Parameter calibration method and device of simulation software, computer equipment and storage medium

Abstract

The invention provides a parameter calibration method and device for simulation software, computer equipment and a storage medium, and the method comprises the steps: obtaining actual mechanical parameter values, determined based on an experiment, of a coal and rock mass, the actual mechanical parameters comprise an actual compressive strength value and an actual tensile strength value, and the actual compressive strength value is greater than the actual tensile strength value; the trained first prediction model is used for prediction according to the actual compressive strength value and the actual tensile strength value of the coal and rock mass, calibration values of contact parameters between simulation particles of the coal and rock mass in discrete element software are obtained, and the trained first prediction model has learned a complex nonlinear relation between the parameters; the actual mechanical parameter value determined by the experiment is predicted through the first prediction model, the calibration value of the contact parameter between the simulation particles of the coal and rock mass in the discrete element software is obtained, rapid and accurate calibration of the contact parameter between the simulation particles in the discrete element software is achieved, and the calibration efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the fields of coal mining, rock mechanics and artificial intelligence technology, and in particular to a method, device, computer equipment and storage medium for parameter calibration of simulation software. Background Art

[0002] In the process of coal mining, the cutting mechanism of the mining equipment directly interacts with the coal and rock mass, and the mechanical properties of the coal and rock mass are important factors that determine the service life of the cutting mechanism of the mining equipment. However, the actual mining time is long, the on-site working conditions are complex, the risk factor is high, and the cost of industrial testing is high. The discrete element method is an important method for simulating the mechanical behavior of granular materials and is widely used in the numerical simulation analysis of the coal and rock cutting process. Therefore, using discrete element software to simulate the coal and rock cutting process of mining equipment is an important means of studying coal and rock cutting.

[0003] However, due to the complex mechanical properties of coal and rock in coal mines, the parameter calibration of discrete element software usually adopts the trial and error method, which requires a lot of experiments and complicated manual debugging, resulting in high time cost and poor accuracy. Therefore, how to accurately and quickly calibrate the parameters of discrete element software is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the present application proposes, on the one hand, a method, apparatus, computer equipment and storage medium for parameter calibration of simulation software, so as to realize accurate and rapid calibration of parameters of discrete element software based on a neural network model obtained through physical experiments and training.

[0006] In one aspect, an embodiment of the present application provides a method for calibrating parameters of simulation software, including:

[0007] Acquire actual mechanical parameter values ​​of the coal rock mass determined based on experiments; wherein the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values;

[0008] The first prediction model obtained through training is used to perform prediction according to the actual compressive strength value and the actual tensile strength value of the coal rock mass, so as to obtain the calibration value of the contact parameter between the simulation particles of the coal rock mass in the discrete element software.

[0009] Another aspect of the present application provides a parameter calibration device for simulation software, including:

[0010] An acquisition module, used to acquire actual mechanical parameter values ​​of the coal rock mass determined based on experiments; wherein the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values;

[0011] The calibration module is used to use the first prediction model obtained through training to predict according to the actual compressive strength value and the actual tensile strength value of the coal rock mass, and obtain the calibration value of the contact parameter between the simulated particles of the coal rock mass in the discrete element software.

[0012] Another aspect of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in the above aspect is implemented.

[0013] Another aspect of the present application provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method described in the aforementioned aspect is implemented.

[0014] Another aspect of the present application provides a computer program product on which a computer program is stored. When the program is executed by a processor, the method described in the above aspect is implemented.

[0015] The parameter calibration method, device, computer equipment and storage medium of the simulation software proposed in the present application obtain actual mechanical parameter values ​​of coal rock mass determined based on experiments, wherein the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values. The first prediction model obtained by training is used to predict the actual compressive strength value and the actual tensile strength value of the coal rock mass, and the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software are obtained. The first prediction model obtained by training has learned the complex nonlinear relationship between the parameters. The actual mechanical parameter values ​​determined by the experiment are predicted by the first prediction model, and the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software are obtained, thereby realizing the rapid and accurate calibration of the contact parameters between the simulated particles in the discrete element software and improving the calibration efficiency and accuracy.

[0016] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0018] Figure 1 A flow chart of a parameter calibration method for a simulation software provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a flow chart of another method for calibrating parameters of simulation software provided in an embodiment of the present application;

[0020] Figure 3 A schematic diagram of particle stacking simulation using discrete element software provided in an embodiment of the present application;

[0021] Figure 4 A schematic diagram of a flow chart of another method for calibrating parameters of simulation software provided in an embodiment of the present application;

[0022] Figure 5 A schematic diagram of a uniaxial compression test and a Brazilian splitting test simulated by a discrete element software provided in an embodiment of the present application;

[0023] Figure 6 A schematic diagram of a test result provided in an embodiment of the present application;

[0024] Figure 7 A schematic diagram of the structure of a parameter calibration device for simulation software provided in an embodiment of the present application;

[0025] Figure 8 A block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0027] The following describes the parameter calibration method, apparatus, computer equipment and storage medium of the simulation software of the embodiment of the present application with reference to the accompanying drawings.

