An aluminum profile extrusion die chamfer optimization method, device, terminal equipment and storage medium

Abstract

The application discloses an aluminum profile extrusion die chamfer optimization method and device, terminal equipment and storage medium, the method is to obtain target aluminum profile material parameters and target extrusion process parameters; the target aluminum profile material parameters and target extrusion process parameters are input into the chamfer parameter prediction model, and the final chamfer parameter combination is generated, and the chamfer of the aluminum profile extrusion die to be optimized is optimized according to the final chamfer parameter combination. The chamfer parameter prediction model is provided with an aluminum profile characteristic sensing layer, which senses the thermoplastic state and oxidation state of the aluminum profile, and is provided with a physical geometric constraint branch layer, which ensures whether the generated chamfer parameter combination meets the physical constraint; by implementing the application, the efficiency of the aluminum profile extrusion die chamfer parameter design can be improved, and the surface defect rate of the finished product is reduced.

Description

Technical Field

[0001] This invention relates to the field of aluminum profile die optimization, and in particular to a method, apparatus, terminal equipment and storage medium for optimizing the chamfer of aluminum profile extrusion dies. Background Technology

[0002] In the field of aluminum profile processing, the chamfer of aluminum profile extrusion die refers to the non-right-angle transition structure formed by machining at the entrance of the die cavity, the transition zone of the working zone, or the die cutting edge. Its core parameters are angle, radius, and transition length. It is a flow guide and stress buffer structure for aluminum profile extrusion. Simply put, it is like a guide ramp for extrusion molding. When the billet (high-temperature aluminum rod) enters the constrained cavity of the die from a free state, the chamfer transition avoids flow obstruction, stress concentration, or surface scratches caused by sharp edges. Appropriate chamfer parameters can reduce the surface defect rate of the finished product.

[0003] The current design of chamfering parameters for aluminum profiles is generally based on the experience of engineers, which is too subjective. When faced with unfamiliar working conditions, engineers can only design a rough range of chamfering parameters and then conduct tests one by one to determine the optimal chamfering parameters. This process is cumbersome and has poor accuracy. Summary of the Invention

[0004] This invention provides a method, apparatus, terminal equipment, and storage medium for optimizing the chamfering of aluminum profile extrusion dies, which can improve the efficiency of chamfering parameter design for aluminum profile extrusion dies and reduce the surface defect rate of finished products.

[0005] The present invention provides a method for optimizing the chamfer of an aluminum profile extrusion die, comprising: obtaining the target aluminum profile material parameters and the target extrusion process parameters; The target aluminum profile material parameters and target extrusion process parameters are input into the chamfer parameter prediction model; wherein, the chamfer parameter prediction model includes: an aluminum profile characteristic perception layer, a feature extraction layer, an initial chamfer parameter generation layer, a feature parameter mapping layer, a physical geometric constraint branch layer, and a fusion layer; Based on the aluminum profile characteristic sensing layer, the thermoplastic state and oxidation state of the aluminum profile are sensed according to the target aluminum profile material parameters and target extrusion process parameters, and thermoplastic feature vector and oxidation feature vector are generated. Based on the feature extraction layer, the thermoplastic feature vector, oxidation risk feature vector, and working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters are fused to generate fused features; The fused features are transmitted to the initial chamfer parameter generation layer to generate an initial chamfer parameter combination; Based on the feature parameter mapping layer, according to the preset aluminum profile characteristic association weight matrix, the thermoplastic feature vector, oxidation risk feature vector and preliminary chamfer parameter combination are mapped and calculated to fine-tune the preliminary parameters for characteristic adaptability and generate the fine-tuned chamfer parameter combination. The fine-tuned chamfer parameter combination is input into the physical geometry constraint branch layer so that the physical geometry constraint branch layer can determine whether the fine-tuned chamfer parameter combination meets the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction. Based on the fusion layer, the fine-tuned chamfer parameter combination is adjusted a second time according to the physical constraint satisfaction to generate the final chamfer parameter combination; Based on the final chamfer parameter combination, the chamfer of the aluminum profile extrusion die to be optimized is then performed. Furthermore, the chamfer parameter prediction model is trained in the following manner: Obtain several samples consisting of different aluminum profile material parameters and extrusion process parameters; wherein each sample includes aluminum profile material parameters and extrusion process parameters; For each sample, based on the aluminum profile extrusion dies corresponding to different chamfer parameter combinations, and the finished product surface defect rate under the aluminum profile material parameters and extrusion process parameters corresponding to the sample, the chamfer parameter combination with the smallest finished product surface defect rate and less than the preset defect rate is selected as the target chamfer parameter combination for the corresponding sample. Using each sample and its corresponding target chamfer parameter combination as input, and the predicted chamfer parameter combination corresponding to each sample as output, the preset neural network model is iteratively trained until the loss function converges, thereby generating the chamfer parameter prediction model. During each training iteration, the loss function value is calculated based on the deviation between the target chamfer parameter combination and the predicted chamfer parameter combination, as well as the physical constraint satisfaction of the predicted chamfer parameter combination. If the loss function value does not converge, the parameters of the neural network model are adjusted.

