Visual analysis method and device for straightening machine production process based on agent model

Abstract

This invention relates to the field of digital twins, specifically to a method and apparatus for visual analysis of the production process of a straightening machine based on a surrogate model. The method includes: establishing a finite element model of a roller straightening machine based on straightening machine parameters; obtaining steel plate stress parameter data during the straightening process based on steel plate production process parameters and the finite element model; constructing a surrogate model for predicting the stress state of the steel plate based on the pre-processed steel plate stress parameter data, and obtaining the prediction results of the surrogate model; and realizing dynamic real-time mapping of the stress state during the steel plate straightening production process based on the prediction results of the surrogate model and a digital twin platform for the straightening machine. This invention utilizes finite element simulation modeling to obtain data that is difficult to measure online during actual production, providing a complete dataset for establishing the surrogate model. By developing a digital twin platform for the straightening machine, the transparency of the steel plate straightening production process is improved, while also providing operators with powerful interactive capabilities.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and in particular to a method and device for visualizing and analyzing a production process of a straightening machine based on an agent model. Background Art

[0002] Straightening machines are crucial for eliminating residual stress and improving plate quality. In order to improve the transparency of the steel plate straightening production process, how to achieve real-time interactive mapping and visualization of the production process and improve the interactivity and intelligence level of the steel plate straightening production process are issues that need to be urgently addressed by existing technologies. Summary of the invention

[0003] In order to solve the technical problem of how to realize real-time interactive mapping and visualization of the production process in the prior art and improve the interactivity and intelligence level of the steel plate straightening production process, the embodiment of the present invention provides a method and device for visual analysis of the straightening machine production process based on an agent model. The technical solution is as follows:

[0004] On the one hand, a visualization analysis method for a straightening machine production process based on an agent model is provided, the method being implemented by a visualization analysis device for a straightening machine production process, the method comprising:

[0005] S1. Obtain straightening machine parameters, and establish a roller straightening machine finite element model according to the straightening machine parameters.

[0006] S2. Obtain steel plate production process parameters, perform finite element simulation based on the steel plate production process parameters and the roller straightening machine finite element model, and obtain steel plate stress parameter data during the straightening process.

[0007] S3. Preprocess the steel plate stress parameter data, construct a steel plate stress state prediction proxy model according to the preprocessed steel plate stress parameter data, and obtain the proxy model prediction result according to the constructed steel plate stress state prediction proxy model.

[0008] S4. Develop a digital twin platform for the straightening machine, and realize dynamic real-time mapping of the stress state of the steel plate straightening production process based on the prediction results of the agent model and the digital twin platform of the straightening machine.

[0009] Optionally, the straightening machine parameters in S1 include: straightening roller diameter, straightening roller length, straightening roller pitch and the number of rollers.

[0010] The steel plate production process parameters in S2 include: steel type, width, thickness, yield limit and elastic modulus.

[0011] Optionally, the process of establishing the finite element model of the roller straightening machine in S1 includes:

[0012] The reverse bending deflection during the straightening process is calculated according to the following formula (1), and the accuracy of the finite element model of the roller straightening machine is determined according to the reverse bending deflection:

[0013] Δf=Δl+Δc

[0014]

[0015]

[0016]

[0017] Where Δf is the reverse bending deflection, Δl is the elastic complex deflection, Δc is the residual deflection, M is the elastic-plastic bending moment, L is the straightening roller distance, E is the elastic modulus of the material, I is the moment of inertia of the material cross section about the neutral axis, σ s is the yield stress, b is the width of the steel plate, h is the thickness of the steel plate, ε s is the maximum elastic strain of the material, is the total deformation curvature, E′ is the work hardening strength modulus of the material, Δw is the yield deflection, and C is the ratio of the residual curvature to the yield curvature.

[0018] Optionally, the preprocessing of the steel plate stress parameter data in S3 includes:

[0019] S31. Use the local outlier factor detection method LOF to detect outliers on the stress parameter data of the steel plate to obtain abnormal points and non-abnormal points.

[0020] Among them, the rule of the local outlier factor detection method LOF is as shown in the following formula (2):

[0021]

[0022] S32, obtaining the stress parameter data of the steel plate at the front and rear positions of the abnormal point, and calculating the mean value according to the following formula (3), and using the mean value to replace the abnormal point.

