Discrete element simulation method of implicit particle environment

Abstract

The invention provides a discrete element simulation method of an implicit particle environment, which comprises the following steps: S1, fully mixing an ore grinding medium and a ground material, taking the mixture as a whole, and measuring a stacking angle; s2, constructing an ore grinding medium discrete element model of implied particles; s3, constructing a discrete element model of the accumulation angle measuring device; and S4, parameter optimization and discrete element simulation. The maximum error between the collision energy spectrum obtained through simulation of the method and conventional simulation is only 3.3%, meanwhile, it is avoided that a large number of mineral particles are added into simulation calculation, a large number of calculation resources are saved, and energy, speed and other information of mutual collision between media in the ore grinding operation process can be rapidly and accurately obtained.

Description

Technical Field

[0001] The invention relates to the technical field of grinding discrete element simulation theory calculation, and in particular to a discrete element simulation method of an implicit particle environment. Background Art

[0002] Discrete element method (DEM) is widely used in the field of grinding. Combined with the simulation software EDEM (Engineering-Discrete Element Method), it can simulate the indicators required for grinding operations, such as mill power consumption, particle collision energy spectrum, and collision energy loss. In the simulation process of grinding operations, the accurate calibration of ore contact characteristic parameters is the key to revealing the material motion characteristics through simulation experiments. Only when the DEM particles have a real shape, the contact parameters are physically reasonable. The collision energy spectrum obtained by simulation can often be combined with the ore fragmentation energy to provide a theoretical basis for setting different grinding media ratios, so as to achieve the purpose of saving mill energy consumption and preventing over-grinding.

[0003] For fine grinding operations (micron-level and finer materials), the fine particle size of the material makes it difficult and inaccurate to obtain intrinsic parameters such as particle recovery coefficient and Poisson's ratio through drop tests and uniaxial compression tests. In addition, conventional modeling auxiliary methods such as 3D laser scanning, automatic particle filling model, and spherical harmonics are no longer able to accurately obtain their morphological structure. Secondly, the fine mill feed in fine grinding operations causes hundreds of millions of particles to participate in the calculation during simulation, which requires a lot of calculations. At present, the accurate acquisition of particle contact parameters, the fineness of particles and their huge number are still the main obstacles in discrete element simulation. Summary of the invention

[0004] The main purpose of the present invention is to provide a discrete element simulation method for an implicit particle environment, so as to achieve the purpose of quickly and accurately obtaining the material movement state during the grinding operation without distortion of the simulation results, thereby saving computing resources.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a discrete element simulation method of an implicit particle environment, the method comprising: S1. Fully mix the grinding media and the material to be ground and treat them as a whole, and measure the stacking angle; S2, constructing a discrete element model of grinding media with implicit particles; S3, constructing a discrete element model of a stacking angle measurement device; S4, parameter optimization, discrete element simulation.

[0006] In the preferred embodiment, when measuring the stacking angle in step S1, a semi-theoretical method is used to formulate a ball loading system for coarse grinding, and a ball loading method based on grinding dynamics is used to formulate a ball loading system for fine grinding.

[0007] In the preferred embodiment, in step S2, after modeling with the aid of EDEM software, the particle size distribution is set with the particle size upper limit as the particle size of particles distributed in the particle size range, and the particle size distribution of the ground material and the grinding medium is set according to the screening test results.

[0008] In the preferred embodiment, when constructing the model in step S3, a physical model of the stacking angle measuring device can be established in COMSOL Multiphysics software according to the actual size, and the physical model can be meshed using free tetrahedrons; Considering the simple structure of the two models, the meshing quality is set to “fine” and the model calibration is set to ordinary physics. After the physical model is established, it is saved as an STL text file and imported into the “Geometries” module in EDEM to complete the construction of the discrete element simulation model of the stacking angle measurement device.

[0009] In the preferred embodiment, the constructed physical model can ignore non-essential components and simplify the model.

[0010] In the preferred embodiment, during the parameter optimization in step S4, the EDEM parameter calibration toolkit is used to perform parameter calibration of only the grinding medium, and the actual measured value of the mixed stacking angle of the grinding medium and the material to be ground is used as the evaluation index. The reverse calibration procedure is executed, and the parameter value is iteratively changed until it meets the actual measured stacking angle response. At this time, there are no particles of the material to be ground in the simulation environment, but the calibrated contact parameters contain the contact information of the particles of the material to be ground with the grinding medium as the carrier.