[0028] Figure 1 A flow chart of a parameter calibration method for a simulation software provided in an embodiment of the present application.

[0029] like Figure 1 As shown, the method comprises the following steps:

[0030] Step 101, obtaining actual mechanical parameter values ​​of the coal rock mass determined based on experiments.

[0031] Among them, the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values.

[0032] In the embodiment of the present application, the experiment can be multiple physical experiments, and the coal rock mass includes coal mines and rocks. As an implementation method, a cylindrical coal rock mass sample of standard size for the experiment is prepared according to the "Recommended Method for Rock Mechanics Test", and the actual mechanical parameters of the coal rock mass are determined through physical experiments. Specifically, the actual compressive strength value of the coal rock mass sample is measured through a uniaxial compression test, for example, the uniaxial compressive strength, denoted as σc The actual tensile strength of the coal rock sample is measured by the Brazilian splitting test, for example, the uniaxial tensile strength, denoted as σ t In the case of multiple experiments, the average value of the multiple test results is taken as the actual compressive strength value σ of the coal rock mass. c and the actual tensile strength value σ t .

[0033] The uniaxial compression test loads the coal rock sample to the failure point and records its maximum bearing pressure F max , according to the maximum bearing pressure F max , calculate the actual compressive strength value:

[0034]

[0035] Where A is the cross-sectional area of ​​the coal rock sample.

[0036] In the Brazilian splitting test, the coal-rock sample is loaded until it breaks and the ultimate load F at the time of rupture is recorded. t , calculate the actual tensile strength value:

[0037]

[0038] Wherein, D is the diameter of the coal rock sample, and T is the thickness at the center of the coal rock sample.

[0039] Step 102 , using the first prediction model obtained through training to perform prediction based on the actual compressive strength value and the actual tensile strength value of the coal rock mass, to obtain the calibration value of the contact parameter between the simulated particles of the coal rock mass in the discrete element software.

[0040] In the embodiment of the present application, the first prediction model is a deep learning model, such as a feedforward artificial neural network model optimized by a genetic algorithm (GA-BP-ANN), wherein the feedforward artificial neural network model optimized by a genetic algorithm combines the advantages of a genetic algorithm (GA) and a feedforward artificial neural network (FANN). Among them, the simulated particles are particles of coal rock mass simulated in discrete element software.

[0041] In one implementation of the embodiment of the present application, a training method for a first prediction model includes: obtaining a training sample set, wherein the training sample set includes a plurality of first training samples, each first training sample includes a simulated internal friction angle value of a simulated particle determined based on discrete element software, and the simulated internal friction angle value is annotated with a set value of a contact parameter between simulated particles, that is, the internal friction angle between simulated particles is determined based on the set value of the contact parameter between simulated particles, that is, the set value of the contact parameter between simulated particles is the value of the internal friction angle. Input each first training sample into the first prediction model to obtain a predicted value of the contact parameter between simulated particles predicted by the first prediction model based on each simulated internal friction angle value.

[0042] According to the difference between the predicted values ​​of the contact parameters between the simulated particles corresponding to the multiple simulated internal friction angle values ​​and the set values ​​of the contact parameters between the marked simulated particles corresponding to the multiple simulated internal friction angle values, the loss function of the first prediction model is determined, wherein the loss function of the first prediction model is a mean square error (MSE) loss function, and the loss function of the first prediction model is determined by the following formula:

[0043]

[0044] Among them, y i is the true value, i.e., the i-th set value of the contact parameter between simulated particles; is the model prediction value, that is, the ith prediction value of the contact parameter between simulated particles; N is the number of the first training samples.

[0045] Then, the parameters of the first prediction model are adjusted according to the loss function of the first prediction model to obtain the trained first prediction model. The first prediction model can use multiple first training samples or multiple batches of first training samples to perform multiple parameter adjustment processes until the loss function is less than the set threshold or the number of model training times reaches the set number, and then the training of the first prediction model is completed, so that the trained first prediction model learns the corresponding relationship between the internal friction angle and the contact parameter between the simulated particles of the coal rock mass. Since the model uses multiple simulated internal friction angle values ​​of the simulated particles for model training during the training process, the multiple internal friction angles can be determined under different experimental conditions. By training based on the internal friction angles under different experimental conditions, the first prediction model can automatically learn the complex nonlinear relationship between the parameters, so that the trained first prediction model has good versatility and adaptability, and can quickly obtain new calibration parameters by changing the input experimental conditions, thereby meeting the rapid calibration requirements under different working conditions and improving the calibration efficiency.

