more than 3 times the standard deviation (e.g., the dθ / dt mean is 0.5° / ms, the standard deviation is 0.1° / ms, and the point detected at 1.0° / ms is an outlier), it is marked as an abnormal fluctuation.

[0095] Data repair: For abnormal data segments, linear interpolation of adjacent points before and after is used for repair (for example, if the values ​​before and after the abnormal point are 0.4° / ms and 0.6° / ms, the repair value is set to 0.5° / ms).

[0096] Within the same time window, calculate the covariance component between the rate of change of the orientation difference angle of the first target monitoring point and the second target monitoring point, as well as the covariance component between the rate of change of the strain energy density of the two monitoring points, construct a two-dimensional covariance matrix, and normalize its off-diagonal elements. Specifically, it includes: within a single time window, extract the sequence of the rate of change of the orientation difference angle (dθ1 / dt) of the first target monitoring point and the sequence of the rate of change of the orientation difference angle (dθ2 / dt) of the second target monitoring point, as well as the sequence of the rate of change of their strain energy density. For example, within a certain window, the dθ1 / dt sequence is [0.2, 0.3, 0.25,...], and the dθ2 / dt sequence is [0.3, 0.4, 0.35,...].

[0097] Covariance calculation: Orientation difference angle synergy: Calculate the covariance Cov(θ1, θ2) of dθ1 / dt and dθ2 / dt, which reflects the synchronization degree of grain boundary slip and twin rotation. If both increase simultaneously, the Cov value is positive; if one increases and the other decreases, the Cov value is negative.

[0098] Energy change synergy: Calculate the covariance Cov(U1, U2) of the rate of change of the strain energy density of the first target monitoring point and the second target monitoring point, which reflects the correlation between grain boundary stress concentration and twin band energy accumulation. For example, when the energy at the grain boundary is rapidly released (the rate of change of the strain energy density of the first target monitoring point shows negative growth), if the energy of the twin band increases synchronously (the rate of change of the strain energy density of the second target monitoring point shows positive growth), then Cov(U1, U2) is negative. Organize the calculation results in matrix form: ; where Var(θ1) and Var(θ2) are the variances of the rate of change of the orientation difference angle of each monitoring point, reflecting their rotational stability.

[0099] Based on the normalized covariance matrix, calculate the Pearson correlation coefficient characterizing the deformation synergy between the grain boundary intersection and the twin band center, as a quantitative index of the influence of local deformation on global twin evolution. Specifically, it includes: normalize the off-diagonal elements (Cov(θ1, θ2)) of the covariance matrix, divide by the product of the standard deviations of the two variables, to obtain the Pearson correlation coefficient ρ(θ1, θ2), with a range of [-1, 1]. For example, if Cov(θ1, θ2) = 0.06, σ(θ1) = 0.2, and σ(θ2) = 0.3, then ρ = 0.06 / (0.2 × 0.3) = 1, indicating a perfect positive correlation.

[0100] Synergy judgment: A preset synergy threshold (e.g., ρ ≥ 0.4 is considered good synergy). If ρ(θ1, θ2) < 0.4, it indicates that grain boundary sliding is decoupled from twin rotation, and local deformation (e.g., grain boundary stress concentration) is not effectively transmitted to the center of the twin band, which may lead to non-uniform twin growth.

[0101] Proliferation rate parameter adjustment: When ρ(θ1, θ2) < 0.4, the proliferation rate parameter k of the corresponding area is reduced proportionally. For example, if the current k = 0.04 / s, ρ = 0.3 (threshold 0.4), k is adjusted to 0.04×(0.3 / 0.4) = 0.03 / s to suppress the excessive growth of twins due to insufficient synergy (for example, the twin density in a certain area is reduced from the predicted 35% to 28%, close to the measured value of 25%).

[0102] Orientation rotation parameter adjustment: Extract Var(θ2) (the variance of the twin band misorientation angle change rate) from the covariance matrix. If Var(θ2) is greater than 1.5 times the initial variance (e.g., initial Var = 0.02, current Var = 0.035), it indicates that the stability of twin rotation has decreased, and the compensation weight of grain boundary sliding on orientation rotation needs to be increased. For example, adjust the orientation rotation parameter α from 0.01rad / MPa・s to 0.01×(0.035 / 0.02) = 0.0175rad / MPa・s to make the model more sensitive to the correction effect of grain boundary sliding on orientation.

