EDEM multi-mesoscopic parameter model generation and processing method

Abstract

The invention provides an EDEM multi-mesoscopic parameter model generation and processing method. The method comprises the steps that (1) an impact penetration simulation model is constructed; (2) reading an initial HDF5 file generated by the impact penetration simulation model, adjusting contact parameters and generating a new simulation model; (3) generating. Bat script files for batch operation and calculation through python codes of python software, performing batch operation on new simulation models, acquiring a simulation result of each time step length by utilizing an EDEMpy tool library, performing secondary processing on the acquired data, and drawing a comparison curve of each group of simulation results; and (4) integrating a code file. Py generated by python software into a Gu i interface, and packaging the code file. Py into a. Exe program file. According to the method, the functions of simultaneously and automatically generating and operating a large number of different parameter models and automatically performing data secondary processing and comparison can be realized, and the problem that the EDEM lacks an automatic multi-group parameter optimization tool when multi-group parameter optimization work is carried out can be effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of discrete element simulation, and particularly relates to a method for generating and processing an EDEM multi-mesoscopic parameter model. Background Art

[0002] Discrete element simulation (DEM, Discrete Element Method) is a numerical calculation method used to simulate and analyze particulate matter, powder flow, and the interaction between particles. It treats particles as discrete objects and precisely simulates the motion and behavior of particles under different working conditions by calculating the mechanical interactions between particles (such as collision, friction, adhesion, etc.), thereby revealing the macro-micro connections of the particle system. This method is widely used in fields such as civil engineering, mining, chemical engineering, food processing, and pharmaceuticals. Especially when designing and optimizing equipment and processes related to particulate matter, it can provide very intuitive and detailed simulation results.

[0003] Currently, the commonly used discrete element simulation software mainly includes PFC and EDEM. Among them, PFC is mainly based on command streams and performs excellently in particulate mechanics simulation, especially suitable for geotechnical engineering and mining fields. PFC has high precision in dealing with the interaction and mechanical properties of particles, but its computing performance is more dependent on hardware configuration, and it may encounter computing efficiency bottlenecks in the simulation of large-scale particle systems. In contrast, EDEM has stronger user-friendliness and an efficient computing engine, supports seamless integration with other engineering simulation software (such as CFD, FEM), can more efficiently handle the complex interaction between particles and fluids, and is suitable for multi-field coupling analysis. At the same time, with its strong parallel computing and cloud computing support, EDEM shows significant computing advantages in large-scale simulations and has become an important tool in the field of particulate material processing, and its functions and performance are difficult to replace by other discrete element software.

[0004] Although the existing EDEM has a fast and friendly workflow and provides a rich result data extraction function, it still faces some challenges when performing multi-group parameter optimization, especially in scenarios where multiple parameters need to be adjusted simultaneously to match the simulation results with experimental data. First, EDEM lacks an automated multi-group parameter optimization tool. Its built-in optimization function is usually only applicable to single-parameter or relatively simple optimization tasks. When multiple parameters are involved, manual adjustment and multiple experiments are often required, which not only makes the workflow cumbersome but also easily introduces human errors. Second, although EDEM supports data extraction, the interface and operations may become relatively complex when performing complex multi-parameter optimization. Especially when a large number of parameter adjustments, secondary data processing, and result comparison are required, the difficulty and complexity of the operations increase significantly.

[0005] At present, developing software that integrates three automated functions of multi-parameter optimization, data secondary processing, and result comparison based on the EDEM platform is still a blank that urgently needs to be filled. Summary of the Invention

[0006] In view of the problems mentioned in the above background technology, such as when carrying out multi-group parameter optimization work, the EDEM software cannot simultaneously automatically generate and run a large number of different parameter models, and the problems of not solving the functions of automatic data secondary processing and comparison, etc., the present invention proposes an EDEM multi-mesoscopic parameter model generation and processing method, which has a reasonable concept. By integrating the development of the graphical user interface and post-processing functions on the original EDEM platform, it not only meets the functions of automatically adjusting contact parameters in batches, generating and running calculation examples, etc., but also simultaneously meets the secondary processing and correction of data such as cone tip resistance and acceleration of the impact penetration simulation model, and outputs a curve graph of the processed results for convenient result verification.