[0046] Furthermore, the first prediction model obtained through training is used to perform predictions based on the actual compressive strength value and the actual tensile strength value of the coal rock mass, and the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software are obtained. The first prediction model obtained through training has good versatility and adaptability. The first prediction model obtained through training can accurately and quickly complete the determination of the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software, and realizes that through simple experiments combined with the first prediction model obtained through training, the calibration values ​​of the basic parameters that are consistent with the experimental results can be determined. The calibration values ​​can also be called target values ​​that are consistent with the actual results.

[0047] The parameter calibration method of the simulation software of the embodiment of the present application obtains the actual mechanical parameter values ​​of the coal rock mass determined based on experiments, wherein the actual mechanical parameter values ​​include the actual compressive strength value and the actual tensile strength value, and uses the trained first prediction model to predict the actual compressive strength value and the actual tensile strength value of the coal rock mass, so as to obtain the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software. The trained first prediction model has learned the complex nonlinear relationship between the parameters. The actual mechanical parameter values ​​determined by the experiment are predicted by the first prediction model, and the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software are obtained, thereby realizing the rapid and accurate calibration of the contact parameters between the simulated particles in the discrete element software, and improving the calibration efficiency and accuracy.

[0048] Based on the previous embodiment, this embodiment provides another method for calibrating parameters of simulation software, which specifically illustrates that before training the first prediction model, it is necessary to first determine the training samples used for training the first prediction model. Figure 2 A flowchart of another method for calibrating parameters of simulation software provided in an embodiment of the present application.

[0049] like Figure 2 As shown, the method may include the following steps:

[0050] Step 201 , using discrete element software to perform a stacking simulation test on the simulated particles according to a plurality of set values ​​of the contact parameters between the simulated particles, and determining the stacking angle values ​​of the simulated particles corresponding to the respective set values ​​of the contact parameters between the simulated particles.

[0051] Among them, the set three-dimensional model is imported into the discrete element software for various simulation or simulation experiments on coal rock mass.

[0052] In one implementation of the embodiment of the present application, the contact parameters between the simulated particles include at least one of the coefficient of restitution, the coefficient of kinetic friction and the coefficient of static friction. Take the contact parameters including the coefficient of restitution, the coefficient of kinetic friction and the coefficient of static friction as an example for illustration, wherein three different set values ​​are set for each of the three parameters, the coefficient of restitution, the coefficient of kinetic friction and the coefficient of static friction, wherein the three set values ​​corresponding to the coefficient of restitution are A1, A2 and A3, the three set values ​​corresponding to the coefficient of kinetic friction are B1, B2 and B3, and the three set values ​​corresponding to the coefficient of static friction are C1, C2 and C3. Using orthogonal experimental design (L9(34)), a total of 9 groups of numerical simulations are required to determine the corresponding set values ​​of the three parameters, which can cover all working conditions. In actual applications, more groups of set value combinations can be tested, and a total of 27 groups of tests can be performed. As an example, Figure 3 A schematic diagram of a particle stacking simulation simulated by a discrete element software provided in an embodiment of the present application is provided. Discrete element simulation is carried out according to each group of parameters set in the aforementioned orthogonal experiment, and simulated particles of coal rock mass with geometric dimensions corresponding to the model requirements built into the discrete element software are established, and contact parameters (restitution coefficient, dynamic friction coefficient and static friction coefficient) between the simulated particles are set. Then, for any one of a plurality of set values ​​of the contact parameters, wherein any one of the set values ​​includes set values ​​of the restitution coefficient, the dynamic friction coefficient and the static friction coefficient, for example, the set value of the restitution coefficient in the contact parameters is A1, the set value corresponding to the dynamic friction coefficient is B2, and the set value corresponding to the static friction coefficient is C1. Under the set value of the contact parameters, the particles are allowed to fall freely from a set height, such as Figure 3 As shown in the figure, a free stack of simulated particles is formed on the ground, and the stacking angle value of the simulated particles is determined. The stacking angle is as follows: Figure 3 φ in.

[0053] Step 202 , according to the stacking angle values ​​of the simulated particles corresponding to the various set values ​​of the contact parameters between the simulated particles, the simulated internal friction angle values ​​of the simulated particles corresponding to the various set values ​​of the contact parameters between the simulated particles are determined.