[0023]

[0024] Where D new The outlier point D i Alternative data, D i-1 is the stress parameter data of the steel plate at the previous position of the abnormal point, D i+1 is the stress parameter data of the steel plate at the position after the abnormal point, and i represents the i-th data point.

[0025] Optionally, constructing a steel plate stress state prediction agent model according to the preprocessed steel plate stress parameter data in S3 includes:

[0026] According to the preprocessed stress parameter data of steel plates, the three-swarm particle swarm differential evolution (PSODE) algorithm is adopted to optimize the lightweight gradient boosting machine (LightGBM) and construct an agent model for predicting the stress state of steel plates.

[0027] Optionally, three swarm particle swarm differential evolution PSODE algorithms include:

[0028] Two of the three populations are used for differential evolution DE optimization, and the other population except the two populations is used for particle swarm PSO optimization.

[0029] Among them, the mutation mode of differential evolution DE adopts the combination of the current best mode and the best mode.

[0030] The base point vector of the best mode is the best individual of the previous generation population, as shown in the following formula (4):

[0031]

[0032] Where V i is a random new individual, G is the evolutionary generation, r 1 , r 2 , r 3 ∈[1,N],r 1 ≠r 2 ≠r 3 ≠i, N is the total number of individuals, X best，G is the optimal individual, F is a constant in the range of [0,1], called the variation factor, are base point vectors.

[0033] The base point vector of the current-best mode is the combination of the target individual and the optimal individual, as shown in the following formula (5):

[0034]

[0035] In the formula, X i，G is the base point vector.

[0036] Optionally, a three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the lightweight gradient boosting machine LightGBM, including:

[0037] The three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the learning rate and learning cycle combination of the lightweight gradient boosting machine LightGBM.

[0038] On the other hand, a visualization analysis device for a straightening machine production process based on an agent model is provided, and the device is applied to a visualization analysis method for a straightening machine production process based on an agent model, and the device comprises:

[0039] The finite element model building module is used to obtain the parameters of the straightening machine and establish the finite element model of the roller straightening machine according to the parameters of the straightening machine.

[0040] The finite element simulation module is used to obtain the steel plate production process parameters, perform finite element simulation based on the steel plate production process parameters and the roller straightening machine finite element model, and obtain the steel plate stress parameter data during the straightening process.

[0041] The prediction module is used to preprocess the steel plate stress parameter data, construct a steel plate stress state prediction agent model according to the preprocessed steel plate stress parameter data, and obtain the agent model prediction result according to the constructed steel plate stress state prediction agent model.

[0042] The output module is used to develop a digital twin platform for the straightening machine. Based on the prediction results of the agent model and the digital twin platform for the straightening machine, dynamic real-time mapping of the stress state of the steel plate straightening production process is achieved.

[0043] Optionally, the straightening machine parameters include: straightening roller diameter, straightening roller length, straightening roller pitch and the number of rollers.

[0044] Steel plate production process parameters include: steel type, width, thickness, yield limit and elastic modulus.

[0045] Optionally, the finite element model building module is further used to:

[0046] The reverse bending deflection during the straightening process is calculated according to the following formula (1), and the accuracy of the finite element model of the roller straightening machine is determined according to the reverse bending deflection:

[0047] Δf=Δl+Δc

[0048]

[0049]

[0050]

[0051] Where Δf is the reverse bending deflection, Δl is the elastic complex deflection, Δc is the residual deflection, M is the elastic-plastic bending moment, L is the straightening roller distance, E is the elastic modulus of the material, I is the moment of inertia of the material cross section about the neutral axis, σ s is the yield stress, b is the width of the steel plate, h is the thickness of the steel plate, ε s is the maximum elastic strain of the material, is the total deformation curvature, E′ is the work hardening strength modulus of the material, Δw is the yield deflection, and C is the ratio of the residual curvature to the yield curvature.

[0052] Optionally, the prediction module is further configured to:

[0053] S31. Use the local outlier factor detection method LOF to detect outliers on the stress parameter data of the steel plate to obtain abnormal points and non-abnormal points.