[0011] In the preferred embodiment, discrete element simulation is performed with optimized contact parameters, and ultimately the purpose of performing discrete element simulation research only with grinding media and calibrated parameters to characterize the actual operating state of the grinding operation is achieved.

[0012] In the preferred embodiment, the specific steps of measuring the stacking angle are: A1. Use a high-definition digital camera to collect data from multiple angles around the material pile images to obtain visual information of the material stacking status.

[0013] A2. For each collected image , using a custom grayscale function Convert it to a grayscale image and then use the binarization function Convert the grayscale image to a binary image. The binary formula is: ,in It is a threshold value adaptively determined according to the grayscale histogram of the image. This formula converts the grayscale image into a binary image, which is convenient for subsequent annotation of the stacking angle. A3. Mark the stacking angle on the binary image using an algorithm based on contour detection and angle calculation Marking stacking angle , the formula is: , Used to get the value of the stacking angle from the processed image.

[0014] In the preferred embodiment, the specific steps of parameter optimization are: B1. Calculation and collection Angle of accumulation The average As the measured stacking angle value, It is expressed as: ,average value Used to combine multiple measurement angles to obtain a stacking angle value representing the overall stacking characteristics; B2. According to the material properties and empirical formula, set the initial range of contact parameters related to the stacking angle. , setting range is arrive ; For the static friction coefficient , setting range is arrive ; For the rolling friction coefficient , setting range is arrive ; B3. Using nonlinear regression model To describe the stacking angle Contact parameters , , The relationship between them is: ,in It is a coefficient obtained by fitting a large amount of experimental data, which establishes a complex nonlinear relationship between the stacking angle and the contact parameters for subsequent parameter optimization; B3. Use genetic algorithm To optimize the contact parameters, in the genetic algorithm, the fitness function is defined as , with the goal of maximizing the fitness function value, the optimal contact parameter combination is iteratively searched through genetic operations such as crossover and mutation. The fitness function is used to evaluate the pros and cons of each individual in the genetic algorithm. The higher the fitness, the better the fit with the actual stacking angle; B4. Combine the optimized contact parameters Substitute into the model The simulated stacking angle is calculated , calculation error ,if Less than the set threshold , then the optimized parameters are considered accurate, formula: Used to measure the accuracy of the optimized parameter simulation stacking angle.

[0015] The present invention provides a discrete element simulation method for an implicit particle environment. The present invention successfully characterizes the operating state of an actual grinding operation (including material particles) by "using grinding media as a carrier to contain contact information of ground material particles, and then conducting discrete element simulation research only with grinding media and calibrated parameters (implicit material particles)". The maximum error between the simulation results of material particles and the simulation results of implicit material particles is only 3.3%. At the same time, it avoids adding a large number of mineral particles to the simulation calculation, saving a lot of computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is the main structural diagram of the cleaning process of the present invention; Figure 2 It is the overall appearance structure diagram of the present invention; Figure 3 It is a structural diagram of the ejection protection shell of the ejection device of the present invention; Figure 4 It is a disassembled structural diagram of the cover plate of the general protective shell of the present invention. DETAILED DESCRIPTION

[0017] Example 1 like Figures 1 to 4 As shown, a discrete element simulation method for an implicit particle environment includes: S1. Fully mix the grinding media and the material to be ground and treat them as a whole, and measure the stacking angle; S2, constructing a discrete element model of grinding media with implicit particles; S3, constructing a discrete element model of a stacking angle measurement device; S4, parameter optimization, discrete element simulation.

[0018] In the preferred embodiment, when measuring the stacking angle in step S1, a semi-theoretical method is used to formulate a ball loading system for coarse grinding, and a ball loading method based on grinding dynamics is used to formulate a ball loading system for fine grinding.

[0019] In the preferred embodiment, in step S2, after modeling with the aid of EDEM software, the particle size distribution is set with the particle size upper limit as the particle size of particles distributed in the particle size range, and the particle size distribution of the ground material and the grinding medium is set according to the screening test results.

[0020] In the preferred embodiment, when constructing the model in step S3, a physical model of the stacking angle measuring device can be established in COMSOL Multiphysics software according to the actual size, and the physical model can be meshed using free tetrahedrons; Considering the simple structure of the two models, the meshing quality is set to “fine” and the model calibration is set to ordinary physics. After the physical model is established, it is saved as an STL text file and imported into the “Geometries” module in EDEM to complete the construction of the discrete element simulation model of the stacking angle measurement device.

