Forging and pressing simulation method and related device

Abstract

The invention discloses a forging and pressing simulation method and a related device, and relates to the technical field of forging and pressing. Current forging and pressing simulation parameters are obtained and input into a pre-trained forging and pressing simulation model to obtain a forging and pressing simulation result; the forging and pressing simulation model is constructed by learning the relationship between forging and pressing simulation parameters and standardized forging and pressing interpolation results through a neural network. Wherein the standardized forging and pressing interpolation result is obtained by performing interpolation processing on the forging and pressing simulation result. The forging simulation result is processed by adopting a flexible interpolation method, so that the consistency of the forging simulation result and the standardized forging interpolation result before and after interpolation can be ensured, and the fixed format requirement of fixed-length output of the neural network can be met. According to the method, the time for generating the forging and pressing simulation result can be greatly shortened when the forging and pressing simulation model is used for processing the current forging and pressing simulation parameters by utilizing the rapid prediction capability of the neural network, so that the rapid prediction requirement on the forging and pressing simulation result is met.

Description

Technical Field

[0001] The present invention relates to the field of forging technology, and more specifically, to a forging simulation method and related devices. Background Art

[0002] In the forging process, the blank undergoes complex deformation under the action of external forces to form the target shape and meet specific mechanical property requirements. However, the nonlinear material behavior, large deformation and complex contact conditions involved in the forging process make process design and optimization difficult. In order to accurately predict the deformation behavior and stress distribution during the forging process, numerical simulation (such as the finite element method) has become an indispensable tool. Numerical simulation can intuitively display the deformation process by simulating process parameters (such as blank geometry, material properties and load conditions), and provide a reliable basis for tool design, defect prevention and process optimization.

[0003] In forging simulation, the finite element method (FEM) is a classic and commonly used numerical solution method. The main process includes inputting the geometric dimensions and material properties of the billet, meshing, numerical solution, and subsequent result processing. However, since the number of mesh nodes in forging simulation changes dynamically with the deformation of the billet, each solution requires a lot of time, resulting in a long time and low efficiency for forging simulation, which makes it difficult to meet the needs of rapid prediction based on forging simulation results. Summary of the invention

[0004] In view of this, the present invention discloses a forging simulation method and related devices to shorten the time of generating forging simulation results, improve the forging simulation efficiency, and meet the demand for rapid prediction based on forging simulation results.

[0005] A forging simulation method, comprising:

[0006] Get the current forging simulation parameters;

[0007] The current forging simulation parameters are input into a pre-trained forging simulation model to obtain a forging simulation result, wherein the forging simulation model is obtained by learning the correspondence between the forging simulation parameters using a neural network and the standardized forging interpolation results obtained using an interpolation method.

[0008] Optionally, the training process of the forging simulation model includes:

[0009] Determine the number of horizontal points in the horizontal direction and the number of vertical points in the vertical direction of the forging simulation result, wherein the number of horizontal points and the number of vertical points determine the resolution of the interpolation grid;

[0010] Batch acquisition uses finite element method to process forging simulation parameters and obtain finite element solutions;

[0011] For each of the finite element solutions, obtaining all corresponding boundary points;

[0012] Connecting all the boundary points in order to form a polygonal curve as the boundary of the interpolation area;

[0013] Determine each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number;

[0014] Using a preset interpolation function to interpolate at the position of each interpolation point to obtain the standardized forging interpolation result;

[0015] The forging simulation parameters are used as training samples, and the standardized forging interpolation results corresponding to the forging simulation parameters are used as sample labels, and a neural network is trained to obtain the forging simulation model.

[0016] Optionally, the determining each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number includes:

[0017] Determine the horizontal coordinate vector of the interpolation point;

[0018] Determine the corresponding interpolation point ordinate based on each interpolation point abscissa in the interpolation point abscissa vector and the number of vertical points;

[0019] Based on each of the interpolation point abscissas and a corresponding one of the interpolation point ordinates, an interpolation point is determined, and each interpolation point on and inside the polygonal curve is obtained.

[0020] Optionally, the process of determining the abscissa vector of the interpolation point includes:

[0021] Determine the maximum interpolation point abscissa and the minimum interpolation point abscissa of the polygonal curve in the horizontal direction;

[0022] According to the maximum interpolation point abscissa, the minimum interpolation point abscissa and the number of horizontal points, the distance between two adjacent interpolation points is determined to obtain the interpolation point abscissas corresponding to each interpolation point in the horizontal direction;

[0023] The interpolation point abscissa vector is obtained based on all the interpolation point abscissas.

[0024] Optionally, the determining the corresponding interpolation point ordinate based on each interpolation point abscissa in the interpolation point abscissa vector and the number of vertical points includes:

[0025] Determine an intersection vector between each of the interpolation point abscissas in the interpolation point abscissa vector and the longitudinal axis of the polygonal curve;

[0026] Based on the number of longitudinal axis intersection points contained in the longitudinal axis intersection vector and in combination with the number of vertical points, determine the number of interpolation points between adjacent intersection points;

[0027] The ordinate of each interpolation point is determined based on the number of interpolation points.