[0007] To solve the above technical problems, an EDEM multi-mesoscopic parameter model generation and processing method provided by the present invention mainly includes the following steps:

[0008] (1) First, construct an impact penetration simulation model with the same size as the indoor impact penetration test;

[0009] (2) Read the initial HDF5 file generated by the impact penetration simulation model constructed in the above step (1) through the EDEMpy tool library of the python software, adjust the contact parameters, and generate a new simulation model;

[0010] (3) Generate a.bat script file for batch operation and calculation through the python code of the python software, batch run the new simulation models generated in step (2), use the EDEMpy tool library to obtain the simulation results of each time step, and perform secondary processing on the obtained data and draw comparison curves of each group of simulation results;

[0011] (4) Integrate the code file.py generated by the python software into the Gui interface and package it into an.exe program file to realize the integration of three automated functions of multi-parameter optimization, data secondary processing, and result comparison.

[0012] For the EDEM multi-mesoscopic parameter model generation and processing method, wherein the specific process of constructing the basic impact penetration simulation model in the above step (1) is as follows:

[0013] (1.1) According to the indoor impact penetration test, create granular materials and set the intrinsic parameters of the granular materials;

[0014] (1.2) Import the geometric model generated by the external modeling software into the EDEM software to generate a cylinder that is consistent with the externally generated geometric model for use in generating particles. Use the particle factory function of the EDEM software to generate particles layer by layer;

[0015] (1.3) Remove the unbalanced force between particles through stress release, so that the particles are in a balanced state;

[0016] (1.4) Import the impact penetration probe geometry model into the EDEM software and generate the basic impact penetration simulation model “edem-base.dem” to carry out the impact penetration simulation test.

[0017] The EDEM multi-micro-parameter model generation and processing method, wherein the specific process of step (2) is: the EDEMpy tool library is used as the Python API interface of the EDEM software, and the HDF5 file generated by the EDEM software simulation is accessed through the python software and the python library developed for EDEM to adjust and read the data of the particle or impact penetration probe; then the initial HDF5 file generated by the impact penetration simulation model in step (1) is read through the EDEM tool library of the python software, and the parent folder where the basic impact penetration simulation model is located is copied in the same directory to generate the required number of multiple groups of identical folders, one folder represents a model file, and the micro-parameters of the particle body and geometric body of the simulation model in each model file are adjusted respectively to generate multiple groups of new simulation models in batches.

[0018] The EDEM multi-micro-parameter model generation and processing method, wherein the specific process of step (3) is:

[0019] (3.1) Generate a batch calculation .bat file through the python code of the python software to batch run the new simulation model generated in step (2);

[0020] (3.2) The attitude transformation matrix of the penetrometer is directly obtained through the EDEMpy tool library, and the cone tip resistance, axial acceleration, inclination angle and effective penetration depth are quickly obtained by coordinate transformation;

[0021] (3.3) The results of the new simulation model generated in the above step (3.1) are exported as a .CSV file through the python script code of the python software, and the time history curves of cone tip resistance, axial acceleration and vertical penetration depth are plotted, and the time history curves are compared with the impact penetration indoor test results.

[0022] The EDEM multi-micro-parameter model generation and processing method, wherein the process of obtaining the posture transformation matrix in step (3.2) is:

[0023] ① First, introduce the simulation results (.h5 files) of each model folder in the EDEMpy tool library h5py;

[0024] ② Then, directly read the attitude transformation matrix Orientation of the penetrometer through the python code:

[0025]

[0026] Among them, G represents the global coordinate; b represents the carrier coordinate; X G , Y G , Z G respectively represent the vector magnitudes in the X, Y, and Z directions in the global coordinate system; X b , Y b , Z b respectively represent the vector magnitudes in the X, Y, and Z directions in the global coordinate system.