[0103] In an embodiment of the present invention, by analyzing deformation synergy in real time through a sliding window, transient decoupling events during loading (such as a sudden drop in ρ value from 0.5 to 0.2 under impact load) can be captured, and the response time is shortened from 500ms of traditional global analysis to 50ms. Sensor noise (such as dθ / dt mutation caused by electromagnetic interference) is eliminated through Z-score detection, so that the correlation coefficient calculation error is reduced from ±20% to ±5%, avoiding model correction deviation caused by misjudgment. When ρ < 0.4 is detected, timely reduction of the k value can effectively suppress the twin density peak in the local area. In a wheel hub spoke case, the corrected twin density standard deviation was reduced from 12% to 6%, and the fatigue crack initiation life was increased by 50%. Through the adjustment of the α value driven by Var (θ2), the fluctuation range of the misorientation angle of the twin band is reduced from ±10° to ±5°, so that the consistency between the twin texture prediction and the EBSD measurement is increased from 70% to 85%, optimizing the anisotropic properties of the material.

[0104] Analysis of covariance simultaneously considers the dynamic correlation between stress (ΔU) and orientation (θ), enabling the model to reflect the true physical process of "grain boundary stress concentration - twin rotation lag" (the traditional model only considers the single factor of stress, with an error of ±30%). The corrected dynamic evolution matrix can predict the abnormal trend of twin density 3 - 5 time windows in advance (for example, a continuous decrease in ρ indicates local overgrowth), providing early warning for the temperature adjustment of hot extrusion dies (such as heating up 10°C in advance to activate dynamic recrystallization), and improving the process debugging efficiency by 40%.

[0105] In a preferred embodiment of the present invention, according to the twin density gradient distribution output by the corrected dynamic evolution matrix, the hot extrusion process parameter group is adjusted. The parameter group includes die temperature, extrusion speed, and strain path, and the twin density is dynamically controlled within the target range, including: Based on the corrected dynamic evolution matrix, the twin density gradient distribution data in the current loading stage is extracted, and the spatial positions and deviation amplitudes of the high - density area and low - density area beyond the target range are identified. Specifically, it includes: extracting the twin density distribution data in the current loading stage (such as the hot extrusion die filling stage) from the corrected dynamic evolution matrix, and generating a three - dimensional grid cloud map with a resolution of 0.1 mm. For example, in the hot extrusion simulation of a magnesium alloy wheel hub, it is found that the twin density at the root of the spoke reaches 45% (the target range is 20% - 30%), and only 15% at the rim edge, forming a significant gradient (Δρ = 30%).

[0106] Abnormal area location: Set density thresholds (upper limit 30%, lower limit 20%), and identify the spatial positions of the high - density area (ρ > 30%) and low - density area (ρ < 20%) through the contour map. For example, the root of the spoke is marked as a high - density abnormal area (volume ratio 15%), and the rim edge is a low - density area (volume ratio 10%).

[0107] Deviation amplitude calculation: Calculate the deviation between the abnormal area and the target value. For example, the high - density area exceeds the threshold by 15%, and the low - density area is below the threshold by 5%, to determine the area that needs to be preferentially regulated.

[0108] If the high-density area exceeds the threshold, the die temperature is reduced to inhibit the dynamic recrystallization rate; otherwise, the temperature is increased to promote twinning homogenization, so as to obtain the adjusted die temperature. According to the local gradient change of the strain path, the extrusion speed is adjusted to control the strain rate distribution, so that the twinning density gradient converges along the preset path, so as to obtain the adjusted extrusion speed. Based on the spatial correlation between the principal stress direction and the twinning sensitive area, the segmented loading strategy of the strain path is re-planned to optimize the stress distribution uniformity, so as to obtain the adjusted strain path parameters. According to the adjusted die temperature, extrusion speed and strain path parameters, the predicted corrected twinning density gradient distribution is obtained and compared with the spatial coverage of the target interval, and the deviation convergence rate is calculated. During the hot extrusion process, the actual twinning density data is collected in real time and dynamically matched with the predicted corrected twinning density gradient distribution. If the deviation convergence rate in the local area does not meet the preset requirements, the process parameter group is iteratively optimized based on the change trend of the Pearson correlation coefficient until the twinning density is dynamically controlled within the target interval, specifically including: If the high-density area exceeds the threshold, reduce the die temperature (e.g., from 380 °C to 350 °C) to reduce twinning proliferation by inhibiting the dynamic recrystallization rate. For example, for every 10 °C decrease in temperature, the activation energy of dynamic recrystallization increases by 5%, and the twinning proliferation rate k decreases by 0.005 / s, causing the high-density area ρ to gradually decrease from 45% to below 30%.