[0028] Optionally, the determining the number of interpolation points between adjacent intersection points based on the number of longitudinal axis intersection points included in the longitudinal axis intersection vector and in combination with the number of vertical points includes:

[0029] Determine the total number of regions formed by adjacent intersection points in the vertical axis intersection vector;

[0030] The number of vertical points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersections.

[0031] Optionally, the determining the number of interpolation points between adjacent intersection points based on the number of longitudinal axis intersection points included in the longitudinal axis intersection vector and in combination with the number of vertical points includes:

[0032] Determine the distance between each adjacent intersection point in the longitudinal axis intersection vector;

[0033] The number of vertical points is distributed according to each of the distances to obtain the number of interpolation points between adjacent intersections.

[0034] Optionally, the determining each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number includes:

[0035] Determine the ordinate vector of the interpolation point;

[0036] Determine the corresponding interpolation point abscissa based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points;

[0037] Based on each interpolation point ordinate and a corresponding interpolation point abscissa, an interpolation point is determined to obtain each interpolation point on and inside the polygonal curve.

[0038] Optionally, the process of determining the ordinate vector of the interpolation point includes:

[0039] Determine the maximum interpolation point ordinate and the minimum interpolation point ordinate of the polygonal curve in the vertical direction;

[0040] According to the maximum interpolation point ordinate, the minimum interpolation point ordinate and the number of vertical points, the distance between two adjacent interpolation points is determined to obtain the interpolation point ordinates corresponding to each interpolation point in the vertical direction;

[0041] The interpolation point ordinate vector is obtained based on all the interpolation point ordinates.

[0042] Optionally, the determining the corresponding interpolation point abscissa based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points includes:

[0043] Determine an intersection vector between each of the interpolation point ordinates in the interpolation point ordinate vector and the horizontal axis of the polygonal curve;

[0044] Based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points, determine the number of interpolation points between adjacent intersection points;

[0045] The horizontal coordinate of each interpolation point is determined based on the number of interpolation points.

[0046] Optionally, the determining the number of interpolation points between adjacent intersection points based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points includes:

[0047] Determine the total number of regions formed by adjacent intersection points in the horizontal axis intersection vector;

[0048] The number of horizontal points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersections.

[0049] Optionally, the determining the number of interpolation points between adjacent intersection points based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points includes:

[0050] Determine the distance between each adjacent intersection point in the horizontal axis intersection point vector;

[0051] The number of horizontal points is distributed according to each of the distances to obtain the number of interpolation points between adjacent intersection points.

[0052] A forging simulation device, comprising:

[0053] A parameter acquisition unit, used to acquire current forging simulation parameters;

[0054] A simulation unit is used to input the current forging simulation parameters into a pre-trained forging simulation model to obtain a forging simulation result, wherein the forging simulation model is obtained by learning the correspondence between the forging simulation parameters using a neural network and the standardized forging interpolation results obtained using an interpolation method.

[0055] A computer storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, any one of the forging simulation methods described above is implemented.

[0056] An electronic device, comprising: a memory and a processor;

[0057] The memory is used to store at least one instruction;

[0058] The processor is used to execute the at least one instruction to implement any one of the forging simulation methods mentioned above.

[0059] It can be seen from the above technical solutions that the present invention discloses a forging simulation method and related devices, which obtain current forging simulation parameters, input the current forging simulation parameters into a pre-trained forging simulation model to obtain forging simulation results, and the forging simulation model is obtained by using a neural network to learn the correspondence between forging simulation parameters and standardized forging interpolation results obtained by an interpolation method. The present application uses a flexible interpolation method to process the forging simulation results, which can not only ensure the consistency of the forging simulation results and the standardized forging interpolation results before and after interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network. By utilizing the rapid prediction ability of the neural network, when the forging simulation model is used to process the current forging simulation parameters, the time to generate the forging simulation results can be greatly shortened, thereby improving the forging simulation efficiency to meet the needs of rapid prediction based on the forging simulation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the disclosed drawings without paying creative work.

[0061] Figure 1 A flow chart of a forging simulation method disclosed in an embodiment of the present invention;

[0062] Figure 2 A flow chart of a training method for a forging simulation model disclosed in an embodiment of the present invention;

[0063] Figure 3 A flow chart of a method for determining interpolation points on and inside a polygonal curve based on the number of horizontal points and the number of vertical points disclosed in an embodiment of the present invention;

[0064] FIG4 (1) is a schematic diagram of various interpolation points on a polygonal curve disclosed in an embodiment of the present invention;

[0065] FIG4 (2) is a schematic diagram of interpolation points on another polygonal curve disclosed in an embodiment of the present invention;

[0066] Figure 5 A flow chart of a method for determining interpolation points on and inside a polygonal curve based on the number of horizontal points and the number of vertical points disclosed in an embodiment of the present invention;

[0067] Figure 6 A schematic diagram of interpolation points on another polygonal curve disclosed in an embodiment of the present invention;

[0068] Figure 7 A schematic diagram of the structure of a forging simulation device disclosed in an embodiment of the present invention;

[0069] Figure 8 The present invention is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] The embodiment of the present invention discloses a forging simulation method and related devices, which obtain current forging simulation parameters, input the current forging simulation parameters into a pre-trained forging simulation model to obtain forging simulation results, and the forging simulation model is obtained by using a neural network to learn the correspondence between forging simulation parameters and standardized forging interpolation results obtained by an interpolation method. The present application uses a flexible interpolation method to process the forging simulation results, which can not only ensure the consistency of the forging simulation results and the standardized forging interpolation results before and after the interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network. By utilizing the rapid prediction capability of the neural network, when the forging simulation model is applied to process the current forging simulation parameters, the time for generating the forging simulation results can be greatly shortened, thereby improving the forging simulation efficiency to meet the needs of rapid prediction based on the forging simulation results.