[0006] Furthermore, the surface defect rate of the finished product for each aluminum profile extrusion die corresponding to the different chamfering parameter combinations, under the aluminum profile material parameters and extrusion process parameters corresponding to the sample, includes: Simulation models of extrusion dies for various aluminum profiles were constructed; the chamfer parameter combinations for different simulation models were different. For each aluminum profile extrusion die simulation model, multiple sets of die flow simulations are performed based on different aluminum profile material parameters and extrusion process parameters. From each set of die flow simulation results, the maximum thickness of the oxide layer in the chamfer area, the maximum stress in the die cavity, the standard deviation of the velocity at the profile exit section, and the decrease in peak force during the extrusion process are extracted. The maximum thickness of the oxide layer in the chamfered area, the maximum stress in the die cavity, the standard deviation of the velocity at the profile exit section, and the decrease in peak force during extrusion are input into the preset defect rate fitting model to obtain the finished surface defect rate of the aluminum profile extrusion die under the corresponding combination of chamfer parameters and aluminum profile material parameters and extrusion process parameters.

[0007] Furthermore, the aluminum profile characteristic sensing layer senses the thermoplastic and oxidation states of the aluminum profile based on the target aluminum profile material parameters and target extrusion process parameters, generating thermoplastic feature vectors and oxidation feature vectors, including: The thermoplastic strength coefficient is calculated based on the extrusion temperature, yield strength, tensile strength, and melting point temperature of the target aluminum profile; the flow resistance coefficient is calculated based on the friction coefficient, extrusion pressure, extrusion speed, and cross-sectional area of ​​the profile; the softening risk coefficient is determined based on the extrusion temperature, melting point temperature, and recrystallization temperature of the target aluminum profile; and a thermoplastic characteristic vector is constructed based on the thermoplastic strength coefficient, flow resistance coefficient, and softening risk coefficient. The oxidation rate coefficient is calculated based on the oxidation reaction activation energy, gas constant, extrusion temperature, and frequency factor; the oxide layer thickness is generated based on the oxidation rate coefficient; and an oxidation feature vector is constructed based on the oxidation rate coefficient and the oxide layer thickness.

[0008] Furthermore, based on the feature extraction layer, the thermoplastic feature vector, oxidation risk feature vector, and the working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters are fused to generate fused features, including: The thermoplastic strength coefficient is correlated with the extrusion temperature and recrystallization temperature to construct a first correlation feature between temperature and thermoplastic strength; the flow resistance coefficient is correlated with the extrusion ratio and friction coefficient to construct a second correlation feature between process and flow resistance; the softening risk coefficient is correlated with the extrusion speed and extrusion temperature to construct a third correlation feature between softening risk coefficients. The oxidation rate coefficient is correlated with extrusion temperature and time to construct a fourth correlation feature to characterize the effect of extrusion temperature and time on the oxidation rate; the oxidation rate coefficient is correlated with extrusion pressure and extrusion speed to construct a fifth correlation feature to characterize the regulatory effect of the process on the oxidation rate; the oxide layer thickness is correlated with extrusion time and extrusion temperature to construct a sixth correlation feature to quantify the oxide thickness corresponding to a unit extrusion temperature. Based on the first association feature, the second association feature, the third association feature, the fourth association feature, the fifth association feature, and the sixth association feature, an association feature vector is generated; The thermoplastic feature vector, oxidation feature vector, working condition feature vector, and associated feature vector are concatenated to obtain an initial high-dimensional feature matrix; The features in the initial high-dimensional feature matrix are weighted and fused according to the attention weights of each feature to generate the fused feature vector.

[0009] Furthermore, the feature parameter mapping layer, based on a preset aluminum profile characteristic association weight matrix, performs mapping calculations on the thermoplastic feature vector, oxidation risk feature vector, and preliminary chamfer parameter combination to fine-tune the preliminary parameters for characteristic adaptation, generating a fine-tuned chamfer parameter combination, including: Based on the aluminum profile characteristic correlation weight matrix, thermoplasticity eigenvector, and oxidation risk eigenvector, calculate the characteristic influence coefficient of the chamfer parameter combination; The parameter adjustment amount is calculated based on the characteristic influence coefficient and the initial chamfer parameter combination; The initial chamfer parameter combination is fine-tuned according to the parameter adjustment amount to generate the fine-tuned chamfer parameter combination.

[0010] Further, the step of inputting the fine-tuned chamfer parameter combination into the physical geometry constraint branch layer allows the physical geometry constraint branch layer to determine whether the fine-tuned chamfer parameter combination satisfies the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction, including: Determine whether each chamfer parameter in the fine-tuned chamfer parameter combination is within the corresponding physical geometric constraint range, and determine the number of parameters that are not within the corresponding physical geometric constraint range. Based on the number of parameters that are not within the range of the corresponding physical geometric constraints, the standard physical constraint satisfaction is linearly reduced to generate the corresponding physical constraint satisfaction.

[0011] Another embodiment of the present invention provides a chamfer optimization device for aluminum profile extrusion dies, comprising: a parameter acquisition module, a model parameter input module, a characteristic sensing module, a feature extraction module, a preliminary chamfer generation module, a chamfer parameter primary adjustment module, a physical constraint module, a chamfer secondary adjustment module, and an optimization module; The parameter acquisition module is used to acquire the target aluminum profile material parameters and the target extrusion process parameters; The model parameter input module is used to input the target aluminum profile material parameters and target extrusion process parameters into the chamfer parameter prediction model; wherein, the chamfer parameter prediction model includes: an aluminum profile characteristic perception layer, a feature extraction layer, an initial chamfer parameter generation layer, a feature parameter mapping layer, a physical geometric constraint branch layer, and a fusion layer; The characteristic sensing module is used to sense the thermoplastic state and oxidation state of the aluminum profile based on the aluminum profile characteristic sensing layer, according to the target aluminum profile material parameters and the target extrusion process parameters, and generate thermoplastic feature vectors and oxidation feature vectors. The feature extraction module is used to fuse the thermoplastic feature vector, the oxidation risk feature vector, and the working condition features corresponding to the target aluminum profile material parameters and the target extrusion process parameters based on the feature extraction layer to generate fused features; The preliminary chamfer generation module is used to transmit the fused features to the initial chamfer parameter generation layer to generate a preliminary chamfer parameter combination; The chamfer parameter adjustment module is used to perform mapping calculations on the thermoplastic feature vector, oxidation risk feature vector and preliminary chamfer parameter combination based on the feature parameter mapping layer and according to the preset aluminum profile characteristic association weight matrix, so as to fine-tune the preliminary parameters for characteristic adaptability and generate the fine-tuned chamfer parameter combination. The physical constraint module is used to input the fine-tuned chamfer parameter combination into the physical geometric constraint branch layer, so that the physical geometric constraint branch layer can determine whether the fine-tuned chamfer parameter combination meets the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction. The chamfer secondary adjustment module is used to perform secondary adjustments on the fine-tuned chamfer parameter combination based on the fusion layer and the physical constraint satisfaction, to generate the final chamfer parameter combination. The optimization module is used to optimize the chamfer of the aluminum profile extrusion die to be optimized based on the final chamfer parameter combination.