[0027] In the method for generating and processing the EDEM multi-mesoscopic parameter model, where the solution process of the cone tip resistance in the step (3.2) is as follows:

[0028] ① First, obtain the acting vector

[0029] Extract the acting vector from the simulation:

[0030]

[0031] Among them, A x , A y , A z respectively represent the force vector magnitudes in the x, y, and z directions in the global coordinate; A represents the force vector on the penetrometer;

[0032] ② Obtain the local Z-axis direction

[0033] Extract the third column from the transformation matrix of the geometry as the Z-axis direction vector:

[0034]

[0035] Among them, z x , z y , z z respectively represent the acceleration vector magnitudes in the x, y, and z directions in the global coordinate; z_axis represents the axial acceleration of the penetrometer;

[0036] ③ Normalize the Z-axis direction vector

[0037] Calculate the vector norm:

[0038]

[0039] Normalize to obtain a unit vector:

[0040]

[0041] ④ Calculate the dot product to obtain the axial tip resistance

[0042] Project vector A onto the unit vector:

[0043]

[0044] The method for generating and processing the EDEM multi-mesoscopic parameter model, wherein the solution process of the axial acceleration in step (3.2) is as follows:

[0045] ① Obtain the velocity vector

[0046] Extract the velocity vectors of the geometric body at the current time step and the next time step from the simulation:

[0047]

[0048] Wherein, V now 、V next represent the velocity vector at the current time step and the adjacent next time step respectively; represent the velocity magnitudes in the x, y, and z directions of the global coordinates at the current time step respectively; represent the velocity magnitudes in the x, y, and z directions of the global coordinates at the current time step respectively;

[0049] ② Calculate the time difference

[0050] Calculate the time difference between the current time step and the next time step:

[0051] Δt = time_next - time_now;

[0052] Wherein, Δt represents the time difference between the current time step and the next time step; time_next represents the moment represented by the adjacent next time step; time_now represents the moment represented by the current time step;

[0053] ③ Calculate the acceleration vector

[0054] Divide the change in velocity by the time difference to obtain the acceleration vector:

[0055]

[0056] ④ Obtain the local Z-axis direction vector

[0057] Extract the third column from the transformation matrix of the geometric body to represent the local Z-axis direction vector:

[0058]

[0059] ⑤ Normalize the Z-axis vector

[0060] Calculate the magnitude of the Z-axis direction vector:

[0061]

[0062] The unit vector obtained by normalization:

[0063]

[0064] ⑥ Calculate the component of the acceleration in the Z-axis direction

[0065] Use the dot product formula to calculate the projection:

[0066]

[0067] The method for generating and processing the EDEM multi-mesoscopic parameter model, wherein the solution process of the tilt angle in the step (3.2) is as follows:

[0068] ① Extract the local Z-axis vector of the geometric body

[0069] Extract the third column from the transformation matrix Geometry_TransformMatrix of the geometric body:

[0070]

[0071] ② Define the ground normal vector

[0072] The ground normal vector is fixed as:

[0073]

[0074] ③ Calculate the vector dot product

[0075] The dot product result is:

[0076] dot_product = z_axis · ground_normal = z z ;

[0077] ④ Calculate the magnitude of the Z-axis vector of the geometric body

[0078] The vector magnitude calculation formula:

[0079]

[0080] ⑤ Calculate the cosine value

[0081] Substitute the dot product and the magnitude into the cosine formula:

[0082]

[0083] ⑥ Calculate the included angle (in radians)

[0084] Through the arccosine function:

[0085] θ = arccos(cosθ);

[0086] ⑦ Convert radians to degrees

[0087] Finally, convert to degree unit:

[0088]

[0089] The method for generating and processing the EDEM multi-mesoscopic parameter model, wherein the solution process of the effective penetration depth in the step (3.2) is as follows:

[0090] ① Obtain the coordinate data of the vertices of the geometric body

[0091] Extract the list of Z coordinates of the vertices from the geometric data at the current time:

[0092] ZCoords = [Z 1 , Z 2 ,..., Z n ;

[0093] ② Find the minimum value

[0094] Use the minimum value function to operate on the list:

[0095] min_z = min(ZCoords).