[0109] Strategy for homogenizing the low-density area: If the low-density area is below the threshold, increase the die temperature (e.g., from 350 °C to 380 °C) to promote atomic diffusion to homogenize the twinning distribution. For example, when the temperature increases by 30 °C, the dynamic recrystallization rate increases by 30%, and the twinning nucleation probability increases by 12%, causing the low-density area ρ to increase from 15% to above 20%.

[0110] Temperature gradient setting: Use zone temperature control for local abnormal areas (e.g., set an independent cooling circuit at the die root), so that the temperature in the high-density area is 20 °C lower than the average, and 10 °C higher in the low-density area, forming a directional temperature gradient to guide twinning homogenization.

[0111] According to the strain rate sensitivity coefficient (e.g., the twinning density of AZ31 magnesium alloy increases by 20% at a strain rate of 10 -3 / s), adjust the extrusion speed. For example, the extrusion speed corresponding to the high-density area is reduced from 5 mm / s to 3 mm / s, so that the strain rate is reduced from 0.1 / s to 0.06 / s to inhibit excessive twinning generation; the speed in the low-density area is increased from 5 mm / s to 7 mm / s, and the strain rate is increased to 0.14 / s to promote twinning nucleation.

[0112] Preset the strain rate distribution curve through finite element simulation (e.g., the strain rate linearly decreases from 0.1 / s to 0.05 / s along the extrusion direction), monitor the strain rate deviation caused by the extrusion speed in real time, and dynamically adjust the speed through a servo motor (accuracy ±0.1 mm / s) to make the twin density gradient converge along the preset path (e.g., the gradient decreases by 5% per 10 mm extrusion stroke).

[0113] Optimization of segmented loading of the strain path, specifically including: Identify the twin-sensitive area (e.g., for {10-12} twin activation, the angle between the principal stress and the c-axis needs to be >60°), and superimpose periodic torsional strain (±5° / s) in the later stage of extrusion to dynamically match the direction of the principal stress with the orientation of the twin band. For example, when simulation finds that the angle between the principal stress and the c-axis in the spoke area is only 45°, twist to increase the angle to 65° to activate the target twin system, and the uniformity of twin density is increased by 25%.

[0114] Segmented loading strategy: Divide the extrusion process into three stages: Initial stage (0-30% stroke): Axial strain is the main, speed is 6 mm / s, quickly fill the die; Intermediate stage (30%-70% stroke): Superimpose radial strain (strain ratio 0.3:1), speed is 4 mm / s, regulate the twin density in the core; Final stage (70%-100% stroke): Torsional strain and axial strain (angular rate 2° / s), speed is 3 mm / s, make the density of the surface layer and the core uniform.

[0115] Input the new parameters (e.g., temperature 350 °C, speed 3 mm / s, torsional angular rate 2° / s) into the dynamic evolution matrix to predict the twin density distribution after 200 s (e.g., at the root of the spoke ρ = 28%, at the edge of the rim ρ = 22%, gradient Δρ = 6%).

[0116] Calculation of the deviation convergence rate: Compare the predicted value with the target interval, and calculate the convergence rate = (initial deviation - remaining deviation) / initial deviation × 100%. For example, the initial high-density deviation is 15%, the predicted remaining deviation is 2%, and the convergence rate = (15 - 2) / 15 ≈ 87% (preset target ≥ 85%).

[0117] Dynamic matching and iteration: During the hot extrusion process, detect the actual twin density through on-line EBSD (e.g., the measured ρ at the root of the spoke is 30%, convergence rate 75% < 85%), trigger parameter iteration: further reduce the speed to 2.5 mm / s, increase the torsional angular rate to 3° / s, until the convergence rate reaches 88% to meet the requirements.

[0118] In the embodiments of the present invention, through temperature zone control and strain rate adjustment, the standard deviation of twin density is reduced from 12% to 5%. Especially in geometrically complex regions (such as the intersection of wheel hub spokes), the density gradient is reduced from 30% to 8%, avoiding the risk of brittle fracture caused by local over-density. For example, for a hot-extruded part of a chassis bracket, after optimization, the uniformity of twin density is increased by 40%, and the room-temperature elongation is increased from 9% to 13%, meeting the requirements of automotive lightweighting for the plasticity of materials.