[0012] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the aluminum profile extrusion die chamfering optimization method provided by the present invention.

[0013] Another embodiment of the present invention also provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to execute the steps of the aluminum profile extrusion die chamfering optimization method provided by the present invention.

[0014] The following benefits can be obtained by implementing the present invention: This application provides a method, apparatus, terminal device, and storage medium for optimizing the chamfering of aluminum profile extrusion dies. The method is based on a chamfering parameter prediction model constructed from a neural network model, directly predicting the optimal chamfering parameters under the target working conditions, i.e., the target aluminum profile material parameters and the target extrusion process parameters. Compared to manual methods based on experience, this reduces over-reliance on personal experience and improves the accuracy and efficiency of chamfering design. Furthermore, the neural network model innovatively incorporates an aluminum profile characteristic sensing layer, combining the plastic state and easy oxidation characteristics of the aluminum profile. This layer actively senses and integrates the thermoplastic and oxidation states of the aluminum profile, guiding the setting of chamfering parameters based on these states. On the one hand, by analyzing the thermoplastic state of the material, the flowability of the metal can be predicted. The chamfering parameters generated based on this ensure smooth metal flow, reducing surface scratches and abrasions caused by poor flow or localized stress concentration, and reducing the degree of oxidation to avoid surface defects caused by the shedding of thick oxide layers, thereby reducing the surface defect rate. In addition, the introduction of a feature parameter mapping layer and a physical geometric constraint branch layer enables fine-tuning of parameter characteristics and verification of physical constraints. This allows for more precise parameter adjustments and ensures that the adjusted parameters conform to physical laws and geometric design requirements, thus avoiding unreasonable designs. Attached Figure Description

[0015] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating a method for optimizing the chamfering of aluminum profile extrusion dies according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of an aluminum profile extrusion die chamfering optimization device provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0019] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0022] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0023] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0024] See Figure 1 An embodiment of the present invention provides a method for optimizing the chamfer of an aluminum profile extrusion die, comprising: 101. Obtain the target aluminum profile material parameters and the target extrusion process parameters.

[0025] Preferably, in this invention, the working parameters of the aluminum profile extrusion process are composed of aluminum profile material parameters and extrusion process parameters.

[0026] Indicatively, aluminum profile material parameters include, but are not limited to, yield strength, tensile strength, melting point, and recrystallization temperature, which determine the inherent properties of the material. Extrusion process parameters, including but not limited to, extrusion temperature, extrusion speed, extrusion pressure, coefficient of friction, extrusion ratio, and extrusion duration, are the extrusion process conditions.

[0027] When predicting the chamfer of aluminum profile extrusion dies, the material parameters of the aluminum profile corresponding to the current model can be obtained to obtain the target aluminum profile material parameters; the relevant extrusion process parameters can be obtained to obtain the corresponding target extrusion process parameters.

[0028] 102. Input the target aluminum profile material parameters and target extrusion process parameters into the chamfer parameter prediction model; wherein, the chamfer parameter prediction model includes: an aluminum profile characteristic perception layer, a feature extraction layer, an initial chamfer parameter generation layer, a feature parameter mapping layer, a physical geometric constraint branch layer, and a fusion layer.

[0029] In a preferred embodiment, the training method of the chamfer parameter prediction model includes: obtaining a number of samples composed of different aluminum profile material parameters and extrusion process parameters; wherein, each sample includes an aluminum profile material parameter and an extrusion process parameter; For each sample, based on the aluminum profile extrusion dies corresponding to different chamfer parameter combinations, and the finished product surface defect rate under the aluminum profile material parameters and extrusion process parameters corresponding to the sample, the chamfer parameter combination with the smallest finished product surface defect rate and less than the preset defect rate is selected as the target chamfer parameter combination for the corresponding sample. Using each sample and its corresponding target chamfer parameter combination as input, and the predicted chamfer parameter combination corresponding to each sample as output, the preset neural network model is iteratively trained until the loss function converges, thereby generating the chamfer parameter prediction model. During each training iteration, the loss function value is calculated based on the deviation between the target chamfer parameter combination and the predicted chamfer parameter combination, as well as the physical constraint satisfaction of the predicted chamfer parameter combination. If the loss function value does not converge, the parameters of the neural network model are adjusted.

[0030] Preferably, the above loss function can be specifically as follows: ; The regression loss, used to measure the difference between the predicted and target chamfer parameters, can be constructed using mean squared error. An illustrative example is: ; Where: N is the number of samples, and M is the dimension of the chamfer parameters (such as chamfer radius, angle, and transition length). Schematically, in this invention, the chamfer parameter combination is a combination of chamfer radius, angle, and transition length. Let j be the predicted chamfer parameter for the i-th sample; Let be the chamfer parameter of the j-th target for the i-th sample.