[0054] Among them, the rule of the local outlier factor detection method LOF is as shown in the following formula (2):

[0055]

[0056] S32, obtaining the stress parameter data of the steel plate at the front and rear positions of the abnormal point, and calculating the mean value according to the following formula (3), and using the mean value to replace the abnormal point.

[0057]

[0058] Where D new The outlier point D i Alternative data, D i-1 is the stress parameter data of the steel plate at the previous position of the abnormal point, D i+1 is the stress parameter data of the steel plate at the position after the abnormal point, and i represents the i-th data point.

[0059] Optionally, the prediction module is further configured to:

[0060] According to the preprocessed stress parameter data of steel plates, the three-swarm particle swarm differential evolution (PSODE) algorithm is adopted to optimize the lightweight gradient boosting machine (LightGBM) and construct an agent model for predicting the stress state of steel plates.

[0061] Optionally, the prediction module is further configured to:

[0062] Two of the three populations are used for differential evolution DE optimization, and the other population except the two populations is used for particle swarm PSO optimization.

[0063] Among them, the mutation mode of differential evolution DE adopts the combination of the current best mode and the best mode.

[0064] The base point vector of the best mode is the best individual of the previous generation population, as shown in the following formula (4):

[0065]

[0066] Where V i is a random new individual, G is the evolutionary generation, r 1 , r 2 , r 3 ∈[1,N],r 1 ≠r 2 ≠r 3≠i, N is the total number of individuals, X best，G is the optimal individual, F is a constant in the range of [0,1], called the variation factor, are base point vectors.

[0067] The base point vector of the current-best mode is the combination of the target individual and the optimal individual, as shown in the following formula (5):

[0068]

[0069] Where, X i，G is the base point vector.

[0070] Optionally, the prediction module is further configured to:

[0071] The three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the learning rate and learning cycle combination of the lightweight gradient boosting machine LightGBM.

[0072] On the other hand, a straightening machine production process visualization analysis device is provided, and the straightening machine production process visualization analysis device comprises: a processor; a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned straightening machine production process visualization analysis methods based on the agent model is implemented.

[0073] On the other hand, a computer-readable storage medium is provided, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned methods for visual analysis of the straightening machine production process based on the agent model.

[0074] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0075] In an embodiment of the present invention, a finite element proxy model is established to quickly predict stress changes in the steel plate straightening process. It is integrated with digital twin technology to achieve real-time simulation and visualization of the straightening production process, thereby improving the interactivity and intelligence level of the entire steel plate straightening system. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0077] Figure 1It is a flow chart of a visual analysis method of a straightening machine production process based on an agent model provided by an embodiment of the present invention;

[0078] Figure 2 It is a diagram of the LightGBM parameter optimization process provided by an embodiment of the present invention;

[0079] Figure 3 It is a block diagram of a visual analysis device for a straightening machine production process based on an agent model provided by an embodiment of the present invention;

[0080] Figure 4 It is a structural schematic diagram of a straightening machine production process visualization analysis device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0082] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0083] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0084] In the embodiments of the present invention, sometimes the subscripts such as W 1 It may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0085] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0086] The embodiment of the present invention provides a straightening machine production process visualization analysis method based on an agent model, which can be implemented by a straightening machine production process visualization analysis device, which can be a terminal or a server. Figure 1 The flowchart of the visualization analysis method of the straightening machine production process based on the agent model is shown. The processing flow of the method may include the following steps:

[0087] S1. Obtain straightening machine parameters, and establish a roller straightening machine finite element model according to the straightening machine parameters.

[0088] In a feasible implementation manner, the key structure and dimensional parameters of the straightening machine are first obtained to establish a finite element model of the roller straightening machine. The key structure and dimensional parameters of the straightening machine may include: straightening roller diameter, straightening roller length, straightening roller pitch and number of rollers, etc.

[0089] Among them, the vertical position of the straightening roller during the straightening process, that is, the determination of the bending amount, determines the accuracy of the finite element simulation model of the straightening process, and the bending amount is determined based on the reverse bending deflection on the plate. The present invention calculates it in the following way:

[0090] Δf=Δl+Δc

[0091]

[0092]

[0093]

[0094] Where Δf is the reverse bending deflection, Δl is the elastic complex deflection, Δc is the residual deflection, M is the elastic-plastic bending moment, L is the straightening roller distance, E is the elastic modulus of the material, I is the moment of inertia of the material cross section about the neutral axis, σ s is the yield stress, b is the width of the steel plate, h is the thickness of the steel plate, ε s is the maximum elastic strain of the material, is the total deformation curvature, E′ is the work hardening strength modulus of the material, Δw is the yield deflection, and C is the ratio of the residual curvature to the yield curvature.