[0119] The closed-loop feedback of the dynamic evolution matrix and on-line detection shortens the parameter adjustment lag time from 30 minutes of the traditional trial-and-error method to 5 minutes. During the extrusion process, it can respond in real time to die temperature fluctuations (such as a deviation of ±5°C) and billet microstructure changes (such as a grain size fluctuation of ±2μm). The segmented loading strategy reduces the process debugging cycle from 20 iterations to 5 times, and the R & D cost is reduced by 60%. It is especially suitable for the production of small-batch customized magnesium alloy components. Through strain path optimization, the twin texture can be directionally controlled (for example, the proportion of {10-12} twinning is reduced from 60% to 40%, and the {10-11} twinning is increased to 50%), reducing the strength difference between tensile and compressive loads of the material from 40 MPa to 15 MPa, and the anisotropy is reduced by 62.5%.

[0120] In another preferred embodiment of the present invention, the above-mentioned closed-loop control signal including the grain boundary slip compensation amount and the dynamic recrystallization threshold is generated according to the twin orientation distribution data and fatigue strength detection data collected in real time, wherein the dynamic recrystallization threshold is dynamically adjusted according to the orientation rotation parameter in the corrected dynamic evolution matrix, and may include: Through an on-line electron backscatter diffraction (EBSD) system, the surface of the specimen is scanned in real time with a spatial resolution of 100 nm to obtain the crystal orientation data (Euler angles φ1, Φ, φ2) of the twin bands and the orientation difference angle distribution. For example, during the hot extrusion process, an orientation distribution map is generated every 5 seconds to identify the spatial distribution ratio of {10-12} tensile twins (orientation difference angle 86°) and {10-11} compressive twins (orientation difference angle 56°).

[0121] Real-time detection of fatigue strength: Using ultrasonic fatigue detection technology, the fatigue crack growth rate (da / dN) and the stress intensity factor range (ΔK) of the specimen are measured in real time. For example, when it is detected that ΔK = 15 MPa√m and da / dN = 1×10⁻ 6 mm / cycle in a certain area, it is determined that this area enters the fatigue damage acceleration stage.

[0122] Calculation of grain boundary slip compensation amount Stress concentration assessment: Calculate the misorientation angle θ at the grain boundary according to the twin orientation distribution (for example, the misorientation θ between adjacent grains is 45°), and combine it with the strain energy density U (measured by EELS to be U = 0.8 MJ / m 3 ), and evaluate the grain boundary sliding resistance. The larger the misorientation angle and the higher the strain energy density, the more difficult the grain boundary sliding is, and the greater the compensation amount required.

[0123] Compensation amount generation logic: Preset compensation amount model: The compensation amount Δγ = k×θ×U, where k is a material constant (for example, for AZ31 magnesium alloy, k = 0.001 / °·MJ / m³). For example, when θ = 45° and U = 0.8 MJ / m³, Δγ = 0.001×45×0.8 = 0.036, that is, a shear strain of 0.036 needs to be introduced by adjusting the loading path to compensate for the insufficient grain boundary sliding.

[0124] Loading path adjustment: Convert the compensation amount into real-time loading instructions, such as superimposing a periodic shear strain of ±Δγ / 2 (frequency 1 Hz) in the principal stress direction, and dynamically apply it through a six-axis loading device to release the grain boundary stress concentration.

[0125] Dynamic recrystallization threshold adjustment: Orientation rotation parameter correlation: Extract the orientation rotation parameter α from the corrected dynamic evolution matrix (for example, α = 0.015 rad / s, indicating that the twin band rotates 0.9° per minute), and establish the mapping relationship between α and the dynamic recrystallization threshold Trec: Trec = T0 + ΔT×α, where T0 is the reference threshold (such as 300 °C), and ΔT is the temperature coefficient (such as 20 °C·s / rad).

[0126] Threshold dynamic calculation: When α = 0.015 rad / s, Trec = 300 + 20×0.015 = 300.3 °C; if α increases to 0.02 rad / s (twin rotation accelerates), then Trec = 300 + 20×0.02 = 300.4 °C, indicating that the temperature needs to be increased to activate recrystallization to offset the tissue inhomogeneity caused by orientation rotation.