[0031] To constrain the loss, a penalty term based on the physical constraint satisfaction can be used to penalize predicted chamfer parameters that do not meet the physical geometric constraints: ; in, Let represent the physical constraint satisfaction of the i-th sample; The constraint satisfaction threshold can be 1 for example.

[0032] In an optional embodiment, aluminum profile extrusion dies with various combinations of chamfering parameters can be prepared in advance. Except for the chamfering parameters, all other die parameters are identical. Then, under different aluminum profile material parameters and extrusion process parameters, each aluminum profile extrusion die is used for actual extrusion to generate actual finished products. The surface defect rate of the finished products is then statistically analyzed. This determines the surface defect rate of the finished product under a set of chamfering parameter combinations, aluminum profile material parameters, and extrusion process parameters. In other words, the surface defect rate of the finished product for each aluminum profile extrusion die corresponding to the different chamfering parameter combinations under the aluminum profile material parameters and extrusion process parameters corresponding to the sample.

[0033] Because the preparation of physical molds for various chamfer parameter combinations is too time-consuming and costly, in other alternative embodiments, based on the mold flow simulation of aluminum profile extrusion dies, simulation models of aluminum profile extrusion dies under different chamfer parameter combinations are first generated, and then simulation is performed. The corresponding simulation indicators are extracted and combined with standards. A multiple linear regression model of the relationship between simulation indicators and defect rate is used to determine the final surface defect rate of the finished product. The specific steps are as follows: The surface defect rate of the finished product under the aluminum profile material parameters and extrusion process parameters corresponding to the aluminum profile extrusion dies according to different chamfer parameter combinations includes: Simulation models of extrusion dies for various aluminum profiles were constructed; the chamfer parameter combinations for different simulation models were different. For each aluminum profile extrusion die simulation model, multiple sets of die flow simulations are performed based on different aluminum profile material parameters and extrusion process parameters. From each set of die flow simulation results, the maximum thickness of the oxide layer in the chamfer area, the maximum stress in the die cavity, the standard deviation of the velocity at the profile exit section, and the decrease in peak force during the extrusion process are extracted. The maximum thickness of the oxide layer in the chamfered area, the maximum stress in the die cavity, the standard deviation of the velocity at the profile exit section, and the decrease in peak force during extrusion are input into the preset defect rate fitting model to obtain the finished surface defect rate of the aluminum profile extrusion die under the corresponding combination of chamfer parameters and aluminum profile material parameters and extrusion process parameters.

[0034] In this embodiment, firstly, based on the geometric parameters of the die, a simulation model of an aluminum profile extrusion die with different chamfer parameter combinations is constructed using existing simulation tools. It should be noted that the simulation models of different aluminum profile extrusion dies have the same geometric parameters except for the chamfer parameter combinations.

[0035] After constructing simulation models of different aluminum profile extrusion dies, multiple die flow simulations were performed for each aluminum profile extrusion die under different aluminum profile material parameters and extrusion process parameters. In each simulation, the simulation parameters remained consistent except for the aluminum profile material parameters and extrusion process parameters.

[0036] After each simulation, the following four indicators are extracted from the simulation results: maximum oxide layer thickness in the chamfered area, maximum stress in the die cavity, standard deviation of the profile exit section velocity, and the decrease in peak force after extrusion. These four indicators are the core for judging the defect rate. For the maximum oxide layer thickness in the chamfered area, the thicker the oxide layer, the easier it is to peel off, leading to scratches and pitting on the profile surface. For the maximum stress in the die cavity, the greater the die stress, the easier it is to deform, resulting in an uneven profile surface. For the standard deviation of the profile exit section velocity, the larger the value, the more uneven the flow velocity of the profile when it exits the die, which is prone to twisting and uneven wall thickness. For the decrease in peak force after extrusion, a decrease that is too fast or too slow indicates unstable extrusion stress, which is prone to defects such as bubbles and cracks.

[0037] After obtaining the above simulation indicators, they are input into the pre-constructed multiple linear regression model (i.e., the above defect rate fitting model) to output the corresponding defect rate.

[0038] It should be noted that the multiple linear regression model establishes the relationship between the maximum thickness of the oxide layer in the chamfered region, the maximum stress in the die cavity, the standard deviation of the profile exit section velocity, and the peak force during extrusion, and the defect rate, without involving chamfer parameters. Therefore, based on a small number of physical aluminum profile extrusion dies, by continuously changing the working conditions, actual aluminum profile extrusion can be performed to determine the maximum thickness of the oxide layer in the chamfered region, the maximum stress in the die cavity, the standard deviation of the profile exit section velocity, the peak force during extrusion, and the actual defect rate under each working condition. Then, multiple linear fitting can be performed to obtain the aforementioned defect rate fitting model. This eliminates the need to prepare a large number of physical aluminum profile extrusion dies; even with only one die, by continuously changing the working parameters, multiple sets of correlation data on the maximum thickness of the oxide layer in the chamfered region, the maximum stress in the die cavity, the standard deviation of the profile exit section velocity, the peak force during extrusion, and the actual defect rate can be obtained, thus constructing a multiple linear regression model. This significantly reduces the cost of die preparation and shortens the cycle time.

[0039] 103. Based on the aluminum profile characteristic sensing layer, the thermoplastic state and oxidation state of the aluminum profile are sensed according to the target aluminum profile material parameters and target extrusion process parameters, and thermoplastic feature vector and oxidation feature vector are generated.