[0072] See also Figure 1 , a flow chart of a forging simulation method disclosed in an embodiment of the present application, the method comprising:

[0073] Step S101, obtaining current forging simulation parameters.

[0074] Among them, the current forging simulation parameters mainly include material properties, geometric parameters, boundary conditions, loading conditions, etc.

[0075] Material properties include: yield strength, hardness, tensile strength, etc.

[0076] Geometric parameters include: parameters that directly affect the size and shape of the forging, such as length, width, thickness, etc., and parameters that affect the stress distribution and forming performance of the forging during the forming process, such as cutting width.

[0077] Boundary conditions and loading cases include:

[0078] Firmware constraints: In simulation, it is necessary to impose fixed constraints on certain parts to simulate the fixed state in actual production.

[0079] Load application: Apply corresponding loads such as pressure, temperature, etc. according to the forging process requirements.

[0080] Step S102: input the current forging simulation parameters into a pre-trained forging simulation model to obtain a forging simulation result.

[0081] The forging simulation model is obtained by using a neural network to learn the corresponding relationship between forging simulation parameters and standardized forging interpolation results obtained by an interpolation method.

[0082] Among them, the number of points contained in all standardized forging interpolation results in this application is consistent to meet the fixed-length output requirements of the neural network.

[0083] The main function of the interpolation method is to estimate or predict the value of unknown data points between known data points. This application uses the interpolation method to process the forging simulation parameters, which can not only ensure the consistency of the finite element solution obtained by using the finite element method to process the forging simulation parameters before and after interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network.

[0084] The predictive ability of neural networks has many advantages, such as processing complex nonlinear problems, high-precision prediction, strong generalization ability, strong adaptability, automatic feature extraction, large-scale data processing ability, and integration with other machine learning technologies. These advantages make neural networks play an important role in the field of predictive analysis. This application obtains a forging simulation model by training the forging interpolation results with a neural network, and uses the predictive ability of the neural network to greatly shorten the forging simulation time.

[0085] In summary, the present application discloses a forging simulation method, which obtains current forging simulation parameters, and inputs the current forging simulation parameters into a pre-trained forging simulation model to obtain forging simulation results. The forging simulation model is obtained by using a neural network to learn the correspondence between forging simulation parameters and standardized forging interpolation results obtained by an interpolation method. The present application uses a flexible interpolation method to process the forging simulation results, which can not only ensure the consistency of the forging simulation results and the standardized forging interpolation results before and after interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network. By utilizing the rapid prediction capability of the neural network, when the forging simulation model is applied to process the current forging simulation parameters, the time to generate the forging simulation results can be greatly shortened, thereby improving the forging simulation efficiency to meet the needs of rapid prediction based on the forging simulation results.

[0086] See also Figure 2 , a flow chart of a training method for a forging simulation model disclosed in an embodiment of the present application, the method comprising:

[0087] Step S201, determining the number of horizontal points in the horizontal direction and the number of vertical points in the vertical direction of the forging simulation results.

[0088] Before starting interpolation, it is first necessary to determine the number of horizontal points Nx in the horizontal direction (x direction) and the number of vertical points Ny in the vertical direction (y direction) of the forging simulation results.

[0089] The number of horizontal points Nx and the number of vertical points Ny determine the resolution of the interpolation grid, that is, how many interpolation points will be generated in the horizontal direction and how many interpolation points will be generated in the vertical direction.

[0090] The values ​​of the horizontal point number Nx and the vertical point number Ny can be determined based on the accuracy of the forging simulation results and computing resources.

[0091] Step S202: Batch obtain all boundary points of the finite element solution obtained after processing the forging simulation parameters using the finite element method.

[0092] Finite Element Method (FEM) is a numerical method for solving mathematical and physical problems. The basic idea of ​​the finite element method is to discretize the continuous solution domain into a finite number of small units, which are connected to each other through nodes to form a grid. Based on this, the finite element method solves the physical quantities on the grid nodes, such as displacement and stress, rather than the physical quantities in the entire area. This application uses the finite element method to solve different forging simulation parameters to obtain different finite element solutions.

[0093] Step S203: For each of the finite element solutions, obtain all corresponding boundary points.

[0094] Step S204: connect all the boundary points in sequence to form a polygonal curve serving as the boundary of the interpolation area.

[0095] All boundary points define the geometric boundary of the interpolation region of interest. The present application connects all boundary points in sequence to form a closed polygon curve, which represents the boundary of the interpolation region.