[0127] Temperature control execution: Adjust the local temperature to Trec with an accuracy of ±1 °C through the heating element built into the mold. For example, in the spoke area with a high twin rotation rate, raise the mold temperature from 350 °C to 350.4 °C to promote dynamic recrystallization and refine the grains.

[0128] Closed-loop control signal generation and feedback: Multi-parameter fusion: Integrate the grain boundary slip compensation amount Δγ (such as 0.036) and the dynamic recrystallization threshold Trec (such as 300.4 °C) into a closed-loop control signal and transmit it to the extruder control system through industrial Ethernet. Example of signal format: [Δγ = 0.036, Trec = 300.4 °C, effective time = 2024-12-26 14:30:00].

[0129] Real-time feedback iteration: Compare the actual twin orientation distribution with the predicted value every 10 seconds. If the orientation difference angle deviation > 5° or the fatigue crack growth rate exceeds the threshold (such as da / dN > 5×10 -7 mm / cycle), automatically adjust the Δγ and Trec parameters to form a "detection - calculation - execution - re-detection" closed loop. For example, if the measured da / dN = 6×10 -7 mm / cycle, increase Δγ to 0.04 and at the same time increase Trec to 300.6 °C until the indicators meet the standards.

[0130] In the embodiment of the present invention, through the dynamic adjustment of the grain boundary slip compensation amount, the stress concentration coefficient Kt at the grain boundary is reduced from 3.2 to 2.1, and the fatigue crack initiation life is increased by 80%. After 50,000 cycles of loading, the crack length of a certain wheel hub product is only 0.2 mm (1.5 mm for the traditional process). The adaptive adjustment of the dynamic recrystallization threshold refines the grain size from 15 μm to 8 μm, and the fatigue crack growth rate da / dN is reduced from 2×10 -6 mm / cycle to 8×10 -7 mm / cycle, approaching the level of aluminum alloy.

[0131] The closed-loop control adjusts the twin ratio of {10-12} and {10-11} from 7:3 to 5:5, reduces the texture strength by 35%, reduces the material anisotropy index from 0.8 to 0.5, and reduces the difference between the tensile and compressive strengths from 50 MPa to 20 MPa. The linkage between the dynamic recrystallization threshold and the orientation rotation parameter reduces the standard deviation of the grain size from 4 μm to 1.5 μm, solving the two-way problems of "overheating coarsening" and "excessive undercooled twins" in the traditional process.

[0132] The system can respond in real time to the initial microstructure fluctuations of the billet (such as a change in grain size of ±2 μm). Through closed-loop adjustment, the qualified rate of twin density is increased from 70% to 98%, and the frequency of manual intervention is reduced by more than 90%. The online detection of fatigue performance replaces the traditional offline test, and the process debugging time is shortened from 72 hours to 8 hours, especially suitable for multi-variety and small-batch production scenarios (such as customized components for new energy vehicles).

[0133] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. Data analysis system for pre-twinned deformation of magnesium alloys for automobiles, characterized in that, Including: A multi-axial stress loading module for establishing a composite stress field model including vertical force, braking force, and lateral force, applying multi-axial proportional loads through a six-axis loading device, and generating a three-dimensional stress tensor dataset; A twin crystal evolution analysis module for calculating the nucleation probability density function of twin crystals based on the three-dimensional stress tensor dataset and constructing a dynamic evolution matrix including nucleation energy barrier, proliferation rate, and orientation rotation; A twin crystal correction module for setting a first target monitoring point and a second target monitoring point in the pre-twin crystal region, where the first target monitoring point is located at the intersection of grain boundaries and the second target monitoring point is located at the center of the twin crystal band, obtaining the orientation difference angle and strain energy density difference data of the two monitoring points in real time, calculating the covariance matrix of the change characteristics of the two monitoring points, and correcting the proliferation rate parameter and orientation rotation parameter in the dynamic evolution matrix according to the correlation coefficient of the covariance matrix to obtain a corrected dynamic evolution matrix; An optimization module for adjusting the hot extrusion process parameter group according to the twin crystal density gradient distribution output by the corrected dynamic evolution matrix, where the parameter group includes die temperature, extrusion speed, and strain path, and dynamically controlling the twin crystal density within a target range; A regulation module for generating a closed-loop control signal including grain boundary slip compensation amount and dynamic recrystallization threshold according to the twin crystal orientation distribution data and fatigue strength detection data collected in real time, where the dynamic recrystallization threshold is dynamically adjusted according to the orientation rotation parameter in the corrected dynamic evolution matrix.