[0040] In a preferred embodiment, the aluminum profile characteristic sensing layer senses the thermoplastic and oxidation states of the aluminum profile based on the target aluminum profile material parameters and target extrusion process parameters, and generates thermoplastic feature vectors and oxidation feature vectors, including: The thermoplastic strength coefficient is calculated based on the extrusion temperature, yield strength, tensile strength, and melting point temperature of the target aluminum profile; the flow resistance coefficient is calculated based on the friction coefficient, extrusion pressure, extrusion speed, and cross-sectional area of ​​the profile; the softening risk coefficient is determined based on the extrusion temperature, melting point temperature, and recrystallization temperature of the target aluminum profile; and a thermoplastic characteristic vector is constructed based on the thermoplastic strength coefficient, flow resistance coefficient, and softening risk coefficient. The oxidation rate coefficient is calculated based on the oxidation reaction activation energy, gas constant, extrusion temperature, and frequency factor; the oxide layer thickness is generated based on the oxidation rate coefficient; and an oxidation feature vector is constructed based on the oxidation rate coefficient and the oxide layer thickness.

[0041] Specifically, the material parameters of the target aluminum profile and the target extrusion process parameters are first normalized. Next, the thermoplastic strength coefficient is calculated using the following formula: ; in, This is the thermoplastic strength coefficient, which reflects the material's ability to undergo plastic deformation at high temperatures; Yield strength; Tensile strength; This refers to the extrusion temperature. This is the melting point temperature.

[0042] The flow resistance coefficient is calculated using the following formula: ; in, is the flow resistance coefficient, which reflects the flow resistance of the material in the mold cavity; μ is the friction coefficient; P is the extrusion pressure; v is the extrusion speed; A is the cross-sectional area of ​​the profile; The softening risk coefficient is calculated using the following formula: ;in, The softening risk factor assesses the risk of excessive softening of a material at high temperatures. This is the recrystallization temperature; The corresponding thermoplastic feature vector is finally generated based on the thermoplastic strength coefficient, flow resistance coefficient, and softening risk coefficient. For the oxidation eigenvector, the oxidation rate coefficient is first calculated using the following formula: ; in, R is the oxidation rate coefficient; Q is the activation energy of the oxidation reaction; R is the gas constant. The oxide layer thickness is then generated using the following formula: ; in, t represents the oxide layer thickness; t represents the extrusion time. Finally, based on the oxide layer thickness and oxidation rate coefficient, the corresponding oxidation feature vector is generated; In this embodiment, the oxidation and thermoplasticity characteristics of the aluminum profile during the extrusion process are perceived through oxidation feature vectors and thermoplasticity feature vectors. Based on these two characteristics, the setting of subsequent chamfering parameter combinations can be guided, making the generated chamfering parameter combinations more consistent with the characteristics of aluminum alloys and improving the accuracy of chamfering combination design.

[0043] 104. Based on the feature extraction layer, the thermoplastic feature vector, oxidation risk feature vector, and the working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters are fused to generate fused features.

[0044] In a preferred embodiment, the step of fusing the thermoplastic feature vector, oxidation risk feature vector, and the working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters based on the feature extraction layer to generate fused features includes: The thermoplastic strength coefficient is correlated with the extrusion temperature and recrystallization temperature to construct a first correlation feature between temperature and thermoplastic strength; the flow resistance coefficient is correlated with the extrusion ratio and friction coefficient to construct a second correlation feature between process and flow resistance; the softening risk coefficient is correlated with the extrusion speed and extrusion temperature to construct a third correlation feature between softening risk coefficients. The oxidation rate coefficient is correlated with extrusion temperature and time to construct a fourth correlation feature to characterize the effect of extrusion temperature and time on the oxidation rate; the oxidation rate coefficient is correlated with extrusion pressure and extrusion speed to construct a fifth correlation feature to characterize the regulatory effect of the process on the oxidation rate; the oxide layer thickness is correlated with extrusion time and extrusion temperature to construct a sixth correlation feature to quantify the oxide thickness corresponding to a unit extrusion temperature. Based on the first association feature, the second association feature, the third association feature, the fourth association feature, the fifth association feature, and the sixth association feature, an association feature vector is generated; The thermoplastic feature vector, oxidation feature vector, working condition feature vector, and associated feature vector are concatenated to obtain an initial high-dimensional feature matrix; The features in the initial high-dimensional feature matrix are weighted and fused according to the attention weights of each feature to generate the fused feature vector.

[0045] In this embodiment of the invention, the working condition characteristics corresponding to the target aluminum profile material parameters and the target extrusion process parameters include: yield strength, tensile strength, melting point, recrystallization temperature, extrusion temperature, extrusion speed, extrusion pressure, coefficient of friction, extrusion ratio, and extrusion time; wherein, yield strength, tensile strength, melting point, and recrystallization temperature are material-related working condition characteristics; extrusion temperature, extrusion speed, extrusion pressure, coefficient of friction, extrusion ratio, and extrusion time are process parameter-related working condition characteristics.

[0046] First, correlation features are constructed based on thermoplasticity feature vector, oxidation risk feature vector and operating condition features to characterize the relationship between operating conditions and properties.