[0096] The method for generating and processing the EDEM multi-mesoscopic parameter model, wherein the specific process of the step (4) is as follows:

[0097] (4.1) First, design the graphical user interface GUI through QtDesigner, generate the corresponding.ui file, and use PyQt5 to convert it into Python code to achieve the interaction between the user and the program;

[0098] (4.2) Subsequently, use PyInstaller to package the entire Python program into an independent EXE executable file. The user only needs to run this EXE file to directly execute the batch processing task without installing the Python environment or dependency libraries.

[0099] Adopting the above technical solution, the present invention has the following beneficial effects:

[0100] The concept of the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention is reasonable, and it can realize the batch generation of the EDEM multi-mesoscopic parameter model and the secondary data processing; it can effectively solve the problems of low efficiency in batch adjusting the model parameters and generating the model and cumbersome secondary data processing of the EDEM model, and greatly reduce the time and labor costs required for EDEM simulation.

[0101] Through the integrated development of the graphical user interface and the post-processing function on the original EDEM platform, the present invention not only satisfies the functions of automatically batch adjusting the contact parameters, generating and running the calculation cases, etc., but also satisfies the secondary processing and correction of data such as the tip resistance and acceleration of the impact penetration simulation model, and outputs a curve graph of the processed results for convenient result verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required to be used in the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0103] Figure 1 is the implementation flowchart of the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention;

[0104] Figure 2 is a schematic diagram of the impact penetration simulation model involved in the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention;

[0105] Figure 3 is the schematic diagram of the principle for obtaining the tip resistance of the impact penetration involved in the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention;

[0106] Figure 4 is the visualization interface diagram of the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention;

[0107] Figure 5 is the result diagram of the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention;

[0108] Figure 6 is the comparison diagram of the output results involved in the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention;

[0109] Figure 7 is the operation flowchart of the EDEM software involved in the method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0110] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work shall fall within the protection scope of the present invention.

[0111] The following further explains and illustrates the present invention in combination with specific implementation manners.

[0112] As Figure 1 shown, a method for generating and processing an EDEM multi-mesoscopic parameter model provided in this embodiment is to first construct a basic impact penetration simulation model; then read the initial HDF5 file generated by the simulation model through the EDEMpy tool library of the python software, adjust the contact parameters and generate a new simulation model; generate a batch operation model of the calculation.bat file through python code, use EDEMpy to obtain the simulation results of each time step, and perform secondary processing on the obtained data (such as tip resistance, axial acceleration, etc.) and plot and output the comparison curves of the simulation results of each group; finally, integrate the above script files into the Gui interface to realize the integration of three automatic functions of multi-parameter optimization, data secondary processing and result comparison.

[0113] The method for generating and processing the EDEM multi-mesoscopic parameter model of the present invention specifically includes the following steps:

[0114] S100, Construction of the basic model

[0115] Taking the impact penetration simulation test as an example; first, according to the indoor impact penetration test, create granular materials and set the intrinsic parameters of the granular materials; then, import the geometric model (constructed by software such as CAD / Solidworks) generated by an external modeling software (such as CAD, etc.) into the EDEM software to generate a cylinder consistent with the externally generated geometric model, and use the particle factory function of the EDEM software to generate particles layer by layer; secondly, since the rapidly generated particle body may have overlaps between particles, resulting in contact forces and causing the particles not to be static, remove the unbalanced forces between the particles through stress release, so that the speed, contact force, etc. of the particles are all a fixed value, and the particle body is in an equilibrium state without affecting the subsequent penetration test; finally, after generating the particle body, import the impact penetration cone geometric model into the EDEM software to carry out the impact penetration simulation test, as Figure 2 shown, import the external 3D penetrometer model to generate a basic impact penetration simulation model "edem-base.dem".

[0116] S200, Batch generation and operation of models with different contact parameters

[0117] The EDEMpy tool library, as the Python API interface of the EDEM software, can access the HDF5 files generated by EDEM simulations through the Python software and the Python library developed for EDEM to adjust and read the data of particles or impact penetration penetrometers, including the intrinsic parameters of particles, contact parameters, etc.; then, through the EDEM tool library of the Python software, read the initial HDF5 file generated by the impact penetration simulation model in step S100, and copy the parent folder (the folder containing the.dem file in the upper-level directory of the directory where the basic impact penetration simulation model is located) in the same directory to generate multiple groups of the same folders in the required quantity. One folder represents one model file, and separately adjust the mesoscopic parameters (focus on the coefficient of restitution, static friction coefficient, dynamic friction coefficient, and JKR free surface energy, and each group of parameters can be inconsistent) of the particle bodies and geometric bodies of the simulation models in each model file; at the same time, generate a batch calculation.bat file to batch generate multiple groups of new simulation models.