[0054] In the embodiment of the present application, for the simulated particles, their stacking angle value is the same as the simulated internal friction angle value. Therefore, the measured stacking angle value of the simulated particles is the simulated internal friction angle value of the simulated particles under the contact parameters, that is, the stacking angle value of the simulated particles corresponding to each set value of the contact parameter between the simulated particles is used as the actual internal friction angle value of the simulated particles corresponding to each set value of the contact parameter between the simulated particles to simulate the internal friction angle value. By performing 9 groups of simulation tests using discrete element software, the simulated internal friction angle values ​​of the simulated particles under 9 contact parameters can be obtained. By simulating the mechanical characteristics of the simulated particles of coal rock mass under various working conditions, the internal friction angle of the simulated particles under various working conditions can be determined, thereby realizing the simulation of various working conditions and meeting the needs of different working conditions.

[0055] Step 203 : determining each first training sample according to each simulated internal friction angle value of the simulated particles and each set value of the contact parameter between the corresponding simulated particles.

[0056] Step 204: Generate a training sample set according to the multiple first training samples.

[0057] In an embodiment of the present application, each simulated internal friction angle value of the simulated particles is used as a training sample, and the contact parameters between the simulated particles corresponding to each simulated internal friction angle value are used as annotation information to generate each first training sample, and multiple first training samples are used as a training sample set, so that the each simulated internal friction angle value of the simulated particles included in the training sample set is measured under various working conditions, including the actual internal friction angle of coal rock under different scenarios and experimental conditions. The first prediction model is trained based on the training sample set, so that the first prediction model can automatically learn the complex nonlinear relationship between parameters and can meet the requirements of different working conditions, so that the first prediction model has good versatility and adaptability, and the reliability of the first prediction model training is improved.

[0058] The parameter calibration method of the simulation software of the embodiment of the present application adopts discrete element software to experimentally determine the simulated internal friction angle values ​​of the simulated particles corresponding to each set value of the contact parameter between the simulated particles according to multiple set values ​​of the contact parameter between the simulated particles, use each simulated internal friction angle value of the simulated particles as a training sample, use the contact parameters between the simulated particles corresponding to each simulated internal friction angle value as annotation information, generate each first training sample, and combine multiple first training samples to form a training sample set, so that each simulated internal friction angle value of the simulated particles included in the training sample set is measured under various working conditions, including the actual internal friction angle of coal rock in different scenarios, and train the first prediction model based on the training sample set, so that the first prediction model can automatically learn the complex nonlinear relationship between parameters and can meet the requirements of different working conditions, so that the first prediction model has good versatility and adaptability, and the reliability of the training of the first prediction model is improved.

[0059] Based on the previous embodiment, this embodiment provides another method for calibrating parameters of simulation software, which specifically describes how to calibrate the bonding parameters between simulation particles of coal and rock mass in discrete element software. Figure 4 A flowchart of another method for calibrating parameters of simulation software provided in an embodiment of the present application.

[0060] like Figure 4 As shown, the method may include the following steps:

[0061] Step 401, obtaining actual mechanical parameter values ​​of the coal rock mass determined based on experiments.

[0062] Among them, the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values.

[0063] Step 402 , using the first prediction model obtained through training to perform prediction based on the actual compressive strength value and the actual tensile strength value of the coal rock mass, obtains the calibration value of the contact parameter between the simulated particles of the coal rock mass in the discrete element software.

[0064] Among them, step 401 and step 402 can refer to the relevant explanations in the above-mentioned embodiments, and the principles are the same, which will not be repeated here.

[0065] Step 403, using discrete element software, based on multiple set values ​​of bonding parameters between simulated particles and calibrated values ​​of contact parameters between simulated particles, to determine the simulated mechanical parameter values ​​of the coal rock mass corresponding to each set value of bonding parameters of simulated particles.

[0066] Among them, the simulated mechanical parameter value, where the simulation is used to indicate the mechanical parameter value, is not obtained by actual physical experimental measurement of the coal rock mass, but is obtained by simulation test simulation using discrete element software.

[0067] In the embodiment of the present application, the simulated particle bonding parameters include normal stiffness, tangential stiffness, normal strength and tangential strength, a total of 4 parameters. The above 4 parameters are designed, and each parameter has 5 set values ​​of orthogonal experiments (L25 (45)), a total of 25 groups can meet the needs of different working conditions, wherein each group is a set value of a bonding parameter, including the set values ​​of the 4 parameters of normal stiffness, tangential stiffness, normal strength and tangential strength, the discrete element software is set to the coal rock isoproportional numerical simulation mode, and the uniaxial compression and Brazilian splitting experiments of the coal rock samples are simulated. Figure 5 As shown in the figure, the calibrated values ​​of the contact parameters between the simulated particles are input into the discrete element software to fix the contact parameters between the simulated particles in the discrete element software. Then, the 25 sets of parameters of the set L25 (45) orthogonal experiment are respectively input into the discrete element software to carry out Figure 5 The uniaxial compression test simulation of the coal rock mass indicated in (a), and Figure 5 The Brazilian splitting test simulation indicated in (b) obtains the simulated compressive strength and simulated tensile strength of the coal rock mass corresponding to each set of bonding parameters, that is, the simulated mechanical parameter values ​​of the coal rock mass corresponding to each set value of the bonding parameters are obtained, wherein the simulated mechanical parameter values ​​include simulated compressive strength and simulated tensile strength, wherein the "simulated" in the simulated compressive strength and simulated tensile strength serves as an identification, and the identification is obtained by simulation.