[0021] In the preferred embodiment, the constructed physical model can ignore non-essential components and simplify the model.

[0022] In the preferred embodiment, during the parameter optimization in step S4, the EDEM parameter calibration toolkit is used to perform parameter calibration of only the grinding medium, and the actual measured value of the mixed stacking angle of the grinding medium and the material to be ground is used as the evaluation index. The reverse calibration procedure is executed, and the parameter value is iteratively changed until it meets the actual measured stacking angle response. At this time, there are no particles of the material to be ground in the simulation environment, but the calibrated contact parameters contain the contact information of the particles of the material to be ground with the grinding medium as the carrier.

[0023] In the preferred embodiment, discrete element simulation is performed with optimized contact parameters, and ultimately the purpose of performing discrete element simulation research only with grinding media and calibrated parameters to characterize the actual operating state of the grinding operation is achieved.

[0024] In the preferred embodiment, the specific steps of measuring the stacking angle are: A1. Use a high-definition digital camera to collect data from multiple angles around the material pile images to obtain visual information of the material stacking status.

[0025] A2. For each collected image , using a custom grayscale function Convert it to a grayscale image and then use the binarization function Convert the grayscale image to a binary image. The binary formula is: ,in It is a threshold value adaptively determined according to the grayscale histogram of the image. This formula converts the grayscale image into a binary image, which is convenient for subsequent annotation of the stacking angle. A3. Mark the stacking angle on the binary image using an algorithm based on contour detection and angle calculation Marking stacking angle , the formula is: , Used to get the value of the stacking angle from the processed image.

[0026] In the preferred embodiment, the specific steps of parameter optimization are: B1. Calculation and collection Angle of accumulation The average As the measured stacking angle value, It is expressed as: ,average value Used to combine multiple measurement angles to obtain a stacking angle value representing the overall stacking characteristics; B2. According to the material properties and empirical formula, set the initial range of contact parameters related to the stacking angle. , setting range is arrive ; For the static friction coefficient , setting range is arrive ; For the rolling friction coefficient , setting range is arrive ; B3. Using nonlinear regression model To describe the stacking angle Contact parameters , , The relationship between them is: ,in It is a coefficient obtained by fitting a large amount of experimental data, which establishes a complex nonlinear relationship between the stacking angle and the contact parameters for subsequent parameter optimization; B3. Use genetic algorithm To optimize the contact parameters, in the genetic algorithm, the fitness function is defined as , with the goal of maximizing the fitness function value, the optimal contact parameter combination is iteratively searched through genetic operations such as crossover and mutation. The fitness function is used to evaluate the pros and cons of each individual in the genetic algorithm. The higher the fitness, the better the fit with the actual stacking angle; B4. Combine the optimized contact parameters Substitute into the model The simulated stacking angle is calculated , calculation error ,if Less than the set threshold , then the optimized parameters are considered accurate, formula: Used to measure the accuracy of the optimized parameter simulation stacking angle.

[0027] Example 2 Further illustrate with reference to Example 1, Figure 1 -6, step 1: measuring the accumulation angle of the material to be ground, the material to be ground is magnetite as an example; The material to be ground is fully mixed, and the angle of accumulation of the material to be ground is measured by means of an angle of accumulation measuring device. The angle of accumulation measuring device includes a funnel, a support plate, a storage container, a level and other components. The dimensions of the funnel are an upper bottom radius of 4.2 cm, a lower bottom radius of 1.8 cm, and a height of 4 cm. The effective volume of the storage container is 20 cm*4.4 cm*15 cm, and the material is an acrylic plate with a thickness of 0.3 cm. The level is used to calibrate the operating platform.

[0028] Further, 880 g of magnetite ore was taken, mixed thoroughly and then slowly added in small amounts from the top of the funnel. After the stacking shape of the magnetite ore particles was stable, a high-definition digital camera was used to capture the material stacking state diagram, and the image was grayscaled and binarized using Matlab software. The stacking angle of the binarized image was marked, and the measurement was repeated 5 times to obtain an average value of the iron ore stacking angle of 32.7° with a standard deviation of 0.15°, that is, the measured value of the static stacking angle of the magnetite ore particles was 32.7°±0.15°.