[0118] S300. Secondary processing of data and result comparison

[0119] The dynamic response characteristics of the impact penetration test are mainly characterized by the time history curves of the tip resistance and the axial acceleration; the EDEM software can only export the basic components such as the force, velocity, and displacement of the penetrometer (in the global coordinate system). When the penetrometer penetrates the soil with an initial inclination angle, it is impossible to obtain characteristic parameters such as the axial acceleration and tip resistance. Therefore, through the Python code of the Python software, generate a batch operation and calculation.bat script file, batch run the new simulation models generated in step S200, use the EDEMpy tool library to obtain the simulation results at each time step, and perform secondary processing on the obtained data and draw the comparison curves of the simulation results of each group (time history curves of quantities such as tip resistance, axial acceleration, and vertical penetration depth); the specific process is as follows:

[0120] S310. First, generate a batch operation and calculation.bat script file through the Python code of the Python software, and batch run the new simulation models generated in the above step S200;

[0121] S320. Then, directly obtain the attitude transformation matrix of the penetrometer through the EDEMpy tool library, and quickly obtain the results such as tip resistance, axial acceleration, tilt angle, and effective penetration depth by coordinate transformation;

[0122] The process of obtaining the above attitude transformation matrix is as follows:

[0123] ① First, introduce the simulation results (.h5 files) of each model folder in the EDEMpy tool library h5py;

[0124] ② Then directly read the attitude transformation matrix Orientation of the penetrometer through the Python code:

[0125]

[0126] Among them, G represents the global coordinate; b represents the carrier coordinate; X G , Y G , Z G respectively represent the vector magnitudes in the X, Y, and Z directions in the global coordinate system; X b , Y b , Z b respectively represent the vector magnitudes in the X, Y, and Z directions in the global coordinate system.

[0127] As Figure 3 shown, the solution process of the above tip resistance is as follows:

[0128] ① First, obtain the acting vector

[0129] Extract the acting vector from the simulation:

[0130]

[0131] Among them, A x , A y , A z respectively represent the force vector magnitudes in the x, y, and z directions in the global coordinate; A represents the force vector on the penetrometer;

[0132] ② Obtain the local Z-axis direction

[0133] Extract the third column from the transformation matrix of the geometry as the Z-axis direction vector:

[0134]

[0135] Among them, z x , z y , z z respectively represent the acceleration vector magnitudes in the x, y, and z directions in the global coordinate; z_axis represents the axial acceleration of the penetrometer;

[0136] ③ Normalize the Z-axis direction vector

[0137] Calculate the vector norm:

[0138]

[0139] Normalize to obtain the unit vector:

[0140]

[0141] ④ Calculate the dot product to obtain the axial tip resistance

[0142] Project vector A onto the unit vector:

[0143]

[0144] The solution process for the above-mentioned axial acceleration is as follows:

[0145] ① Obtain the velocity vector

[0146] Extract the velocity vectors of the geometric body at the current time step and the next time step from the simulation:

[0147]

[0148] Among them, V now and V next represent the velocity vector at the current time step and the velocity vector at the adjacent next time step respectively; represent the magnitudes of the velocities in the x, y, and z directions of the global coordinates at the current time step respectively; represent the magnitudes of the velocities in the x, y, and z directions of the global coordinates at the current time step respectively;

[0149] ② Calculate the time difference

[0150] Calculate the time difference between the current time step and the next time step:

[0151] Δt = time_next - time_now;

[0152] Among them, Δt represents the time difference between the current time step and the next time step; time_next represents the moment represented by the adjacent next time step; time_now represents the moment represented by the current time step;

[0153] ③ Calculate the acceleration vector

[0154] Divide the change in velocity by the time difference to obtain the acceleration vector:

[0155]

[0156] ④ Obtain the local Z-axis direction vector

[0157] Extract the third column from the transformation matrix of the geometric body to represent the local Z-axis direction vector:

[0158] ⑤ Normalize the Z-axis vector

[0159] Calculate the magnitude of the Z-axis direction vector:

[0160]

[0161] The unit vector obtained by normalization:

[0162]

[0163] ⑥ Calculate the component of the acceleration in the Z-axis direction

[0164] Calculate the projection using the dot product formula:

[0165]

[0166] The solution process for the above-mentioned inclination angle is as follows:

[0167] ① Extract the local Z-axis vector of the geometric body

[0168] Extract the third column from the transformation matrix Geometry_TransformMatrix of the geometric body:

[0169] ② Define the ground normal vector

[0170] The ground normal vector is fixed as:

[0171]

[0172] ③ Calculate the dot product of vectors

[0173] The result of the dot product is:

[0174] dot_product = z_axis · ground_normal = z z ; ④ Calculate the modulus length of the Z-axis vector of the geometric body. The modulus length calculation formula:

[0175]

[0176] ⑤ Calculate the cosine value

[0177] Substitute the dot product and the modulus length into the cosine formula:

[0178]

[0179] ⑥ Calculate the included angle (in radians)

[0180] Through the arccosine function:

[0181] θ = arccos(cosθ);

[0182] ⑦ Convert radians to degrees

[0183] Finally, convert to the degree unit:

[0184]

[0185] The solution process for the above-mentioned effective penetration depth is as follows:

[0186] ① Extract the coordinate data of the vertices of the geometric body. Extract the list of Z coordinates of the vertices from the geometric data at the current time:

[0187] ZCoords = [Z 1 , Z 2 ,..., Z n ;

[0188] ② Find the minimum value

[0189] Operate on the list using the minimum value function:

[0190] min_z = min(ZCoords).

[0191] S330. Output the results of the new simulation model generated in step S310 as a.CSV file through the Python script code of the Python software, draw the time history curves of quantities such as tip resistance, axial acceleration, and vertical penetration depth, and compare the time history curves with the results of the impact penetration indoor test.

[0192] S400. Provide a convenient operation method through the visualization interface developed using Qt Designer and PyQt5, which is especially suitable for non-technical users. First, design the graphical user interface (GUI) through Qt Designer to generate the corresponding.ui file, and use PyQt5 to convert it into Python code to achieve the interaction between the user and the program; through the simple and intuitive interface, the user can easily input parameters, trigger batch processing operations, and view the task progress in real time. Subsequently, use PyInstaller to package the entire Python program into an independent EXE executable file. The user only needs to run this EXE file to directly execute batch processing tasks without installing the Python environment or dependency libraries, as Figure 4 shown. By providing a friendly visualization interface and an independent EXE file, non-technical users can use complex batch processing tools without obstacles, greatly improving the usability and ease of use of the application.

[0193] The operation results of the present invention:

[0194] Taking the impact penetration simulation test as an example, through the parameter input operation of the visualization interface, the software automatically generates and runs the simulation model.

[0195] And perform secondary processing on the obtained data (such as noise reduction, coordinate system conversion, etc.) and draw and output the comparison curves of each group of simulation results. Finally, obtain the simulation results and curves in the parameter_calibration folder, as Figure 5 shown (only one simulation).

[0196] By using the same impact penetration simulation model, manual simulation of EDEM was carried out with the same penetration speed and mesoscopic parameters (restitution coefficient, static friction coefficient, dynamic friction coefficient, etc.), and the output tip penetration resistance, acceleration and penetration depth were compared to determine the accuracy of the simulation process of the software for batch generation and secondary processing of multi-mesoscopic parameter models based on the EDEM platform. The comparison results are as Figure 6 shown. The simulation results of the software are almost the same as those of the manual simulation, indicating the accuracy of the software simulation.