[0095] S2. Obtain steel plate production process parameters, perform finite element simulation based on the steel plate production process parameters and the roller straightening machine finite element model, and obtain steel plate stress parameter data during the straightening process.

[0096] In a feasible implementation, according to the production process parameters of strip steel of different specifications, finite element simulation of the straightening process is performed to obtain the stress parameter information of the steel plate during the straightening process. The production process parameters of the steel plate may include: steel type, width, thickness, yield limit and elastic modulus.

[0097] S3. Preprocess the steel plate stress parameter data, construct a steel plate stress state prediction proxy model according to the preprocessed steel plate stress parameter data, and obtain the proxy model prediction result according to the constructed steel plate stress state prediction proxy model.

[0098] In a feasible implementation manner, the acquired steel plate stress parameter data is preprocessed and input into LightGBM (Light Gradient Boosting Machine) optimized by three-swarm PSODE (Particle Swarm Optimization Differential Evolution) algorithm to establish a steel plate stress state prediction agent model, and the preprocessed steel plate stress parameter data is input into the established steel plate stress state prediction agent model to obtain the agent model prediction result.

[0099] Optionally, the preprocessing of the steel plate stress parameter data in S3 may include the following steps S31-S32:

[0100] S31. First, use LOF (Local Outlier Factor, local outlier factor detection method) to detect outliers in the data.

[0101] In LOF, each data point is assigned an outlier factor LOF that depends on the neighborhood density to determine whether the data point is an outlier. The following rules are formulated:

[0102]

[0103] S32, obtain abnormal point D i The stress parameter data of the steel plate at the front and rear positions are obtained, and the mean value is calculated according to the following formula (3), and the mean value is used to replace the abnormal point.

[0104]

[0105] Where D new The outlier point D i Alternative data, D i-1 is the stress parameter data of the steel plate at the previous position of the abnormal point, D i+1 is the stress parameter data of the steel plate at the position after the abnormal point, and i represents the i-th data point.

[0106] Optionally, constructing a steel plate stress state prediction agent model according to the preprocessed steel plate stress parameter data in S3 includes:

[0107] According to the preprocessed stress parameter data of steel plates, the three-swarm particle swarm differential evolution (PSODE) algorithm is adopted to optimize the lightweight gradient boosting machine (LightGBM) and construct an agent model for predicting the stress state of steel plates.

[0108] Optionally, three swarm particle swarm differential evolution PSODE algorithms include:

[0109] Two of the three populations are used for differential evolution DE optimization, and the other population except the two populations is used for particle swarm PSO optimization.

[0110] Among them, the mutation mode of differential evolution DE adopts the combination of the current best mode and the best mode.

[0111] The base point vector of the best mode is the best individual of the previous generation population, as shown in the following formula (4):

[0112]

[0113] Where V i is a random new individual, G is the evolutionary generation, r 1 , r 2 , r 3 ∈[1,N],r 1 ≠r 2 ≠r 3 ≠i, N is the total number of individuals, X best，G is the optimal individual, F is a constant in the range of [0,1], called the variation factor, ranging from 0 to 7. are base point vectors, is called the deviation disturbance term.

[0114] The base point vector of the current-best mode is the combination of the target individual and the optimal individual, as shown in the following formula (5):

[0115]

[0116] Where, X i，G is the base point vector.

[0117] Optionally, a three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the lightweight gradient boosting machine LightGBM, including:

[0118] The three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the combination of learning rate (LearnRate) and learning cycle (NumLearningCycles) of the lightweight gradient boosting machine LightGBM.

[0119] S4. Develop a digital twin platform for the straightening machine, and realize dynamic real-time mapping of the stress state of the steel plate straightening production process based on the prediction results of the agent model and the digital twin platform of the straightening machine.