[0068] It should be noted that the orthogonal experiment (L25(45)) refers to conducting 25 uniaxial compression tests on coal rock to simulate the simulated compressive strength of the coal rock; and 25 Brazilian splitting tests to simulate the simulated tensile strength of the coal rock, but it is not limited to 25 times, and more experiments can be conducted.

[0069] Step 404: determine each second training sample according to the simulated mechanical parameter value of the coal rock mass corresponding to each set value of the bonding parameter.

[0070] In an embodiment of the present application, for each setting value of the bonding parameter, experimental simulation is performed through discrete element software to obtain the simulated mechanical parameter value of the coal rock mass corresponding to the setting value of the bonding parameter, and the multiple simulated mechanical parameter values ​​obtained are used as training samples of the second prediction model, and each simulated mechanical parameter value is calibrated to the corresponding setting value of the bonding parameter.

[0071] Step 405: Use multiple second training samples to train the second prediction model to obtain a trained second prediction model.

[0072] The second training sample is used to distinguish it from the first training sample, and both "first" and "second" serve as identifiers for identifying training samples for training different models or different stages of models.

[0073] The second prediction model is a deep learning model, such as a feedforward artificial neural network model optimized by a genetic algorithm (GA-BP-ANN), wherein the feedforward artificial neural network model optimized by a genetic algorithm combines the advantages of a genetic algorithm (GA) and a feedforward artificial neural network (FANN).

[0074] The training method of the second prediction model can be implemented in the following way:

[0075] Each second training sample is input into the second prediction model to obtain the predicted value of the simulated particle bonding parameter corresponding to each simulated mechanical parameter value, and the loss function of the second prediction model is determined according to the difference between the predicted value of the simulated particle bonding parameter corresponding to the multiple simulated mechanical parameter values ​​and the set value of the simulated particle bonding parameter corresponding to the multiple simulated mechanical parameter values, wherein the loss function of the second prediction model can be a mean square error loss function, which can be determined by the following formula:

[0076]

[0077] Among them, k i is the true value, i.e., the i-th setting value of the simulation particle bonding parameter. is the model prediction value, that is, the ith prediction value of the simulated inter-particle bonding parameter, and N is the number of samples.

[0078] Then, the parameters of the second prediction model are adjusted according to the loss function of the second prediction model to obtain the trained second prediction model. The second prediction model can use multiple training samples or multiple batches of training samples to perform multiple parameter adjustment processes until the loss function is less than the set threshold or the number of model training times reaches the set number, and then the training of the second prediction model is completed, so that the trained second prediction model learns the corresponding relationship between the mechanical parameter values ​​of the coal rock mass and the bonding parameters between the simulated particles. Since the model uses the mechanical parameter values ​​of multiple coal rock masses of simulated particles for model training during the training process, the mechanical parameter values ​​of the coal rock mass can be determined under different experimental conditions. By training based on the mechanical parameter values ​​of the coal rock mass under different experimental conditions, the second prediction model can automatically learn the complex nonlinear relationship between the parameters, so that the trained second prediction model has good versatility and adaptability, and can quickly obtain new calibration parameters by changing the input experimental conditions, thereby meeting the rapid calibration requirements under different working conditions and improving the calibration efficiency.

[0079] Step 406, determining the actual internal friction angle value of the coal rock mass according to the actual compressive strength and the actual tensile strength of the coal rock mass.

[0080] In one implementation of the embodiment of the present application, according to the actual compressive strength σ of the coal rock mass c and the actual tensile strength σ t , the calculation to determine the actual internal friction angle φ of the coal rock mass is as follows:

[0081]

[0082] Where c is the shear strength of the coal rock mass.

[0083]

[0084] Among them, φ is the actual internal friction angle of the coal rock mass.

[0085] Step 407 , using the trained second prediction model to perform prediction based on the actual internal friction angle value of the coal rock mass, to obtain the calibration value of the bonding parameter between the simulated particles of the coal rock mass in the discrete element software.

[0086] In the embodiment of the present application, the second prediction model obtained by training has good versatility and adaptability. The second prediction model obtained by training is used based on the actual internal friction angle to accurately and quickly complete the determination of the calibration values ​​of the bonding parameters between the simulated particles of the coal rock mass in the discrete element software. It is achieved by combining a simple experiment with the second prediction model obtained by training to determine the calibration values ​​of the bonding parameters that are consistent with the experimental results. The calibration value can also be called a target value that is consistent with the actual results.