[0047] Specifically, the first association feature is generated using the following formula: ;in, The first associated feature is the quantification of the effect of temperature on the enhancement / weakening of thermoplastic strength; The second association feature is generated using the following formula: ;in, This is the second correlation feature, which directly reflects the process's adjustment of flow resistance; This refers to the extrusion ratio; The third association feature is generated using the following formula: ;in, This is the third correlation feature, which directly reflects the inhibitory effect of process parameters on softening risk; This refers to the extrusion speed; The fourth association feature is generated using the following formula: ;in, The fourth correlation feature quantifies the effect of temperature-time on the oxidation rate in a single time period; The fifth association feature is generated using the following formula: ;in, This is the fifth correlation feature, which reflects the indirect regulatory effect of process parameters on the oxidation rate; P is the extrusion pressure. The sixth association feature is generated using the following formula: ;in, The sixth associated feature is the oxidation thickness corresponding to the extrusion temperature as its quantification unit. Based on the first, second, third, fourth, fifth, and sixth association features mentioned above, an association feature vector is generated.

[0048] After obtaining the associated feature vector, the thermoplastic feature vector, oxidation feature vector, working condition feature, and associated feature vector are concatenated in sequence to obtain an initial high-dimensional feature matrix. Then, attention weights for each feature are learned based on a neural network. The attention weights are used to weight each feature to obtain weighted features that highlight the features that are more important for chamfer optimization. Finally, the weighted features are deeply fused and dimensionally compressed through a 2-layer fully connected network to obtain the final fused feature vector.

[0049] 105. The fusion feature is transmitted to the initial chamfer parameter generation layer to generate a preliminary chamfer parameter combination.

[0050] Specifically, in this embodiment, the fused feature vector is input into the initial chamfer parameter generation layer to generate the initial chamfer radius, angle, and transition length, thus obtaining the initial chamfer parameter combination.

[0051] 106. Based on the feature parameter mapping layer, according to the preset aluminum profile characteristic association weight matrix, the thermoplastic feature vector, oxidation risk feature vector and preliminary chamfer parameter combination are mapped and calculated to fine-tune the preliminary parameters for characteristic adaptation and generate the fine-tuned chamfer parameter combination.

[0052] In a preferred embodiment, the feature parameter mapping layer performs mapping calculations on the thermoplastic feature vector, oxidation risk feature vector, and preliminary chamfer parameter combination according to a preset aluminum profile characteristic association weight matrix, in order to fine-tune the preliminary parameters for characteristic adaptability and generate a fine-tuned chamfer parameter combination, including: Based on the aluminum profile characteristic correlation weight matrix, thermoplasticity eigenvector, and oxidation risk eigenvector, calculate the characteristic influence coefficient of the chamfer parameter combination; The parameter adjustment amount is calculated based on the characteristic influence coefficient and the initial chamfer parameter combination; The initial chamfer parameter combination is fine-tuned according to the parameter adjustment amount to generate the fine-tuned chamfer parameter combination.

[0053] Specifically, the aluminum profile property correlation weight matrix is ​​used to define the influence weight of thermoplastic / oxidation properties on each chamfer parameter. It is determined based on engineering experience and historical data statistics, and its specific expression is as follows: ; Wherein, W is the correlation weight matrix of the above aluminum profile characteristics; , , These represent the weights of the influence of the thermoplastic strength coefficient on the chamfer radius, angle, and transition length, respectively. , , These represent the weights of the influence of the flow resistance coefficient on the chamfer radius, angle, and transition length, respectively. , , These represent the weights of the softening risk coefficient on the chamfer radius, angle, and transition length, respectively. , , These represent the weights of the influence of the oxidation rate coefficient on the chamfer radius, angle, and transition length, respectively. , , The weights of the influence of oxide layer thickness on chamfer radius, angle, and transition length are respectively. Indicative ; Based on the aforementioned aluminum profile characteristic correlation weight matrix, the characteristic influence coefficient of each chamfer parameter is calculated to quantify the comprehensive impact of characteristics on individual parameters: C=X 特性 ×W; Among them, X 特性 =[ , , , , ];C=[ , , ]; The characteristic influence coefficient of the chamfer radius; This is the characteristic influence coefficient of the chamfer angle; The characteristic influence coefficient for the transition length; Next, based on the characteristic influence coefficient and the initial chamfer parameter combination, the fine-tuning amount (ΔR, Δθ, ΔL) of each chamfer parameter is calculated. The fine-tuning magnitude is positively correlated with the characteristic influence coefficient, while limiting the maximum fine-tuning ratio (to avoid abrupt parameter changes). The specific formula is as follows: Fine-tuning amount ΔR for chamfer radius: ΔR = R init × ×0.3, (maximum fine-tuning ±30%) Fine-tuning amount Δθ for chamfer angle: Δθ = θ init × ×0.2, (maximum fine-tuning ±20%); Fine-tuning amount ΔL for transition length: ΔL=L init × ×0.25, (maximum fine-tuning ±25%) Rinit is the initial chamfer radius in the initial chamfer parameter combination; θinit is the initial chamfer angle; This is the initial transition length.

[0054] After determining the fine-tuning amount, it is necessary to determine the adjustment direction of each parameter, and the adjustment direction of each parameter is based on... , , , , The physical meaning of these five characteristics is determined; for When it is greater than the corresponding reference value, the adjustment direction of each chamfer parameter is: R is adjusted upwards, θ is adjusted downwards; L is adjusted downwards; this conforms to the physical law that when thermoplasticity is good, a small radius + a small angle + a short transition can meet the flow requirements. for When the value is greater than the corresponding reference value, the adjustment direction of each chamfer parameter is: R is adjusted upwards, θ is adjusted upwards, and L is adjusted upwards. This indicates that when the resistance is high, a large radius + a large angle + a long transition can reduce the resistance. When the value is less than the corresponding reference value, the adjustment direction of each chamfer parameter is reversed. When the value is equal to the corresponding reference value, each chamfer parameter does not need to be adjusted.