[0120] In a feasible implementation, the digital twin system can obtain data information from the basic automation level, process automation level, production execution system and enterprise resource planning system in real time.

[0121] The method of the present invention establishes a finite element model of a roller straightening machine, performs finite element simulation according to the production process parameters of strip steel of different specifications, obtains the stress parameter information of the steel plate in the straightening process, and establishes a steel plate stress state prediction agent model based on the three-population PSODE algorithm to optimize LightGBM. With the help of digital twin technology and agent model integration, a three-dimensional digital twin platform for the straightening machine is developed, which realizes real-time interactive mapping and visualization of the production process, and improves the interactivity and intelligence level of the steel plate straightening production process.

[0122] A 9-roller straightening machine in a steel plant is used as an implementation object of the present invention. This straightening machine uses 4 hydraulic cylinders to control the inclination of the straightening rollers, so as to achieve a state in which the gap between the straightening rollers increases successively.

[0123] Step S1: Obtain the key structure and dimensional parameters of the straightener to establish a finite element model of the roller straightener. The key structure and dimensional parameters of the straightener are shown in Table 1. The finite element modeling of the straightener is completed using ABAQUS software based on the bending amount. The steel plate is divided into C3DBR units, and the division unit is approximately 40*20.6*2.25 (length, width and height), with a total of approximately 70,000 units. The contact between the steel plate and the straightening roller is defined as surface-to-surface contact, and the friction coefficient is 0.3.

[0124] Table 1

[0125] Straightening roller diameter / mm Straightening roller length / mm Straightening roller distance / mm Roller number / piece 360 4400 450 7

[0126] Step S2: According to the production process parameters of strip steel of different specifications, finite element simulation is performed to obtain the stress parameter information of the steel plate in the straightening process and extract data. Taking a certain specification as an example, as shown in Table 2, the finite element simulation obtains the stress parameter information of the steel plate in the straightening process, providing a complete data set for the data preprocessing in step S3.

[0127] Table 2

[0128]

[0129] Step S3: pre-process the acquired steel plate stress parameter data as follows:

[0130] (1) First, the local outlier factor detection method (LOF) is used to detect outliers in the data.

[0131] (2) Replace the outlier point with the mean of its preceding and following position data.

[0132] (3) The three-population PSODE algorithm optimizes LightGBM by optimizing the combination of LightGBM’s learning rate (LearnRate) and learning cycle (NumLearningCycles). The optimization results are as follows: Figure 2As shown, the optimal parameters of LightGBM are: LearnRate=0.071082, NumLearningCyclees=500.

[0133] (4) The input and output parameters of LightGBM in Table 3 are input into LightGBM optimized by the three-species PSODE algorithm to establish a steel plate stress state prediction agent model.

[0134] Table 3

[0135]

[0136] Step S4: Develop a digital twin platform for the straightening machine to achieve dynamic real-time mapping of the stress state of the steel plate straightening production process based on the prediction results of the agent model.

[0137] In an embodiment of the present invention, a finite element proxy model is established to quickly predict stress changes in the steel plate straightening process. It is integrated with digital twin technology to achieve real-time simulation and visualization of the straightening production process, thereby improving the interactivity and intelligence level of the entire steel plate straightening system.

[0138] Figure 3 The block diagram of a visual analysis device for a straightening machine production process based on an agent model according to an exemplary embodiment is shown. The device is used for a visual analysis method for a straightening machine production process based on an agent model. Figure 3 The device includes a finite element model building module 310, a finite element simulation module 320, a prediction module 330, and an output module 630. Among them:

[0139] The finite element model building module 310 is used to obtain the straightening machine parameters and build the roller straightening machine finite element model according to the straightening machine parameters.

[0140] The finite element simulation module 320 is used to obtain the steel plate production process parameters, perform finite element simulation according to the steel plate production process parameters and the roller straightening machine finite element model, and obtain the steel plate stress parameter data during the straightening process.

[0141] The prediction module 330 is used to preprocess the steel plate stress parameter data, construct a steel plate stress state prediction proxy model based on the preprocessed steel plate stress parameter data, and obtain the proxy model prediction result based on the constructed steel plate stress state prediction proxy model.