[0087] The parameter calibration method of the simulation software of the embodiment of the present application obtains the actual internal friction angle value of the coal rock mass determined based on the experiment, uses the trained second prediction model to predict according to the actual internal friction angle value of the coal rock mass, and obtains the calibration value of the bonding parameter between the simulated particles of the coal rock mass in the discrete element software. The second prediction model obtained by training with the actual internal friction angle value of the coal rock mass obtained under different experimental conditions or working conditions has learned the complex nonlinear relationship between the parameters. The actual internal friction angle value determined by the experiment is predicted by the second prediction model, and the calibration value of the bonding parameter between the simulated particles of the coal rock mass in the discrete element software is obtained, thereby realizing the rapid and accurate calibration of the bonding parameters between the simulated particles in the discrete element software and improving the calibration efficiency and accuracy.

[0088] It should be understood that compared with the related art, the parameter calibration of discrete element software relies on manual experience or manual adjustment, resulting in limited calibration accuracy and large errors between simulation results and experimental results. The calibration method of this application uses a small amount of experimental data and simulation data obtained by orthogonal experiments as training samples, which improves the accuracy of parameter calibration and reduces the number of experiments and costs. By using the internal friction angle, compressive strength and tensile strength of the coal rock mass as input parameters of the trained prediction model, the trained prediction model automatically identifies and optimizes the relationship between parameters, ensuring that the calibrated parameters are more in line with the actual mechanical properties of the coal rock mass, and effectively reducing the simulation error.

[0089] Based on the above embodiment, the discrete element software can also be verified based on the calibration values ​​of the contact parameters and the calibration values ​​of the bonding parameters:

[0090] As an implementation method, discrete element software is used to carry out proportional uniaxial compression tests and Brazilian splitting tests on coal-rock masses based on the calibrated values ​​of contact parameters and the calibrated values ​​of bonding parameters, and simulated values ​​are calculated, including simulated compressive strength values, simulated tensile strength values ​​and simulated internal friction angle values ​​of the coal-rock masses. The values ​​are compared with the experimental values ​​determined by the aforementioned real physical experiments, including the determined actual compressive strength, actual tensile strength and actual internal friction angle values, to determine the error rate, which includes the sum of the error rates of compressive strength, tensile strength and internal friction angle, where the error rate δ between the actual value and the simulated value is:

[0091]

[0092] Among them, y exp is the experimental value, y sim is the simulation value. If the error is within the allowable range, the calibration is completed and the accurate discrete element software parameters are obtained; if it does not meet the requirements, the first prediction model and the second prediction model are iteratively optimized again until the accuracy requirements are met.

[0093] Based on the above embodiments, Figure 6 A test result diagram provided in an embodiment of the present application is shown in FIG. Figure 6 As shown in the figure, the tensile strength of the coal rock mass measured in the experiment is 0.4MPa; the compressive strength is 18MPa. After the discrete element software parameters are calibrated, the tensile strength and compressive strength change curves obtained during the discrete element calculation process show that the ultimate tensile strength of the coal rock mass is 0.375MPa, and the compressive strength is 19.03MPa, which is very close to the experimental results, indicating that the method of this application is very efficient and feasible.

[0094] In order to implement the above embodiment, the present application also proposes a parameter calibration device for simulation software.

[0095] Figure 7 A schematic diagram of the structure of a parameter calibration device for simulation software provided in an embodiment of the present application.

[0096] like Figure 7 As shown, the device comprises:

[0097] The acquisition module 71 is used to acquire actual mechanical parameter values ​​of the coal rock mass determined based on experiments; wherein the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values.

[0098] The calibration module 72 is used to use the trained first prediction model to predict according to the actual compressive strength value and the actual tensile strength value of the coal rock mass, and obtain the calibration value of the contact parameter between the simulated particles of the coal rock mass in the discrete element software.

[0099] Furthermore, in a possible implementation of the embodiment of the present application, the device further includes a training module, wherein the training module is used to:

[0100] Acquire a training sample set; wherein each first training sample in the training sample set includes a simulated internal friction angle value of a simulated particle determined based on the discrete element software, and the simulated internal friction angle value marks a set value of a contact parameter between simulated particles;

[0101] Inputting each of the first training samples into the first prediction model to obtain a predicted value of the contact parameter between the simulated particles corresponding to each of the simulated internal friction angle values;

[0102] Determining a loss function of the first prediction model according to a difference between a predicted value of the simulated inter-particle contact parameter corresponding to a plurality of simulated internal friction angle values ​​and a set value of the simulated inter-particle contact parameter corresponding to a plurality of simulated internal friction angle values;

[0103] The parameters of the first prediction model are adjusted according to the loss function of the first prediction model to obtain a trained first prediction model.