[0096] Step S205: determine each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number.

[0097] The horizontal point number Nx determines how many interpolation points will be generated in the horizontal direction, and the vertical point number Ny determines how many interpolation points will be generated in the vertical direction. Therefore, based on the horizontal point number Nx and the vertical point number Ny, the interpolation points on and inside the polygon curve can be determined.

[0098] Step S206: using a preset interpolation function to perform interpolation at the position of each interpolation point to obtain the standardized forging interpolation result.

[0099] The preset interpolation function in this application can be any one or more combinations of linear interpolation, nearest neighbor interpolation, cubic spline interpolation, bilinear interpolation, bicubic interpolation, Lagrange interpolation, Newton's interpolation, Chebyshev interpolation, radial basis function interpolation and multivariate interpolation. It can be selected according to the data characteristics and the required forging simulation accuracy, and this application does not limit it here.

[0100] Step S207: using the forging simulation parameters as training samples, using the standardized forging interpolation results corresponding to the forging simulation parameters as sample labels, and training a neural network to obtain the forging simulation model.

[0101] In summary, the present application adopts the interpolation method to process the forging simulation parameters when training the forging simulation model. Therefore, it can not only ensure the accuracy of the forging interpolation results before and after the interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network. The neural network can quickly obtain the prediction ability of the results, so that when the forging simulation model is used to process the current forging simulation parameters, the time for generating the forging simulation results can be greatly shortened, and the forging simulation efficiency can be improved to meet the needs of rapid prediction based on the forging simulation results.

[0102] In one embodiment, see Figure 3 , a flowchart of a method for determining each interpolation point on and inside a polygonal curve based on the number of horizontal points and the number of vertical points disclosed in an embodiment of the present application, that is, step S204 may specifically include:

[0103] Step S301, determine the horizontal coordinate vector of the interpolation point.

[0104] Specifically, the maximum interpolation point abscissa xmax and the minimum interpolation point abscissa xmin of the polygonal curve in the horizontal direction are determined.

[0105] According to the maximum interpolation point abscissa xmax, the minimum interpolation point abscissa xmin and the number of horizontal points Nx, the distance between two adjacent interpolation points is determined to obtain the interpolation point abscissas corresponding to each interpolation point in the horizontal direction.

[0106] An interpolation point abscissa vector is obtained based on all the interpolation point abscissas.

[0107] It should be noted that the maximum interpolation point horizontal coordinate xmax and the minimum interpolation point horizontal coordinate xmin define the interpolation range along the horizontal direction. According to the interpolation range in the horizontal direction and the number of horizontal points Nx, the interpolation point horizontal coordinate corresponding to each interpolation point in the horizontal direction can be calculated.

[0108] The interpolation point abscissa vector x can be expressed as:

[0109] x=[xmin, xmin+Δx,…,xmax].

[0110] Among them, △x is the distance between two adjacent interpolation points in the horizontal direction.

[0111] The expression of △x is as follows:

[0112] .

[0113] Step S302: Determine the corresponding interpolation point vertical coordinate based on each interpolation point horizontal coordinate in the interpolation point horizontal coordinate vector and the number of vertical points.

[0114] Step S303: determine an interpolation point based on each of the interpolation point abscissas and a corresponding interpolation point ordinate, and obtain interpolation points on and inside the polygonal curve.

[0115] Each interpolation point is composed of an interpolation point horizontal coordinate and an interpolation point vertical coordinate. When all the interpolation point horizontal coordinates and their corresponding interpolation point vertical coordinates are determined, the interpolation points on and inside the polygonal curve can be obtained. For details, see Figures 4 (1) and 4 (2), which are schematic diagrams of the interpolation points on and inside the polygonal curve respectively.

[0116] In one embodiment, based on each interpolation point abscissa and the number of vertical points in the interpolation point abscissa vector, the process of determining the corresponding interpolation point ordinate may specifically include:

[0117] Determine the intersection vector of each interpolation point abscissa in the interpolation point abscissa vector and the ordinate of the polygonal curve;

[0118] Based on the number of longitudinal axis intersection points contained in the longitudinal axis intersection vector and in combination with the number of vertical points, determine the number of interpolation points between adjacent intersection points;

[0119] The ordinate of each interpolation point is determined based on the number of interpolation points.

[0120] Among them, there are many methods for determining the number of interpolation points between adjacent intersections based on the number of longitudinal axis intersection points contained in the longitudinal axis intersection vector and in combination with the number of vertical points. This application provides two implementation methods, which are as follows:

[0121] Method 1

[0122] Determine the total number of regions formed by adjacent intersection points in the vertical axis intersection vector;

[0123] The number of vertical points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersection points.

[0124] Let's take an example to illustrate:

[0125] The number of vertical points is represented by Ny.

[0126] If there are two intersection points in the vertical axis intersection vector J, namely J1 and J2, then Ny vertical coordinate points y are evenly distributed between J1 and J2, that is, the number of interpolation points between J1 and J2 is Ny.