2. The data analysis system for pre-twinned deformation of magnesium alloy for automobiles according to claim 1, wherein, Establishing a composite stress field model including vertical force, braking force, and lateral force, applying multi-axial proportional loads through a six-axis loading device, and generating a three-dimensional stress tensor dataset, including: Defining the mapping rules between stress components in each direction and load ratios based on the mechanical coupling relationship of vertical force, braking force, and lateral force, and generating an initial parameter set of the composite stress field model; According to the initial parameter set, synchronously applying a vertical direction compression load, a braking direction shear load, and a lateral direction tensile load through a six-axis loading device, and dynamically adjusting the loading amplitude and phase difference of each axis based on a preset proportional coefficient to make the composite stress field consistent with the stress distribution of the actual working conditions of the vehicle; Real-time collecting the spatial stress distribution data output by the six-axis loading device, and generating a three-dimensional stress tensor dataset including principal stress direction, shear stress amplitude, and equivalent stress density through tensor superposition operation to fuse stress components in each direction.

3. The magnesium alloy pre-twinning deformation data analysis system for automobiles according to claim 2, wherein Real-time collecting the spatial stress distribution data output by the six-axis loading device, and generating a three-dimensional stress tensor dataset including principal stress direction, shear stress amplitude, and equivalent stress density through tensor superposition operation to fuse stress components in each direction, including: Synchronously collecting the original spatial stress distribution data in the vertical direction, braking direction, and lateral direction through the strain sensors and force sensors built in the six-axis loading device; Performing filtering and noise reduction processing on the original data, and uniformly converting the stress components in each direction to a global three-dimensional reference coordinate system based on a pre-calibrated sensor coordinate system transformation matrix; Performing tensor superposition operation on the vertical compression stress component, braking shear stress component, and lateral tensile stress component in time series according to the stress components in the converted global coordinate system to generate an instantaneous three-dimensional stress tensor. Perform eigenvalue decomposition on the instantaneous three-dimensional stress tensor, determine the principal stress direction through the eigenvector corresponding to the maximum eigenvalue, and record its dynamic trajectory varying with the loading time to obtain the eigenvalue decomposition result; Based on the eigenvalue decomposition result, extract the difference between the intermediate eigenvalue and the minimum eigenvalue as the maximum shear stress amplitude, and synchronously calculate the distribution state of each shear stress plane; Substitute the second invariant of the instantaneous three-dimensional stress tensor into the von Mises equivalent criterion, calculate the equivalent stress density, and generate a continuously distributed equivalent stress density field in the three-dimensional space grid by Gaussian interpolation method; Correlate the principal stress direction, shear stress amplitude, and equivalent stress density according to the time stamp and spatial coordinates, and construct a three-dimensional stress tensor dataset with hierarchical indexing.

4. The data analysis system for pre-twinned deformation of magnesium alloy for automobiles according to claim 3, characterized in that Substitute the second invariant of the instantaneous three-dimensional stress tensor into the von Mises equivalent criterion, calculate the equivalent stress density, and generate a continuously distributed equivalent stress density field in the three-dimensional space grid by Gaussian interpolation method, including: Based on the components of the instantaneous three-dimensional stress tensor, solve the second invariant through the sum-of-squares operation of the stress deviator tensor, and correlate the time-space coordinates during the loading process; Substitute the second invariant into the von Mises equivalent criterion, calculate the scalar value of the three-dimensional equivalent stress density at the current loading moment, and mark its corresponding spatial position; In the three-dimensional space grid, with the marked spatial position as the interpolation node, perform spatial smoothing interpolation on the discrete equivalent stress density scalar values through the Gaussian kernel function according to the preset interpolation radius and weight function to generate a continuously distributed equivalent stress density field; Compare the equivalent stress density field with the preset twin nucleation energy threshold. If the equivalent stress density in the local area exceeds the threshold range, readjust the weight function parameters of the Gaussian interpolation until the density field matches the nucleation energy barrier parameters of the dynamic evolution matrix to obtain the corrected equivalent stress density field.