[0055] for When the value is greater than the corresponding reference value, the adjustment direction of each chamfer parameter is: R is adjusted down, θ is adjusted down, and L is adjusted up. This indicates that during softening, a small radius + small angle + long transition can avoid stress concentration in the mold. When the value is less than the corresponding reference value, the adjustment direction of each chamfer parameter is opposite. When the value is equal to the corresponding reference value, each chamfer parameter does not need to be adjusted.

[0056] for When the value is greater than the corresponding reference value, the adjustment direction of each chamfer parameter is: R is adjusted down, θ is adjusted up; L is adjusted down; this indicates that when oxidation is fast, a small radius + a large angle + a short transition can reduce oxide layer adhesion; when the value is less than the corresponding reference value, the adjustment direction of each chamfer parameter is opposite; when the value is equal to the corresponding reference value, each chamfer parameter does not need to be adjusted.

[0057] for When it is greater than the corresponding reference value, the adjustment direction of each chamfer parameter is: R is adjusted down, θ is adjusted up; L is adjusted down; this means that when the oxide layer is thick, a small radius + a large angle + a short transition can reduce oxide layer adhesion; when it is less than the corresponding reference value, the adjustment direction of each chamfer parameter is opposite; when it is equal to the corresponding reference value, each chamfer parameter does not need to be adjusted.

[0058] It should be noted that the baseline values ​​for each characteristic feature can be set in advance according to the actual situation.

[0059] The corresponding adjustment direction can be determined based on the values ​​of each characteristic feature. Then, based on the adjustment direction and the corresponding parameter adjustment amount, the initial chamfer parameter combination is fine-tuned to generate the fine-tuned chamfer parameter combination.

[0060] Additionally, it should be noted that when different characteristics conflict in their fine-tuning directions for the same parameter (e.g.) θ needs to be adjusted downwards. When θ is required to be adjusted upwards, the characteristic with the larger weight in the aluminum profile characteristic correlation weight matrix shall be used as the standard (e.g., if...). > Then according to The corresponding direction is fine-tuned (θ).

[0061] 107. Input the fine-tuned chamfer parameter combination into the physical geometry constraint branch layer so that the physical geometry constraint branch layer can determine whether the fine-tuned chamfer parameter combination meets the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction.

[0062] In a preferred embodiment, the step of inputting the fine-tuned chamfer parameter combination into the physical geometry constraint branch layer, so that the physical geometry constraint branch layer determines whether the fine-tuned chamfer parameter combination satisfies the preset chamfer parameter constraints and generates the corresponding physical constraint satisfaction, includes: Determine whether each chamfer parameter in the fine-tuned chamfer parameter combination is within the corresponding physical geometric constraint range, and determine the number of parameters that are not within the corresponding physical geometric constraint range. Based on the number of parameters that are not within the range of the corresponding physical geometric constraints, the standard physical constraint satisfaction is linearly reduced to generate the corresponding physical constraint satisfaction.

[0063] Specifically, firstly, the physical geometric constraint range values ​​of each chamfer parameter are preset according to the actual situation; for example, the physical geometric constraint range value of the chamfer radius can be: 0.5mm ≤ R ≤ 5.0mm; this range can ensure the strength of the mold and the material flow guidance effect, avoiding stress concentration due to being too small or affecting the mold structure due to being too large; the physical geometric constraint range value of the chamfer angle can be: 30° ≤ θ ≤ 60°; this range can ensure the feasibility of chamfer processing and the smoothness of material flow; the physical geometric constraint range value of the transition length can be: 1.0mm ≤ L ≤ 3.0mm; this range can balance the chamfer transition effect and the overall size of the mold, avoiding excessive length or shortness affecting the extrusion stability.

[0064] The standard physical constraint satisfaction is set to 1; whenever a chamfer parameter is outside its corresponding physical geometric constraint range, the value is reduced by 0.25; finally, the total reduction value is determined based on the number of parameters outside the corresponding physical geometric constraint range, and the final physical constraint satisfaction is obtained.

[0065] 108. Based on the fusion layer, the fine-tuned chamfer parameter combination is adjusted a second time according to the physical constraint satisfaction to generate the final chamfer parameter combination.

[0066] Specifically, when the physical constraint satisfaction is not 1, a secondary adjustment is needed to the fine-tuned chamfer parameter combination. Illustratively, during adjustment, first identify the chamfer parameters that are not within the corresponding physical geometric constraint range and treat them as the chamfer parameters to be adjusted. If the chamfer parameter to be adjusted is lower than the minimum value of its physical geometric constraint range, then adjust it to the minimum value of its corresponding physical geometric constraint range; if the chamfer parameter to be adjusted is greater than the maximum value of its physical geometric constraint range, then adjust it to the maximum value of its corresponding physical geometric constraint range, ultimately obtaining the final chamfer parameter combination.

[0067] 109. Based on the final chamfer parameter combination, optimize the chamfer of the aluminum profile extrusion die to be optimized.

[0068] Specifically, the existing chamfer parameter combination of the aluminum profile extrusion die to be optimized is compared with the final chamfer parameter combination. Based on the comparison results, the required adjustment amount of the existing chamfer parameter combination is determined, and corresponding adjustments are made according to the required adjustment amount to achieve the optimization of the aluminum profile extrusion die.