[0127] If there are three intersection points in the vertical axis intersection vector J, namely: J1, J2 and J3, the area between J1 and J2 and the area between J2 and J3 can be interpolated separately, with Ny / 2 points interpolated in each area, or unequal interpolation can be performed according to the size and shape of the area.

[0128] If there are four intersection points in the vertical axis intersection vector J, namely: J1, J2, J3 and J4, then the area between J1 and J2 and the area between J3 and J4 can be interpolated respectively, and Ny / 2 points are interpolated in each area.

[0129] If the number of intersection points in the longitudinal axis intersection vector J exceeds four, it is necessary to reasonably distribute the Ny points between the segments according to the shape of the polygonal curve and the distribution of the intersection points.

[0130] Method 2:

[0131] Determine the distance between each adjacent intersection point in the longitudinal axis intersection vector;

[0132] The number of vertical points is distributed according to each distance to obtain the number of interpolation points between adjacent intersection points.

[0133] In practical applications, Ny can also be allocated according to the distance between intersection points.

[0134] For example, assuming that there are four intersection points in the vertical axis intersection vector J, namely: J1, J2, J3 and J4, the interpolation can be performed according to the following steps:

[0135] 1. Calculate the distance d1 between J1 and J2 and the distance d2 between J3 and J4;

[0136] 2. According to the ratio of d1 and d2, Ny points are distributed to the two regions in proportion. For example, if the ratio of d1 and d2 is 2:1, 2Ny / 3 ordinate points y are interpolated between J1 and J2, and Ny / 3 ordinate points y are interpolated between J3 and J4. If Ny cannot be divided evenly, one more point can be considered in one interval and one less point in the other interval to ensure the consistency of the total number of points.

[0137] In one embodiment, see Figure 5 , a flowchart of a method for determining each interpolation point on and inside a polygonal curve based on the number of horizontal points and the number of vertical points disclosed in an embodiment of the present application may specifically include:

[0138] Step S401, determine the vertical coordinate vector of the interpolation point.

[0139] Specifically, determining the maximum interpolation point ordinate ymax and the minimum interpolation point ordinate ymin of the polygonal curve in the vertical direction;

[0140] According to the maximum interpolation point ordinate ymax, the minimum interpolation point ordinate ymin and the vertical point number Ny, the distance between two adjacent interpolation points is determined to obtain the interpolation point ordinates corresponding to each interpolation point in the vertical direction;

[0141] The interpolation point ordinate vector is obtained based on all the interpolation point ordinates.

[0142] It should be noted that the maximum interpolation point ordinate ymax and the minimum interpolation point ordinate ymin define the interpolation range along the horizontal direction. According to the interpolation range in the horizontal direction and the number of horizontal points Ny, the interpolation point ordinate corresponding to each interpolation point in the horizontal direction can be calculated.

[0143] The interpolation point ordinate vector y can be expressed as:

[0144] y=[ymin, ymin+△y,…,ymax].

[0145] Among them, △y is the distance between two adjacent interpolation points in the vertical direction.

[0146] The expression of △y is as follows:

[0147] .

[0148] Step S402: Determine the corresponding interpolation point horizontal coordinate based on each interpolation point vertical coordinate in the interpolation point vertical coordinate vector and the number of horizontal points.

[0149] Step S403: determine an interpolation point based on the ordinate of each interpolation point and a corresponding abscissa of the interpolation point, and obtain each interpolation point on the polygonal curve and inside the polygonal curve.

[0150] Each interpolation point consists of an interpolation point abscissa and an interpolation point ordinate. When all the interpolation point abscissas and their corresponding interpolation point ordinates are determined, the interpolation points on and inside the polygonal curve can be obtained. For details, see Figure 6 Schematic diagram of various interpolation points on the polygonal curve shown.

[0151] In one embodiment, based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points, the process of determining the corresponding interpolation point abscissa may specifically include:

[0152] Determine an intersection vector between each of the interpolation point ordinates in the interpolation point ordinate vector and the horizontal axis of the polygonal curve;

[0153] Based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points, determine the number of interpolation points between adjacent intersection points;

[0154] The horizontal coordinate of each interpolation point is determined based on the number of interpolation points.

[0155] There are many methods for determining the number of interpolation points between adjacent intersection points based on the number of intersection points contained in the horizontal axis intersection vector and the number of horizontal points. This application provides two implementation methods, which are as follows:

[0156] Method 1

[0157] Determine the total number of regions formed by adjacent intersection points in the horizontal axis intersection vector;

[0158] The number of horizontal points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersection points.

[0159] Let's take an example to illustrate:

[0160] The number of horizontal points is represented by Nx.

[0161] If there are two intersection points in the horizontal axis intersection vector K, namely K1 and K2, then Nx horizontal coordinate points y are evenly distributed between K1 and K2, that is, the number of interpolation points between K1 and K2 is Nx.

[0162] If there are three intersection points in the horizontal axis intersection vector K, namely: K1, K2 and K3, the area between K1 and K2 and the area between K2 and K3 can be interpolated separately, with Nx / 2 points interpolated in each area, or unequal interpolation can be performed according to the size and shape of the area.