[0104] In a possible implementation manner of the embodiment of the present application, the device further includes a determination module, the determination module being configured to:

[0105] Using the discrete element software to perform a stacking simulation test on the simulated particles according to a plurality of set values ​​of the contact parameters between the simulated particles, and determining the stacking angle values ​​of the simulated particles corresponding to each set value of the contact parameters between the simulated particles;

[0106] Determining the simulated internal friction angle values ​​of the simulated particles corresponding to each set value of the contact parameter between the simulated particles according to the stacking angle values ​​of the simulated particles corresponding to each set value of the contact parameter between the simulated particles;

[0107] Determining each of the first training samples according to each simulated internal friction angle value of the simulated particles and each set value of the contact parameter between the corresponding simulated particles;

[0108] The training sample set is generated according to the plurality of the first training samples.

[0109] In a possible implementation manner of the embodiment of the present application, the device further includes: a second training module, the second training module being configured to:

[0110] Determine the simulated mechanical parameter value of the coal rock mass corresponding to each set value of the bonding parameter of the simulated particles according to the multiple set values ​​of the bonding parameter between the simulated particles and the calibrated value of the contact parameter between the simulated particles using the discrete element software;

[0111] Determine each second training sample according to the simulated mechanical parameter value of the coal rock mass corresponding to each set value of the bonding parameter;

[0112] The second prediction model is trained using a plurality of second training samples to obtain a trained second prediction model.

[0113] In a possible implementation manner of the embodiment of the present application, the second training module is further used to:

[0114] Inputting each second training sample into the second prediction model to obtain a prediction value of the simulated inter-particle bonding parameter corresponding to each simulated mechanical parameter value;

[0115] Determining a loss function of the second prediction model according to a difference between a predicted value of the simulated inter-particle bonding parameter corresponding to the multiple simulated mechanical parameter values ​​and a set value of the simulated inter-particle bonding parameter corresponding to the multiple simulated mechanical parameter values;

[0116] The parameters of the second prediction model are adjusted according to the loss function of the second prediction model to obtain a trained second prediction model.

[0117] In a possible implementation of the embodiment of the present application, the calibration module 72 is further configured to:

[0118] Determine the actual internal friction angle value of the coal rock mass according to the actual compressive strength and the actual tensile strength of the coal rock mass;

[0119] The trained second prediction model is used to predict the actual internal friction angle value of the coal rock mass, and the calibration value of the bonding parameter between the simulated particles of the coal rock mass in the discrete element software is obtained. It should be noted that the above explanation of the method embodiment is also applicable to the device of this embodiment, which will not be repeated here.

[0120] The parameter calibration method of the simulation software of the embodiment of the present application obtains the actual mechanical parameter values ​​of the coal rock mass determined based on experiments, and uses the trained first prediction model to predict according to the actual compressive strength value and the actual tensile strength value of the coal rock mass, so as to obtain the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software. The trained first prediction model has learned the complex nonlinear relationship between the parameters. The actual mechanical parameter values ​​determined by the experiment are predicted by the first prediction model, and the calibration values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software are obtained, thereby realizing the rapid and accurate calibration of the contact parameters between the simulated particles in the discrete element software and improving the calibration efficiency and accuracy.

[0121] In order to implement the above embodiments, the present application also proposes a non-temporary computer-readable storage medium. When the instructions in the storage medium are executed by a processor, the method described in the above method embodiments is implemented.

[0122] In order to implement the above embodiments, the present application also proposes a computer program product. When the instruction processor in the computer program product executes, the method described in the above method embodiments is implemented.

[0123] Figure 8 A block diagram of a computer device provided in an embodiment of the present application. Figure 8The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0124] like Figure 8 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0125] The bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnection (PCI) bus.

[0126] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0127] The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 8 not shown, usually called a "hard drive"). Although Figure 8Not shown, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a compact disc read only memory (CD-ROM), a digital versatile disc read only memory (DVD-ROM), or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present application.

[0128] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0129] The computer device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, a pointing device, a display 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. In addition, the computer device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 20. As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 8 Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0130] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the methods mentioned in the above embodiments.

[0131] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0132] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0133] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0135] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0136] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0137] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0138] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for calibrating parameters of simulation software, characterized in that: include: Acquire actual mechanical parameter values ​​of the coal rock mass determined based on experiments; wherein the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values; The first prediction model obtained through training is used to perform prediction according to the actual compressive strength value and the actual tensile strength value of the coal rock mass, so as to obtain the calibration value of the contact parameter between the simulation particles of the coal rock mass in the discrete element software.