[0142] The output module 340 is used to develop a digital twin platform for the straightening machine, and realize dynamic real-time mapping of the stress state of the steel plate straightening production process according to the prediction results of the agent model and the digital twin platform of the straightening machine.

[0143] Optionally, the straightening machine parameters include: straightening roller diameter, straightening roller length, straightening roller pitch and the number of rollers.

[0144] Steel plate production process parameters include: steel type, width, thickness, yield limit and elastic modulus.

[0145] Optionally, the finite element model building module 310 is further used to:

[0146] The reverse bending deflection during the straightening process is calculated according to the following formula (1), and the accuracy of the finite element model of the roller straightening machine is determined according to the reverse bending deflection:

[0147] Δf=Δl+Δc

[0148]

[0149]

[0150]

[0151] Where Δf is the reverse bending deflection, Δl is the elastic complex deflection, Δc is the residual deflection, M is the elastic-plastic bending moment, L is the straightening roller distance, E is the elastic modulus of the material, I is the moment of inertia of the material cross section about the neutral axis, σ s is the yield stress, b is the width of the steel plate, h is the thickness of the steel plate, ε s is the maximum elastic strain of the material, is the total deformation curvature, E′ is the work hardening strength modulus of the material, Δw is the yield deflection, and C is the ratio of the residual curvature to the yield curvature.

[0152] Optionally, the prediction module 330 is further configured to:

[0153] S31. Use the local outlier factor detection method LOF to detect outliers on the stress parameter data of the steel plate to obtain abnormal points and non-abnormal points.

[0154] Among them, the rule of the local outlier factor detection method LOF is as shown in the following formula (2):

[0155]

[0156] S32, obtain abnormal point D i The stress parameter data of the steel plate at the front and rear positions are obtained, and the mean value is calculated according to the following formula (3), and the mean value is used to replace the abnormal point.

[0157]

[0158] Where D new The outlier point D i Alternative data, D i-1is the stress parameter data of the steel plate at the previous position of the abnormal point, D i+1 is the stress parameter data of the steel plate at the position after the abnormal point, and i represents the i-th data point.

[0159] Optionally, the prediction module 330 is further configured to:

[0160] According to the preprocessed stress parameter data of steel plates, the three-swarm particle swarm differential evolution (PSODE) algorithm is adopted to optimize the lightweight gradient boosting machine (LightGBM) and construct an agent model for predicting the stress state of steel plates.

[0161] Optionally, the prediction module 330 is further configured to:

[0162] Two of the three populations are used for differential evolution DE optimization, and the other population except the two populations is used for particle swarm PSO optimization.

[0163] Among them, the mutation mode of differential evolution DE adopts the combination of the current best mode and the best mode.

[0164] The base point vector of the best mode is the best individual of the previous generation population, as shown in the following formula (4):

[0165]

[0166] Where V i is a random new individual, G is the evolutionary generation, r 1 , r 2 , r 3 ∈[1,N],r 1 ≠r 2 ≠r 3 ≠i, N is the total number of individuals, X best，G is the optimal individual under the best mode, F is a constant in the range of [0,1], called the variation factor, are all base point vectors;

[0167] The base point vector of the current-best mode is the combination of the target individual and the optimal individual, as shown in the following formula (5):

[0168]

[0169] Where, X i，G is the base point vector.

[0170] Optionally, the prediction module 330 is further configured to:

[0171] The three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the learning rate and learning cycle combination of the lightweight gradient boosting machine LightGBM.

[0172] In an embodiment of the present invention, a finite element proxy model is established to quickly predict stress changes in the steel plate straightening process. It is integrated with digital twin technology to achieve real-time simulation and visualization of the straightening production process, thereby improving the interactivity and intelligence level of the entire steel plate straightening system.

[0173] Figure 4 Schematic diagram of a straightening machine production process visualization analysis device provided by an embodiment of the present invention. Figure 4 As shown, the straightening machine production process visualization analysis device may include the above Figure 3 The visual analysis device for the production process of a straightening machine based on an agent model is shown. Optionally, the visual analysis device 410 for the production process of a straightening machine may include a first processor 2001 .

[0174] Optionally, the straightening machine production process visualization analysis device 410 may further include a memory 2002 and a transceiver 2003 .