[0163] If there are four intersection points in the horizontal axis intersection vector K, namely: K1, K2, K3 and K4, then the area between K1 and K2 and the area between K3 and K4 can be interpolated respectively, and Nx / 2 points are interpolated in each area.

[0164] If the number of intersection points in the horizontal axis intersection vector K exceeds four, it is necessary to reasonably distribute the Nx points between the segments according to the shape of the polygonal curve and the distribution of the intersection points.

[0165] Method 2:

[0166] Determine the distance between each adjacent intersection point in the horizontal axis intersection vector;

[0167] Allocate the number of horizontal points according to each distance to obtain the number of interpolation points between adjacent intersection points.

[0168] In practical applications, Nx may also be allocated according to the distances between intersection points.

[0169] For example, assuming that there are four intersection points in the vertical axis intersection vector K, namely: K1, K2, K3 and K4, the interpolation can be performed according to the following steps:

[0170] 1. Calculate the distance d1 between K1 and K2 and the distance d2 between K3 and K4;

[0171] 2. According to the ratio of d1 and d2, distribute Nx points to the two areas in proportion. For example, if the ratio of d1 and d2 is 2:1, interpolate 2Nx / 3 ordinate points x between K1 and K2, and interpolate Nx / 3 ordinate points x between K3 and K4. If Nx cannot be divided evenly, consider adding one more point in one interval and one less point in the other interval to ensure the consistency of the total number of points.

[0172] Corresponding to the above method embodiments, the present application discloses a forging simulation device.

[0173] See also Figure 7 , a schematic diagram of the structure of a forging simulation device disclosed in an embodiment of the present application, the device may include:

[0174] The parameter acquisition unit 501 is used to acquire the current forging simulation parameters.

[0175] Among them, the current forging simulation parameters mainly include material properties, geometric parameters, boundary conditions, loading conditions, etc.

[0176] The simulation unit 502 is used to input the current forging simulation parameters into a pre-trained forging simulation model to obtain a forging simulation result.

[0177] The forging simulation model is obtained by learning the correspondence between forging simulation parameters using a neural network and standardized forging interpolation results obtained using an interpolation method.

[0178] The main function of the interpolation method is to estimate or predict the value of unknown data points between known data points. This application uses the interpolation method to process the forging simulation parameters, which can not only ensure the consistency of the finite element solution obtained by using the finite element method to process the forging simulation parameters before and after interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network.

[0179] The predictive ability of neural networks has many advantages, such as processing complex nonlinear problems, high-precision prediction, strong generalization ability, strong adaptability, automatic feature extraction, large-scale data processing ability, and integration with other machine learning technologies. These advantages make neural networks play an important role in the field of predictive analysis. This application obtains a forging simulation model by training the forging interpolation results with a neural network, and uses the predictive ability of the neural network to greatly shorten the forging simulation time.

[0180] In summary, the present application discloses a forging simulation device, which obtains current forging simulation parameters, and inputs the current forging simulation parameters into a pre-trained forging simulation model to obtain forging simulation results. The forging simulation model is obtained by using a neural network to learn the correspondence between forging simulation parameters and standardized forging interpolation results obtained by an interpolation method. The present application uses a flexible interpolation method to process the forging simulation results, which can not only ensure the consistency of the forging simulation results and the standardized forging interpolation results before and after interpolation, but also meet the fixed format requirements of the fixed-length output of the neural network. By utilizing the rapid prediction capability of the neural network, when the forging simulation model is applied to process the current forging simulation parameters, the time to generate the forging simulation results can be greatly shortened, thereby improving the forging simulation efficiency to meet the needs of rapid prediction based on the forging simulation results.

[0181] In one embodiment, the forging simulation device may further include:

[0182] Model training unit, used to train forging simulation model.

[0183] The model training unit can be used specifically for:

[0184] Determine the number of horizontal points in the horizontal direction and the number of vertical points in the vertical direction of the forging simulation result, wherein the number of horizontal points and the number of vertical points determine the resolution of the interpolation grid;

[0185] Batch acquisition uses finite element method to process forging simulation parameters and obtain finite element solutions;

[0186] For each of the finite element solutions, obtaining all corresponding boundary points;

[0187] Connecting all the boundary points in order to form a polygonal curve as the boundary of the interpolation area;

[0188] Determine each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number;

[0189] Using a preset interpolation function to interpolate at the position of each interpolation point to obtain the standardized forging interpolation result;

[0190] The forging simulation parameters are used as training samples, and the standardized forging interpolation results corresponding to the forging simulation parameters are used as sample labels, and a neural network is trained to obtain the forging simulation model.

[0191] In one embodiment, the process of the model training unit determining each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number may specifically include:

[0192] Determine the horizontal coordinate vector of the interpolation point;

[0193] Determine the corresponding interpolation point ordinate based on each interpolation point abscissa in the interpolation point abscissa vector and the number of vertical points;

[0194] Based on each of the interpolation point abscissas and a corresponding one of the interpolation point ordinates, an interpolation point is determined, and each interpolation point on and inside the polygonal curve is obtained.