[0029] Step 2: Measure the stacking angle of the grinding medium and the particles, taking steel balls as an example; The magnetite ore particles and steel balls are fully mixed, and the implicit particle stacking angle is measured with the help of a lifting cylinder device. The lifting cylinder device includes a fixed pulley, a lifting line, a lifting cylinder, a loading platform, a speed controller, a power supply, an electric motor, and a driver. The electric motor lifting speed is set to 3mm / s, and the magnetite ore and steel balls are fully mixed and placed in the lifting cylinder. The lifting cylinder is slowly and uniformly lifted by the electric motor until the material is completely separated from the lifting cylinder. After the shape of the material accumulation body is stable, repeat the measurement 5 times and take the average value as the static stacking angle of the material.

[0030] Furthermore, 880 g of magnetite ore and 47, 37 and 12 φ15 cm, φ20 cm and φ25 cm steel balls were taken, and the magnetite ore and steel balls were fully mixed and placed in the lifting cylinder to measure their stacking angles. The measured mixed stacking angle of the ground material and the grinding medium was 26.4°, and the standard deviation was 0.21, that is, the actual measured value of the static stacking angle of the implicit particles was 26.4°±0.21°.

[0031] Preferably, it is reliable enough to formulate the ball loading system by using the semi-theoretical method for coarse grinding (balanced grading is used for grading), and it is reliable enough to formulate the ball loading system by using the ball loading method based on grinding dynamics for fine grinding.

[0032] Preferably, the lifting speed of the electric motor is as small as possible to ensure that the material falls slowly and naturally, and the resulting stacking angle can truly reflect the contact parameters between the material particles.

[0033] Preferably, when measuring the stacking angle, a high-definition digital camera is used to horizontally surround the material pile at 90° intervals, and a total of 4 material stacking state diagrams are collected, and the material stacking angles are marked, and the average of the 4 stacking angles is taken as the actual stacking angle of this measurement.

[0034] Step 3: Construct discrete element models of magnetite particles and steel balls; Construct a discrete element model of magnetite particles and steel balls in EDEM. The steel balls have regular shapes, so they are modeled as a single sphere in EDEM, and 47 Ø15cm steel balls, 37 φ20cm steel balls, and 12 φ25cm steel balls are constructed using φ15cm as the benchmark model. In this scheme, considering that the particle shape has a strong influence on the calculation amount, but has a weak influence on its own energy dissipation, stiffness, density, and porosity, the iron ore model is also constructed with a single sphere and the particle size distribution is set.

[0035] Preferably, the upper limit of the particle size is taken as the particle size of the particles distributed in the particle size range (e.g., the particle size of the particles distributed in the particle size range of 0-2.36 mm is taken as 2.36 mm), and the iron ore particle and steel ball model and its particle size distribution are constructed according to the screening test results. The following table 1 shows the setting details of the particle size distribution of the ground material and the grinding medium in EDEM:

[0036] Preferably, contact parameters are physically reasonable only if the DEM particles have realistic shapes.

[0037] Step 4: Constructing a discrete element model of the stacking angle measurement device; In EDEM, the physical models of the implicit particle stacking angle measurement device and the iron ore stacking angle measurement device were established in COMSOL Multiphysics software according to the actual size, and the free tetrahedron was used to mesh the physical models. Considering the simple structure of the two models, the meshing quality was set to "fine" and the model calibration was set to ordinary physics. After the physical model was established, it was saved as an STL text file and imported into the "Geometries" module in EDEM to complete the construction of the discrete element simulation model of the stacking angle measurement device.

[0038] It should be noted that in order to reduce the amount of calculation, the constructed model is a simplified model that ignores non-essential components such as the support plate and motor, which does not affect the simulation results.

[0039] Example 3 The present invention will be further described below in conjunction with specific implementation process: The EDEM parameter calibration toolkit is used to calibrate the parameters of the grinding medium only. The actual measured value of the mixed stacking angle of the grinding medium and the ground material is used as the evaluation index. The reverse calibration procedure is performed to iteratively change the parameter value until it meets the actual measured stacking angle response. The parameter optimization is performed and compared to verify the correctness of the implicit particle discrete element simulation method.