[0197] As Figure 7 shown, the operation process of the EDEM software in the method for generating and processing multi-mesoscopic parameter models of the present invention is as follows:

[0198] (1) Select and input the installation location of the EDEM software in the computer;

[0199] (2) Select the basic simulation model to be operated;

[0200] (3) Input the number of simulation models to be run in batches, click confirm to confirm, and the software will copy the same number of model folders;

[0201] (4) Input the modified values of the parameters concerned by each simulation model in turn, and click modify to modify;

[0202] (5) Modify the settings such as the total solution time as needed, and click run deck to run multiple groups of simulation models;

[0203] (6) The running status of the simulation model will appear in real time at the lower right corner of the EDEM software. After the operation is completed, click result to perform secondary data processing and result output on the generated model, and the output results are in the corresponding model folder;

[0204] (7) The simulation is completed.

[0205] The method for generating and processing multi-mesoscopic parameter models of the present invention can realize the functions of simultaneously and automatically generating and running a large number of models with different parameters and automatically performing secondary data processing and comparison, and can effectively solve the problem that EDEM lacks an automated multi-parameter optimization tool for carrying out multi-group parameter optimization work.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating and processing an EDEM multi-micro-parameter model, characterized in that , mainly including the following steps: (1) First, construct an impact penetration simulation model with the same size as the indoor impact penetration test; (2) reading the initial HDF5 file generated by the impact penetration simulation model constructed in the above step (1) through the EDEMpy tool library of the python software, adjusting the contact parameters and generating a new simulation model; (3) Generate a .bat script file for batch operation and calculation through the python code of the python software, batch run the new simulation model generated in step (2), use the EDEMpy tool library to obtain the simulation results of each time step, perform secondary processing on the obtained data, and draw comparison curves of each group of simulation results; (4) The code file .py generated by the python software is integrated into the Gui interface and packaged into an .exe program file to realize the integration of three automated functions: multi-parameter optimization, secondary data processing, and result comparison.

2. The EDEM multi-micro-parameter model generation and processing method according to claim 1, characterized in that: The specific process of constructing the basic impact penetration simulation model in step (1) is as follows: (1.1) Create granular materials and set the intrinsic parameters of granular materials based on indoor impact penetration tests; (1.2) Import the geometric model generated by the external modeling software into the EDEM software to generate a cylinder that is consistent with the externally generated geometric model for use in generating particles. Use the particle factory function of the EDEM software to generate particles layer by layer; (1.3) Remove the unbalanced force between particles through stress release, so that the particles are in a balanced state; (1.4) Import the impact penetration probe geometry model into the EDEM software and generate the basic impact penetration simulation model "edem-base.dem" to carry out the impact penetration simulation test.

3. The EDEM multi-micro-parameter model generation and processing method according to claim 1, characterized in that: The specific process of step (2) is as follows: the EDEMpy tool library is used as the Python API interface of the EDEM software. Through the python software and the python library developed for EDEM, the HDF5 file generated by the EDEM software simulation is accessed to adjust and read the data of the particle or impact penetration probe; then the initial HDF5 file generated by the impact penetration simulation model in step (1) is read through the EDEM tool library of the python software, and the parent folder where the basic impact penetration simulation model is located is copied in the same directory to generate the required number of multiple groups of identical folders, one folder represents a model file, and the microscopic parameters of the particle body and geometric body of the simulation model in each model file are adjusted respectively to generate multiple groups of new simulation models in batches.

4. The EDEM multi-micro-parameter model generation and processing method according to claim 1, characterized in that: The specific process of step (3) is as follows: (3.1) Generate a batch calculation .bat file through the python code of the python software to batch run the new simulation model generated in step (2); (3.2) The attitude transformation matrix of the penetrometer is directly obtained through the EDEMpy tool library, and the cone tip resistance, axial acceleration, inclination angle and effective penetration depth are quickly obtained by coordinate transformation; (3.3) The results of the new simulation model generated in the above step (3.1) are exported as a .CSV file through the python script code of the python software, and the time history curves of cone tip resistance, axial acceleration and vertical penetration depth are plotted, and the time history curves are compared with the impact penetration indoor test results.