[0195] In one embodiment, the process of determining the abscissa vector of the interpolation point by the model training unit may specifically include:

[0196] Determine the maximum interpolation point abscissa and the minimum interpolation point abscissa of the polygonal curve in the horizontal direction;

[0197] According to the maximum interpolation point abscissa, the minimum interpolation point abscissa and the number of horizontal points, the distance between two adjacent interpolation points is determined to obtain the interpolation point abscissas corresponding to each interpolation point in the horizontal direction;

[0198] The interpolation point abscissa vector is obtained based on all the interpolation point abscissas.

[0199] In one embodiment, the model training unit determines the corresponding interpolation point vertical coordinate based on each interpolation point horizontal coordinate in the interpolation point horizontal coordinate vector and the number of vertical points, including:

[0200] Determine an intersection vector between each of the interpolation point abscissas in the interpolation point abscissa vector and the longitudinal axis of the polygonal curve;

[0201] Based on the number of longitudinal axis intersection points contained in the longitudinal axis intersection vector and in combination with the number of vertical points, determine the number of interpolation points between adjacent intersection points;

[0202] The ordinate of each interpolation point is determined based on the number of interpolation points.

[0203] In one embodiment, the process of the model training unit determining the number of interpolation points between adjacent intersection points based on the number of longitudinal axis intersection points included in the longitudinal axis intersection vector and in combination with the number of vertical points may specifically include:

[0204] Determine the total number of regions formed by adjacent intersection points in the vertical axis intersection vector;

[0205] The number of vertical points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersections.

[0206] In one embodiment, the model training unit determines the number of interpolation points between adjacent intersection points based on the number of longitudinal axis intersection points included in the longitudinal axis intersection vector and in combination with the number of vertical points, including:

[0207] Determine the distance between each adjacent intersection point in the longitudinal axis intersection vector;

[0208] The number of vertical points is distributed according to each of the distances to obtain the number of interpolation points between adjacent intersections.

[0209] In one embodiment, the model training unit determines each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number, including:

[0210] Determine the ordinate vector of the interpolation point;

[0211] Determine the corresponding interpolation point abscissa based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points;

[0212] Based on each interpolation point ordinate and a corresponding interpolation point abscissa, an interpolation point is determined to obtain each interpolation point on and inside the polygonal curve.

[0213] In one embodiment, the process of determining the ordinate vector of the interpolation point by the model training unit includes:

[0214] Determine the maximum interpolation point ordinate and the minimum interpolation point ordinate of the polygonal curve in the vertical direction;

[0215] According to the maximum interpolation point ordinate, the minimum interpolation point ordinate and the number of vertical points, the distance between two adjacent interpolation points is determined to obtain the interpolation point ordinates corresponding to each interpolation point in the vertical direction;

[0216] The interpolation point ordinate vector is obtained based on all the interpolation point ordinates.

[0217] In one embodiment, the model training unit determines the corresponding interpolation point abscissa based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points, including:

[0218] Determine an intersection vector between each of the interpolation point ordinates in the interpolation point ordinate vector and the horizontal axis of the polygonal curve;

[0219] Based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points, determine the number of interpolation points between adjacent intersection points;

[0220] The horizontal coordinate of each interpolation point is determined based on the number of interpolation points.

[0221] It should be noted that for the specific working principles of each component in the device embodiment, please refer to the corresponding part of the method embodiment, which will not be repeated here.

[0222] Corresponding to the above-mentioned embodiment, the present application further discloses a computer storage medium, which stores at least one instruction, and when the at least one instruction is executed by a processor, the steps shown in the forging simulation method embodiment are implemented.

[0223] Corresponding to the above embodiment, Figure 8 As shown, the present invention also provides an electronic device, which may include: a processor 1 and a memory 2;

[0224] The processor 1 and the memory 2 communicate with each other via a communication bus 3;

[0225] Processor 1, configured to execute at least one instruction;

[0226] Memory 2, used to store at least one instruction;

[0227] The processor 1 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0228] The memory 2 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0229] Among them, the processor executes at least one instruction to implement the steps shown in the forging simulation method embodiment.

[0230] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0231] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0232] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A forging simulation method, characterized in that: include: Get the current forging simulation parameters; The current forging simulation parameters are input into a pre-trained forging simulation model to obtain a forging simulation result, wherein the forging simulation model is obtained by learning the correspondence between the forging simulation parameters using a neural network and the standardized forging interpolation results obtained using an interpolation method.

2. The forging simulation method according to claim 1, characterized in that: The training process of the forging simulation model includes: Determine the number of horizontal points in the horizontal direction and the number of vertical points in the vertical direction of the forging simulation result, wherein the number of horizontal points and the number of vertical points determine the resolution of the interpolation grid; Batch acquisition uses finite element method to process forging simulation parameters and obtain finite element solutions; For each of the finite element solutions, obtaining all corresponding boundary points; Connecting all the boundary points in order to form a polygonal curve as the boundary of the interpolation area; Determine each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number; Using a preset interpolation function to interpolate at the position of each interpolation point to obtain the standardized forging interpolation result; The forging simulation parameters are used as training samples, and the standardized forging interpolation results corresponding to the forging simulation parameters are used as sample labels, and a neural network is trained to obtain the forging simulation model.