[0040] Step 1: Parameter optimization Based on the universal material model database of EDEM, the reference range of contact parameters of magnetite particles can be obtained by inputting the simulation scale, material stacking density and stacking angle: collision recovery coefficient is 0~0.75; static friction coefficient is 0.2~1.16; rolling friction coefficient is 0~0.5. Using Design-Expert software, based on the principle of orthogonal rotation combination experiment, the collision recovery coefficient ( X 1) Static friction coefficient ( X 2) and rolling friction coefficient ( X 3) For the experimental variables, establish an orthogonal experimental table and conduct parameter simulation tests.

[0041] For the simulation static stacking angle ( Y 1) The variance analysis results showed that F = 34.85, P < 0.0001, and the regression model was extremely significant. The collision recovery coefficient ( X 1) Static friction coefficient ( X 2) Rolling friction coefficient ( X 3) and the static stacking angle ( Y 1) The actual value regression equation, the determination coefficient R2=0.96, the corrected determination coefficient R 2adj =0.93, indicating that the regression equation has a good fit. The equation can be used to analyze the experimental results and then the response surface diagram can be drawn based on the equation.

[0042] Actual measurement value of static stacking angle Y 1=32.7° is substituted into the regression equation, and the response surface method is used to Y 1 is the evaluation index to further optimize each parameter and obtain the optimal parameter combination: X 1=0.375; X 2=0.68; X3=0.25, at this time, the calculated simulation value of the static stacking angle of magnetite particles is 32.48°. In order to verify the accuracy of the optimal parameter combination, the above parameter values ​​were used for simulation verification. The average static stacking angle obtained from 5 repeated simulation verification experiments was 32.3°, and the standard deviation was 0.13°, that is, the simulated static stacking angle was 32.3°±0.13°. Compared with the actual measured value of the static stacking angle of magnetite particles (32.7°±0.15°), the error between the two was calculated to be 0.7%, indicating that the optimized contact parameters are more accurate. Therefore, the collision recovery coefficient between magnetite particles is determined to be 0.375, the static friction coefficient is 0.68, and the rolling friction coefficient is 0.25. The other parameters are determined in the same way.

[0043] Step 2: Verification of implicit particle discrete element simulation method In order to verify the correctness of the implicit particle environment discrete element simulation method, the movement process of finely ground magnetite was simulated by using two methods: actual contact parameters and implicit particle environment discrete element simulation.

[0044] During one cycle of the mill operation (a complete cycle is 3-3.9s after the mill runs smoothly), the two simulation methods obtain the total energy loss, normal energy loss and tangential energy loss caused by the collision of steel balls in the mill, and the relationship between the total energy loss, normal energy loss and tangential energy loss over time (collision energy spectrum) remains highly consistent.

[0045] Furthermore, the collision energy is divided into 10 -9 J. 10 -8 J. 10 -7 J. 10 -6 J. 10 -5 J. 10 -4 J. 10 -3 J, 0.01 J, 0.1 J, and 10 -9 ~ 10 -6 J. 10 -6 ~ 10 -3 J. 10 -3 ~ 0.1 J three energy level intervals.

[0046] Furthermore, the energy spectrum includes three indicators: total energy loss, normal energy loss, and tangential energy loss caused by steel ball collision within a rotation period. The total energy loss, normal energy loss, and tangential energy loss all have strong peaks, and their distribution is similar to Gaussian distribution. Among them, the maximum collision energy level is about 0.07J and the collision frequency is low. From the perspective of total energy loss, the main collision energy of ceramic balls and steel balls is concentrated in 10 -6 ~ 10 -3 J energy level interval, in the low energy level interval 10 -9 ~10 -6 J and high energy interval 10-3 The energy consumed by ~ 0.1 J is relatively small. The collision energy spectrum shows that the total energy loss, normal energy loss and tangential energy loss generated by the steel ball in the environment with particles and implicit particles are highly consistent, with a maximum error of only 3.3%, indicating that the medium collision energy spectrum obtained by the discrete element simulation method of the implicit particle environment can characterize the collision energy between media in the actual simulation process, and then study the impact of the medium on the ore particles. The following Table 2 shows the details of the medium collision energy distribution obtained by simulation in the environment with particles and implicit particles of the present invention.

[0047]

[0048] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A discrete element simulation method for an implicit particle environment, characterized by: The method includes: S1. Fully mix the grinding media and the material to be ground and treat them as a whole, and measure the stacking angle; S2, constructing a discrete element model of grinding media with implicit particles; S3, constructing a discrete element model of a stacking angle measurement device; S4, parameter optimization, discrete element simulation.