5. The EDEM multi-micro-parameter model generation and processing method according to claim 4, characterized in that , the process of obtaining the posture transformation matrix in step (3.2) is: ①First, import the simulation results (.h5 files) of each model folder of the EDEMpy tool library h5py; ② Then directly read the orientation matrix of the penetrometer through the python code: Where G represents the global coordinates; b represents the carrier coordinates; X G , Y G , Z G Respectively represent the vector size in the X, Y, and Z directions in the global coordinate system; X b , Y b , Z b They represent the vector sizes in the X, Y, and Z directions in the global coordinate system respectively.

6. The EDEM multi-micro-parameter model generation and processing method according to claim 4, characterized in that , the solution process of the cone tip resistance in step (3.2) is: ①First obtain the action vector Extract the action vector from the simulation: Among them, A x , A y , A z They represent the magnitude of the force vector in the x, y, and z directions in the global coordinates respectively; A represents the force vector acting on the penetrometer; ②Get the local Z-axis direction Extract the third column from the geometry's transformation matrix as the Z-axis direction vector: Among them, z x 、z y 、z z They represent the magnitude of the acceleration vector in the x, y, and z directions in the global coordinates respectively; z_axis represents the axial acceleration of the penetrometer; ③Normalize the Z-axis vector Calculate the vector modulus: Normalize to get the unit vector: ④ Calculate the dot product to obtain the axial cone tip resistance Project vector A to the unit vector:

7. The EDEM multi-micro-parameter model generation and processing method according to claim 4, characterized in that , the solution process of the axial acceleration in step (3.2) is: ①Get velocity vector Extract the velocity vectors of a geometry at the current time step and the next time step from the simulation: Among them, V now 、V next Represent the velocity vector of the current time step and the velocity vector of the next adjacent time step respectively; Respectively represent the velocity magnitude in the global coordinate x, y, and z directions of the current time step; Respectively represent the velocity magnitude in the global coordinate x, y, and z directions of the current time step; ②Calculate the time difference Calculate the time difference between the current time step and the next time step: Δt = time_next - time_now; Among them, Δt represents the time difference between the current time step and the next time step; time_next represents the moment represented by the next adjacent time step; time_now represents the moment represented by the current time step; ③Calculate the acceleration vector Divide the velocity change by the time difference to get the acceleration vector: ④Get the local Z-axis direction vector Extract the third column from the geometry's transformation matrix, which represents the local Z-axis direction vector: ⑤Normalize the Z-axis vector Calculate the modulus of the vector in the Z-axis direction: The normalized unit vector is: ⑥Calculate the component of acceleration in the Z-axis direction Use the dot product formula to calculate the projection:

8. The EDEM multi-micro-parameter model generation and processing method according to claim 4, characterized in that , the solution process of the inclination angle in step (3.2) is: ① Extract the local Z-axis vector of the geometric body Extract the third column from the geometry's transformation matrix Geometry_TransformMatrix: ②Define the ground normal vector The ground normal vector is fixed to: ③Calculate vector dot product The dot product result is: dot_product=z_axis·ground_normal=z z ; ④ Calculate the Z-axis modulus of the geometric body Vector modulus calculation formula: ⑤Calculate the cosine value Substituting the dot product and the modulus into the cosine formula: ⑥Calculate the angle (radians) Through the inverse cosine function: θ = arccos(cosθ); ⑦ Radians to Angle Finally converted to angle units:

9. The EDEM multi-micro-parameter model generation and processing method according to claim 4, characterized in that , the solution process of the effective penetration depth in step (3.2) is: ① Get the coordinate data of the geometric vertices Extract the Z coordinate list of vertices from the geometry data at the current time: ZCoords=[Z1,Z2,…,Z n ]; ②Find the minimum value Use the minimum function to operate on a list: min_z=min(ZCoords).

10. The EDEM multi-micro-parameter model generation and processing method according to claim 1, characterized in that: The specific process of step (4) is as follows: (4.1) First, design the graphical user interface GUI through Qt Designer, generate the corresponding .ui file, and use PyQt5 to convert it into Python code to realize the interaction between users and programs; (4.2) Then, PyInstaller is used to package the entire Python program into an independent EXE executable file. Users only need to run the EXE file to directly execute batch processing tasks without installing the Python environment or dependent libraries.

Images