2. The method according to claim 1, characterized in that The training method of the first prediction model comprises: Acquire a training sample set; wherein each first training sample in the training sample set includes a simulated internal friction angle value of a simulated particle determined based on the discrete element software, and the simulated internal friction angle value marks a set value of a contact parameter between simulated particles; Inputting each of the first training samples into the first prediction model to obtain a predicted value of the contact parameter between the simulated particles corresponding to each of the simulated internal friction angle values; Determining a loss function of the first prediction model according to a difference between a predicted value of the simulated inter-particle contact parameter corresponding to a plurality of simulated internal friction angle values ​​and a set value of the simulated inter-particle contact parameter corresponding to a plurality of simulated internal friction angle values; The parameters of the first prediction model are adjusted according to the loss function of the first prediction model to obtain a trained first prediction model.

3. The method according to claim 2, characterized in that Before obtaining the training sample set, the following steps are included: Using the discrete element software to perform a stacking simulation test on the simulated particles according to a plurality of set values ​​of the contact parameters between the simulated particles, and determining the stacking angle values ​​of the simulated particles corresponding to each set value of the contact parameters between the simulated particles; Determining the simulated internal friction angle values ​​of the simulated particles corresponding to each set value of the contact parameter between the simulated particles according to the stacking angle values ​​of the simulated particles corresponding to each set value of the contact parameter between the simulated particles; Determining each of the first training samples according to each simulated internal friction angle value of the simulated particles and each set value of the contact parameter between the corresponding simulated particles; The training sample set is generated according to the plurality of the first training samples.

4. The method according to claim 1, characterized in that After the first prediction model obtained through training is used to predict the actual compressive strength and the actual tensile strength of the coal rock mass, and the calibrated values ​​of the contact parameters between the simulated particles of the coal rock mass in the discrete element software are obtained, the method further includes: Determine the simulated mechanical parameter value of the coal rock mass corresponding to each set value of the bonding parameter of the simulated particles according to the multiple set values ​​of the bonding parameter between the simulated particles and the calibrated value of the contact parameter between the simulated particles using the discrete element software; Determine each second training sample according to the simulated mechanical parameter value of the coal rock mass corresponding to each set value of the bonding parameter; The second prediction model is trained using a plurality of second training samples to obtain a trained second prediction model.

5. The method according to claim 4, characterized in that The method of training the second prediction model using a plurality of second training samples to obtain the trained second prediction model includes: Inputting each second training sample into the second prediction model to obtain a prediction value of the simulated inter-particle bonding parameter corresponding to each simulated mechanical parameter value; Determining a loss function of the second prediction model according to a difference between a predicted value of the simulated inter-particle bonding parameter corresponding to the multiple simulated mechanical parameter values ​​and a set value of the simulated inter-particle bonding parameter corresponding to the multiple simulated mechanical parameter values; The parameters of the second prediction model are adjusted according to the loss function of the second prediction model to obtain a trained second prediction model.

6. The method according to claim 5, characterized in that After the second prediction model is trained by using a plurality of second training samples to obtain the trained second prediction model, the method further includes: Determine the actual internal friction angle value of the coal rock mass according to the actual compressive strength and the actual tensile strength of the coal rock mass; The trained second prediction model is used to perform prediction according to the actual internal friction angle value of the coal rock mass, so as to obtain the calibration value of the bonding parameter between the simulated particles of the coal rock mass in the discrete element software.

7. A parameter calibration device for simulation software, characterized in that: include: An acquisition module, used to acquire actual mechanical parameter values ​​of the coal rock mass determined based on experiments; wherein the actual mechanical parameter values ​​include actual compressive strength values ​​and actual tensile strength values; The calibration module is used to use the first prediction model obtained through training to predict according to the actual compressive strength value and the actual tensile strength value of the coal rock mass, and obtain the calibration value of the contact parameter between the simulated particles of the coal rock mass in the discrete element software.

8. The device according to claim 7, characterized in that The device also includes a training module, wherein the training module is used to: Acquire a training sample set; wherein each training sample in the training sample set includes a simulated internal friction angle value of a simulated particle determined based on the discrete element software, and the simulated internal friction angle value marks a set value of a contact parameter between simulated particles; Inputting each of the first training samples into the first prediction model to obtain a predicted value of the contact parameter between the simulated particles corresponding to each of the simulated internal friction angle values; Determining the loss function of the first prediction model according to the difference between the predicted values ​​of the simulated inter-particle contact parameters corresponding to the multiple simulated internal friction angle values ​​and the set values ​​of the annotated simulated inter-particle contact parameters corresponding to the multiple internal friction angles; The parameters of the first prediction model are adjusted according to the loss function of the first prediction model to obtain a trained first prediction model.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images