3. The forging simulation method according to claim 2, characterized in that: Determining each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number includes: Determine the horizontal coordinate vector of the interpolation point; Determine the corresponding interpolation point ordinate based on each interpolation point abscissa in the interpolation point abscissa vector and the number of vertical points; Based on each of the interpolation point abscissas and a corresponding one of the interpolation point ordinates, an interpolation point is determined, and each interpolation point on and inside the polygonal curve is obtained.

4. The forging simulation method according to claim 3, characterized in that: The process of determining the interpolation point horizontal coordinate vector includes: Determine the maximum interpolation point abscissa and the minimum interpolation point abscissa of the polygonal curve in the horizontal direction; According to the maximum interpolation point abscissa, the minimum interpolation point abscissa and the number of horizontal points, the distance between two adjacent interpolation points is determined to obtain the interpolation point abscissas corresponding to each interpolation point in the horizontal direction; The interpolation point abscissa vector is obtained based on all the interpolation point abscissas.

5. The forging simulation method according to claim 3 or 4, characterized in that: The determining the corresponding interpolation point ordinate based on each interpolation point abscissa in the interpolation point abscissa vector and the number of vertical points comprises: Determine an intersection vector between each of the interpolation point abscissas in the interpolation point abscissa vector and the longitudinal axis of the polygonal curve; Based on the number of longitudinal axis intersection points contained in the longitudinal axis intersection vector and in combination with the number of vertical points, determine the number of interpolation points between adjacent intersection points; The ordinate of each interpolation point is determined based on the number of interpolation points.

6. The forging simulation method according to claim 5, characterized in that: The determining the number of interpolation points between adjacent intersection points based on the number of longitudinal axis intersection points included in the longitudinal axis intersection vector and in combination with the number of vertical distribution points includes: Determine the total number of regions formed by adjacent intersection points in the vertical axis intersection vector; The number of vertical points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersections.

7. The forging simulation method according to claim 5, characterized in that: The determining the number of interpolation points between adjacent intersection points based on the number of longitudinal axis intersection points included in the longitudinal axis intersection vector and in combination with the number of vertical distribution points includes: Determine the distance between each adjacent intersection point in the longitudinal axis intersection vector; The number of vertical points is distributed according to each of the distances to obtain the number of interpolation points between adjacent intersections.

8. The forging simulation method according to claim 2, characterized in that: Determining each interpolation point on and inside the polygonal curve based on the horizontal point number and the vertical point number includes: Determine the ordinate vector of the interpolation point; Determine the corresponding interpolation point abscissa based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points; Based on each interpolation point ordinate and a corresponding interpolation point abscissa, an interpolation point is determined to obtain each interpolation point on and inside the polygonal curve.

9. The forging simulation method according to claim 8, characterized in that: The process of determining the interpolation point ordinate vector comprises: Determine the maximum interpolation point ordinate and the minimum interpolation point ordinate of the polygonal curve in the vertical direction; According to the maximum interpolation point ordinate, the minimum interpolation point ordinate and the number of vertical points, the distance between two adjacent interpolation points is determined to obtain the interpolation point ordinates corresponding to each interpolation point in the vertical direction; The interpolation point ordinate vector is obtained based on all the interpolation point ordinates.

10. The forging simulation method according to claim 8 or 9, characterized in that: The determining the corresponding interpolation point abscissa based on each interpolation point ordinate in the interpolation point ordinate vector and the number of horizontal points comprises: Determine an intersection vector between each of the interpolation point ordinates in the interpolation point ordinate vector and the horizontal axis of the polygonal curve; Based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points, determine the number of interpolation points between adjacent intersection points; The horizontal coordinate of each interpolation point is determined based on the number of interpolation points.

11. The forging simulation method according to claim 10, characterized in that: The determining the number of interpolation points between adjacent intersection points based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points includes: Determine the total number of regions formed by adjacent intersection points in the horizontal axis intersection vector; The number of horizontal points is distributed according to the total number of regions to obtain the number of interpolation points between adjacent intersections.

12. The forging simulation method according to claim 10, characterized in that: The determining the number of interpolation points between adjacent intersection points based on the number of intersection points included in the horizontal axis intersection vector and in combination with the number of horizontal points includes: Determine the distance between each adjacent intersection point in the horizontal axis intersection point vector; The number of horizontal points is distributed according to each of the distances to obtain the number of interpolation points between adjacent intersection points.

13. A forging simulation device, characterized in that: include: A parameter acquisition unit, used to acquire current forging simulation parameters; A simulation unit is used to input the current forging simulation parameters into a pre-trained forging simulation model to obtain a forging simulation result, wherein the forging simulation model is obtained by learning the correspondence between the forging simulation parameters using a neural network and the standardized forging interpolation results obtained using an interpolation method.

14. A computer storage medium, characterized in that: The computer storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, the forging simulation method according to any one of claims 1 to 12 is implemented.

15. An electronic device, characterized in that: The electronic device comprises: a memory and a processor; The memory is used to store at least one instruction; The processor is used to execute the at least one instruction to implement the forging simulation method as described in any one of claims 1 to 12.

Images