2. The discrete element simulation method for an implicit particle environment according to claim 1, characterized in that: When measuring the stacking angle in step S1, a semi-theoretical method is used to formulate a ball loading system for coarse grinding, and a ball loading method based on grinding dynamics should be selected for fine grinding to formulate a ball loading system.

3. The discrete element simulation method for an implicit particle environment according to claim 1, characterized in that: Step S2: After modeling with the aid of EDEM software, the particle size distribution is set with the upper limit of the particle size as the particle size of particles distributed in the particle size range, and the particle size distribution of the ground material and the grinding medium is set according to the screening test results.

4. The discrete element simulation method for an implicit particle environment according to claim 1, characterized in that: When constructing the model in step S3, a physical model of the stacking angle measuring device can be established in COMSOL Multiphysics software according to the actual size, and the physical model is meshed using free tetrahedrons; Considering the simple structure of the two models, the meshing quality is set to "fine" and the model calibration is set to general physics. After the physical model is established, it is saved as an STL text file and imported into the "Geometries" module in EDEM to complete the construction of the discrete element simulation model of the stacking angle measurement device.

5. The discrete element simulation method for an implicit particle environment according to claim 1, characterized in that: The constructed physical model ignores non-essential components and simplifies the model.

6. The discrete element simulation method for an implicit particle environment according to claim 1, characterized in that: During the parameter optimization in step S4, the EDEM parameter calibration toolkit is used to calibrate the parameters of only the grinding medium. The actual measured value of the mixed stacking angle of the grinding medium and the material to be ground is used as the evaluation index. The reverse calibration procedure is executed to iteratively change the parameter value until it meets the actual measured stacking angle response. At this time, there are no particles of the material to be ground in the simulation environment, but the calibrated contact parameters contain the contact information of the particles of the material to be ground with the grinding medium as the carrier.

7. The discrete element simulation method for an implicit particle environment according to claim 6, characterized in that: Discrete element simulation is performed with the optimized contact parameters, and ultimately the purpose of conducting discrete element simulation research only with grinding media and calibrated parameters is achieved to characterize the actual operating status of the grinding operation.

8. The discrete element simulation method for an implicit particle environment according to claim 6, characterized in that: The specific steps to measure the stacking angle are: A1. Use a high-definition digital camera to collect data from multiple angles around the material pile images to obtain visual information of the material accumulation status; A2. For each collected image , using a custom grayscale function Convert it to a grayscale image and then use the binarization function Convert the grayscale image to a binary image. The binary formula is: ,in It is a threshold value adaptively determined according to the grayscale histogram of the image. This formula converts the grayscale image into a binary image, which is convenient for subsequent annotation of the stacking angle. A3. Mark the stacking angle on the binary image using an algorithm based on contour detection and angle calculation Marking stacking angle , the formula is: , Used to get the value of the stacking angle from the processed image.

9. The discrete element simulation method for an implicit particle environment according to claim 6, characterized in that: The specific steps of parameter optimization are: B1. Calculation and collection Angle of accumulation The average As the measured stacking angle value, It is expressed as: ,average value Used to combine multiple measurement angles to obtain a stacking angle value representing the overall stacking characteristics; B2. According to the material properties and empirical formula, set the initial range of contact parameters related to the stacking angle; for the collision recovery coefficient , setting range is arrive ; For the static friction coefficient , setting range is arrive ; For the rolling friction coefficient , setting range is arrive ; B3. Using nonlinear regression model To describe the stacking angle Contact parameters , , The relationship between them is: ,in It is a coefficient obtained by fitting a large amount of experimental data, which establishes a complex nonlinear relationship between the stacking angle and the contact parameters for subsequent parameter optimization; B3. Use genetic algorithm To optimize the contact parameters, in the genetic algorithm, the fitness function is defined as , with the goal of maximizing the fitness function value, the optimal contact parameter combination is iteratively searched through genetic operations such as crossover and mutation. The fitness function is used to evaluate the pros and cons of each individual in the genetic algorithm. The higher the fitness, the better the fit with the actual stacking angle; B4. Combine the optimized contact parameters Substitute into the model The simulated stacking angle is calculated , calculation error ,if Less than the set threshold , then the optimized parameters are considered accurate, formula: Used to measure the accuracy of the optimized parameter simulation stacking angle.

Images