5. The data analysis system for pre-twinned deformation of magnesium alloy for automobiles according to claim 4, characterized in that, In the three-dimensional space grid, with the marked spatial position as the interpolation node, perform spatial smoothing interpolation on the discrete equivalent stress density scalar values through the Gaussian kernel function according to the preset interpolation radius and weight function to generate a continuously distributed equivalent stress density field, including: With the marked spatial position as the interpolation node, dynamically set the interpolation radius of each node according to the geometric characteristics and stress gradient distribution characteristics of the magnesium alloy specimen, where the interpolation radius in the high stress gradient area is smaller than that in the low gradient area; Based on the non-uniformity of the local stress distribution and the material anisotropy parameters, select the Gaussian kernel function weight coefficient that matches the stress state at the current interpolation node; Centered on the interpolation node, traverse the adjacent grid cells in the three-dimensional space grid, and calculate the Gaussian weighted average stress density value in each cell according to the weight coefficient and interpolation radius to complete the spatial smooth transition of the discrete scalar values; Map the Gaussian weighted average stress density value to the vertices of the three-dimensional grid, and connect the adjacent vertices through the bicubic spline interpolation algorithm to generate an equivalent stress density field with spatial continuity and differential continuity; Compare the equivalent stress density field after interpolation with the stress distribution data measured by the sensor. If the interpolation error in the local area exceeds the preset tolerance, adaptively reduce the interpolation radius until the density field meets the accuracy requirements of the dynamic evolution matrix, and generate a continuously distributed equivalent stress density field.

6. The data analysis system for pre-twinned deformation of magnesium alloy for automobiles according to claim 5, wherein Calculate the nucleation probability density function of twins based on the three-dimensional stress tensor dataset, and construct a dynamic evolution matrix including the nucleation energy barrier, proliferation rate, and orientation rotation, including: Based on the principal stress direction, shear stress amplitude, and equivalent stress density field in the three-dimensional stress tensor dataset, combined with the relationship between the crystal slip system of magnesium alloy and the critical resolved shear stress of twins, establish a nucleation probability density function to quantify the twinning nucleation tendency at different spatial positions and loading times; Calculate the nucleation energy barrier parameter in the local area according to the dynamic difference between the equivalent stress density field and the preset twinning nucleation energy threshold, and characterize the degree of energy accumulation required to overcome the lattice resistance; Based on the time series change rate of the shear stress amplitude and the stability of the principal stress direction, calibrate the twin proliferation rate parameter and correlate the spatial gradient distribution of the nucleation probability density function; Construct an orientation rotation parameter matrix by matching the dynamic trajectory of the principal stress direction with the magnesium alloy crystal orientation database to describe the crystallographic reorientation behavior during twin growth; Integrate the nucleation energy barrier parameter, proliferation rate parameter, and orientation rotation parameter matrix in the time and space dimensions to generate a dynamic evolution matrix.

7. The data analysis system for pre-twinned deformation of magnesium alloy for automobiles according to claim 6, wherein Set a first target monitoring point and a second target monitoring point in the pre-twin region. The first target monitoring point is located at the intersection of grain boundaries, and the second target monitoring point is located at the center of the twin band. Obtain the orientation difference angle and strain energy density difference data of the two monitoring points in real time, calculate the covariance matrix of the change characteristics of the two monitoring points, and correct the proliferation rate parameter and orientation rotation parameter in the dynamic evolution matrix according to the correlation coefficient of the covariance matrix to obtain the corrected dynamic evolution matrix, including: According to the grain boundary topology structure and twin band morphology characteristics in the pre-twin region, set a first target monitoring point at the intersection of grain boundaries and a second target monitoring point at the geometric center of the twin band, and calibrate the initial crystal orientation and strain energy density baseline value of the two monitoring points; Collect the crystal orientation difference angle data of the first target monitoring point and the second target monitoring point in real time, and obtain the local strain energy density difference data of the two monitoring points to form a time series monitoring dataset; Based on the monitoring dataset, extract the covariance components of the orientation difference angle change rate and the strain energy density change rate respectively, construct a covariance matrix reflecting the deformation coordination characteristics of the two monitoring points, and calculate the correlation coefficient quantifying the correlation strength between local and global deformations; If the correlation coefficient indicates that the deformation coordination between the grain boundary intersection and the center of the twin band is lower than the preset threshold, proportionally reduce the proliferation rate parameter in the corresponding area of the dynamic evolution matrix to inhibit the excessive growth of non-uniform twins caused by local stress concentration; Dynamically adjust the orientation rotation parameter in the dynamic evolution matrix according to the variance component of the orientation difference angle change rate in the covariance matrix to enhance the compensation weight of grain boundary slip to the crystallographic reorientation behavior of the twin band, so as to obtain the corrected dynamic evolution matrix.