[0175] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0176] Combine the following Figure 4 The various components of the straightening machine production process visualization analysis device 410 are specifically introduced:

[0177] The first processor 2001 is the control center of the straightening machine production process visualization analysis device 410, which can be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention, such as one or more microprocessors (digital signal processor, DSP), or one or more field programmable gate arrays (field programmable gate array, FPGA).

[0178] Optionally, the first processor 2001 can execute various functions of the straightening machine production process visualization analysis device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0179] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.

[0180] In a specific implementation, as an embodiment, the straightening machine production process visualization analysis device 410 may also include multiple processors, such as Figure 4 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0181] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0182] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently, and may be accessed through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0183] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0184] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0185] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0186] It should be noted that Figure 4 The structure of the straightening machine production process visualization analysis device 410 shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0187] In addition, the technical effects of the straightening machine production process visualization analysis device 410 can refer to the technical effects of the straightening machine production process visualization analysis method based on the agent model described in the above method embodiment, and will not be repeated here.

[0188] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0189] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0190] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0191] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0192] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0193] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0194] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0195] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0196] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0197] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0198] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0199] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0200] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A visual analysis method for the production process of a straightening machine based on an agent model, characterized in that: The method comprises: S1, obtaining straightening machine parameters, and establishing a roller straightening machine finite element model according to the straightening machine parameters; S2, obtaining steel plate production process parameters, performing finite element simulation according to the steel plate production process parameters and a roller straightening machine finite element model, and obtaining steel plate stress parameter data during the straightening process; S3, preprocessing the steel plate stress parameter data, constructing a steel plate stress state prediction proxy model according to the preprocessed steel plate stress parameter data, and obtaining a proxy model prediction result according to the constructed steel plate stress state prediction proxy model; S4. Develop a digital twin platform for a straightening machine, and realize dynamic real-time mapping of the stress state of the steel plate straightening production process according to the prediction results of the agent model and the digital twin platform for the straightening machine; The process of establishing the finite element model of the roller straightening machine in S1 includes: The reverse bending deflection during the straightening process is calculated according to the following formula (1), and the accuracy of the finite element model of the roller straightening machine is determined according to the reverse bending deflection: Δf=Δl+Δc Where Δf is the reverse bending deflection, Δl is the elastic complex deflection, Δc is the residual deflection, M is the elastic-plastic bending moment, L is the straightening roller distance, E is the elastic modulus of the material, I is the moment of inertia of the material cross section about the neutral axis, σ s is the yield stress, b is the width of the steel plate, h is the thickness of the steel plate, ε s is the maximum elastic strain of the material, is the total deformation curvature, E′ is the work hardening strength modulus of the material, ΔW is the yield deflection, and C is the ratio of residual curvature to yield curvature; The step S3 of constructing a steel plate stress state prediction agent model according to the preprocessed steel plate stress parameter data includes: According to the preprocessed steel plate stress parameter data, the three-swarm particle swarm differential evolution PSODE algorithm is used to optimize the lightweight gradient boosting machine LightGBM to build a steel plate stress state prediction agent model; The three-swarm particle swarm differential evolution PSODE algorithm includes: Two of the three populations are used for differential evolution DE optimization, and another population other than the two populations is used for particle swarm PSO optimization; Among them, the mutation mode of differential evolution DE adopts the combination of the current best mode and the best mode; The base point vector of the best mode is the best individual of the previous generation population, as shown in the following formula (2): Where V i is a random new individual, G is the evolutionary generation, r1, r2, r3∈[1,N], r1≠r2≠r3≠i, N is the total number of individuals, X best，G is the optimal individual, F is a constant in the range of [0,1], called the variation factor, are all base point vectors; The base point vector of the current-best mode is the combination of the target individual and the optimal individual, as shown in the following formula (3): Where, X i，G is the base point vector; The three-swarm particle swarm differential evolution PSODE algorithm is used to optimize the lightweight gradient boosting machine LightGBM, including: The three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the learning rate and learning cycle combination of the lightweight gradient boosting machine LightGBM.

2. The visual analysis method for the production process of a straightening machine based on an agent model according to claim 1 is characterized in that: The straightening machine parameters in S1 include: straightening roller diameter, straightening roller length, straightening roller pitch and number of rollers; The steel plate production process parameters in S2 include: steel type, width, thickness, yield limit and elastic modulus.