8. The data analysis system for pre-twinned deformation of magnesium alloy for automobiles according to claim 7, characterized in that, According to the grain boundary topology and twin band morphology characteristics of the pre-twinned region, the first target monitoring point is set at the intersection of the grain boundaries, and the second target monitoring point is set at the geometric center of the twin band. The initial crystal orientation and strain energy density baseline values ​​of the two monitoring points are calibrated, including: Based on the equivalent stress density field and the twin nucleation probability density function in the three-dimensional stress tensor data set, the region with an equivalent stress density higher than a preset threshold is selected as a pre-twinning candidate region, and the final monitoring region is determined in combination with the hot spot analysis of the grain boundary topological structure; Scanning the grain boundary network of the monitoring area, identifying the geometric feature points where three or more grain boundaries intersect, and determining the corresponding intersection points as the first target monitoring points according to the stress concentration degree; In the monitoring area, according to the geometric symmetry of the twin band morphology and the strain energy density distribution gradient, the geometric center point of the longitudinal extension direction of the twin band is determined and used as the second target monitoring point; Perform initial crystal orientation scans on the two monitoring points to obtain initial Euler angle data under unloaded conditions, and verify the consistency of the crystal structure based on the magnesium alloy crystallography database; Under zero load condition, the initial local strain field of two monitoring points is measured, and the initial strain energy density value is calculated in combination with the material elastic modulus parameters; In the initial loading stage, the strain energy density change rate of the two monitoring points is monitored in real time to calibrate the initial crystal orientation and strain energy density baseline value of the two monitoring points.

9. The magnesium alloy pre-twinning deformation data analysis system for automobiles according to claim 8, characterized in that Based on the monitoring data set, the covariance components of the orientation misalignment angle change rate and the strain energy density change rate are extracted respectively, a covariance matrix reflecting the synergistic characteristics of the deformation of the two monitoring points is constructed, and the correlation coefficient of the correlation strength between the local and global deformation is calculated and quantified, including: Performing time window segmentation on the misorientation angle and strain energy density data in the monitoring data set, extracting the change rate sequence in each time window, and eliminating abnormal fluctuation data segments caused by sensor noise or transient interference through a statistical outlier detection algorithm; In the same time window, the covariance components between the orientation misalignment angle change rates of the first target monitoring point and the second target monitoring point, as well as the covariance components between the strain energy density change rates of the two monitoring points are calculated respectively, a two-dimensional covariance matrix is ​​constructed, and its non-diagonal elements are normalized; Based on the normalized covariance matrix, the Pearson correlation coefficient characterizing the deformation synergy between the grain boundary intersection and the twin zone center was calculated as a quantitative indicator of the influence of local deformation on the global twin evolution.

10. The magnesium alloy pre-twinning deformation data analysis system for automobiles according to claim 9, characterized in that According to the twin density gradient distribution output by the modified dynamic evolution matrix, the hot extrusion process parameter group is adjusted, and the parameter group includes the die temperature, the extrusion speed, and the strain path, so that the twin density is dynamically controlled within the target range, including: Based on the corrected dynamic evolution matrix, the twin density gradient distribution data of the current loading stage is extracted to identify the spatial position and deviation amplitude of the high-density and low-density areas beyond the target range; If the high-density area exceeds the threshold, the die temperature is reduced to inhibit the dynamic recrystallization rate. Conversely, the temperature is increased to promote twinning homogenization to obtain the adjusted die temperature. According to the local gradient change of the strain path, the extrusion speed is adjusted to control the strain rate distribution, so that the twinning density gradient converges along the preset path to obtain the adjusted extrusion speed. Based on the spatial correlation between the principal stress direction and the twinning sensitive area, the segmented loading strategy of the strain path is re-planned to optimize the stress distribution uniformity to obtain the adjusted strain path parameters. According to the adjusted die temperature, extrusion speed and strain path parameters, the corrected twinning density gradient distribution is predicted and compared with the spatial coverage of the target interval to calculate the deviation convergence rate. During the hot extrusion process, the actual twinning density data is collected in real time and dynamically matched with the predicted and corrected twinning density gradient distribution. If the deviation convergence rate in the local area does not meet the preset requirements, the process parameter group is iteratively optimized based on the change trend of the Pearson correlation coefficient until the twinning density is dynamically controlled within the target interval.

Images