3. The visual analysis method for the production process of a straightening machine based on an agent model according to claim 1 is characterized in that: The preprocessing of the steel plate stress parameter data in S3 includes: S31, using a local outlier factor detection method LOF to perform outlier detection on the stress parameter data of the steel plate to obtain abnormal points and non-abnormal points; Among them, the rule of the local outlier factor detection method LOF is as shown in the following formula (4): S32, obtaining the stress parameter data of the steel plate at the front and rear positions of the abnormal point, and calculating the mean value according to the following formula (5), and using the mean value to replace the abnormal point; Where D new The outlier point D i Alternative data, D i-1 is the stress parameter data of the steel plate at the previous position of the abnormal point, D i+1 is the stress parameter data of the steel plate at the position after the abnormal point, and i represents the i-th data point.

4. A visual analysis device for a straightening machine production process based on an agent model, wherein the visual analysis device for a straightening machine production process based on an agent model is used to implement a visual analysis method for a straightening machine production process based on an agent model as claimed in any one of claims 1 to 3, characterized in that: The device comprises: A finite element model building module, used for acquiring straightening machine parameters and building a roller straightening machine finite element model according to the straightening machine parameters; A finite element simulation module is used to obtain steel plate production process parameters, perform finite element simulation according to the steel plate production process parameters and a roller straightening machine finite element model, and obtain steel plate stress parameter data during the straightening process; A prediction module is used to preprocess the steel plate stress parameter data, construct a steel plate stress state prediction proxy model according to the preprocessed steel plate stress parameter data, and obtain a proxy model prediction result according to the constructed steel plate stress state prediction proxy model; An output module is used to develop a digital twin platform of a straightening machine, and realize dynamic real-time mapping of the stress state of the steel plate straightening production process according to the prediction results of the proxy model and the digital twin platform of the straightening machine; The process of establishing the finite element model of the roller straightening machine includes: The reverse bending deflection during the straightening process is calculated according to the following formula (1), and the accuracy of the finite element model of the roller straightening machine is determined according to the reverse bending deflection: Δf=Δl+Δc Where Δf is the reverse bending deflection, Δl is the elastic complex deflection, Δc is the residual deflection, M is the elastic-plastic bending moment, L is the straightening roller distance, E is the elastic modulus of the material, I is the moment of inertia of the material cross section about the neutral axis, σ s is the yield stress, b is the width of the steel plate, h is the thickness of the steel plate, εs is the maximum elastic strain of the material, is the total deformation curvature, E′ is the work hardening strength modulus of the material, Δw is the yield deflection, and C is the ratio of residual curvature to yield curvature; The method of constructing a steel plate stress state prediction agent model according to the preprocessed steel plate stress parameter data comprises: According to the preprocessed steel plate stress parameter data, the three-swarm particle swarm differential evolution PSODE algorithm is used to optimize the lightweight gradient boosting machine LightGBM to build a steel plate stress state prediction agent model; The three-swarm particle swarm differential evolution PSODE algorithm includes: Two of the three populations are used for differential evolution DE optimization, and another population other than the two populations is used for particle swarm PSO optimization; Among them, the mutation mode of differential evolution DE adopts the combination of the current best mode and the best mode; The base point vector of the best mode is the best individual of the previous generation population, as shown in the following formula (2): Where V i is a random new individual, G is the evolutionary generation, r1,r2,r3∈[1,N],r1≠r2≠r3≠i, N is the total number of individuals, X best，G is the optimal individual, F is a constant in the range of [0,1], called the variation factor, are all base point vectors; The base point vector of the current-best mode is the combination of the target individual and the optimal individual, as shown in the following formula (3): Where, X i，G is the base point vector; The three-swarm particle swarm differential evolution PSODE algorithm is used to optimize the lightweight gradient boosting machine LightGBM, including: The three-swarm particle swarm differential evolution (PSODE) algorithm is used to optimize the learning rate and learning cycle combination of the lightweight gradient boosting machine LightGBM.

5. A visual analysis device for the production process of a straightening machine, characterized in that: The straightening machine production process visualization analysis equipment includes: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 3 is implemented.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 3.

Images