[0069] like Figure 2 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides a chamfering optimization device for aluminum profile extrusion dies, comprising: It includes: parameter acquisition module, model parameter input module, feature perception module, feature extraction module, preliminary chamfer generation module, chamfer parameter primary adjustment module, physical constraint module, chamfer secondary adjustment module, and optimization module; The parameter acquisition module is used to acquire the target aluminum profile material parameters and the target extrusion process parameters; The model parameter input module is used to input the target aluminum profile material parameters and target extrusion process parameters into the chamfer parameter prediction model; wherein, the chamfer parameter prediction model includes: an aluminum profile characteristic perception layer, a feature extraction layer, an initial chamfer parameter generation layer, a feature parameter mapping layer, a physical geometric constraint branch layer, and a fusion layer; The characteristic sensing module is used to sense the thermoplastic state and oxidation state of the aluminum profile based on the aluminum profile characteristic sensing layer, according to the target aluminum profile material parameters and the target extrusion process parameters, and generate thermoplastic feature vectors and oxidation feature vectors. The feature extraction module is used to fuse the thermoplastic feature vector, the oxidation risk feature vector, and the working condition features corresponding to the target aluminum profile material parameters and the target extrusion process parameters based on the feature extraction layer to generate fused features; The preliminary chamfer generation module is used to transmit the fused features to the initial chamfer parameter generation layer to generate a preliminary chamfer parameter combination; The chamfer parameter adjustment module is used to perform mapping calculations on the thermoplastic feature vector, oxidation risk feature vector and preliminary chamfer parameter combination based on the feature parameter mapping layer and according to the preset aluminum profile characteristic association weight matrix, so as to fine-tune the preliminary parameters for characteristic adaptability and generate the fine-tuned chamfer parameter combination. The physical constraint module is used to input the fine-tuned chamfer parameter combination into the physical geometric constraint branch layer, so that the physical geometric constraint branch layer can determine whether the fine-tuned chamfer parameter combination meets the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction. The chamfer secondary adjustment module is used to perform secondary adjustments on the fine-tuned chamfer parameter combination based on the fusion layer and the physical constraint satisfaction, to generate the final chamfer parameter combination. The optimization module is used to optimize the chamfer of the aluminum profile extrusion die to be optimized based on the final chamfer parameter combination.

[0070] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the aluminum profile extrusion die chamfering optimization method provided by any of the above-described method embodiments of the present invention.

[0071] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0072] Based on the above embodiments of the aluminum profile extrusion die chamfering optimization method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the aluminum profile extrusion die chamfering optimization method of any embodiment of the present invention.

[0073] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0074] The terminal device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0075] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0076] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the XXX method described in any of the above-described method embodiments of the present invention.

[0077] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0078] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for optimizing the chamfering of aluminum profile extrusion dies, characterized in that, include: Obtain the target aluminum profile material parameters and the target extrusion process parameters; The target aluminum profile material parameters and target extrusion process parameters are input into the chamfer parameter prediction model; wherein, the chamfer parameter prediction model includes: an aluminum profile characteristic perception layer, a feature extraction layer, an initial chamfer parameter generation layer, a feature parameter mapping layer, a physical geometric constraint branch layer, and a fusion layer; Based on the aluminum profile characteristic sensing layer, the thermoplastic state and oxidation state of the aluminum profile are sensed according to the target aluminum profile material parameters and target extrusion process parameters, and thermoplastic feature vector and oxidation feature vector are generated. Based on the feature extraction layer, the thermoplastic feature vector, oxidation risk feature vector, and working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters are fused to generate fused features; The fused features are transmitted to the initial chamfer parameter generation layer to generate an initial chamfer parameter combination; Based on the feature parameter mapping layer, according to the preset aluminum profile characteristic association weight matrix, the thermoplastic feature vector, oxidation risk feature vector and preliminary chamfer parameter combination are mapped and calculated to fine-tune the preliminary parameters for characteristic adaptability and generate the fine-tuned chamfer parameter combination. The fine-tuned chamfer parameter combination is input into the physical geometry constraint branch layer so that the physical geometry constraint branch layer can determine whether the fine-tuned chamfer parameter combination meets the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction. Based on the fusion layer, the fine-tuned chamfer parameter combination is adjusted a second time according to the physical constraint satisfaction to generate the final chamfer parameter combination; Based on the final chamfer parameter combination, the chamfer of the aluminum profile extrusion die to be optimized is optimized.

2. The method for optimizing the chamfering of aluminum profile extrusion dies as described in claim 1, characterized in that, The chamfer parameter prediction model is trained using the following method: Obtain several samples consisting of different aluminum profile material parameters and extrusion process parameters; wherein each sample includes aluminum profile material parameters and extrusion process parameters; For each sample, based on the aluminum profile extrusion dies corresponding to different chamfer parameter combinations, and the finished product surface defect rate under the aluminum profile material parameters and extrusion process parameters corresponding to the sample, the chamfer parameter combination with the smallest finished product surface defect rate and less than the preset defect rate is selected as the target chamfer parameter combination for the corresponding sample. Using each sample and its corresponding target chamfer parameter combination as input, and the predicted chamfer parameter combination corresponding to each sample as output, the preset neural network model is iteratively trained until the loss function converges, thereby generating the chamfer parameter prediction model. During each training iteration, the loss function value is calculated based on the deviation between the target chamfer parameter combination and the predicted chamfer parameter combination, as well as the physical constraint satisfaction of the predicted chamfer parameter combination. If the loss function value does not converge, the parameters of the neural network model are adjusted.

3. The method for optimizing the chamfering of aluminum profile extrusion dies as described in claim 2, characterized in that, The surface defect rate of the finished product under the aluminum profile material parameters and extrusion process parameters corresponding to the aluminum profile extrusion dies according to different chamfer parameter combinations includes: Simulation models of extrusion dies for various aluminum profiles were constructed; the chamfer parameter combinations for different simulation models were different. For each aluminum profile extrusion die simulation model, multiple sets of die flow simulations are performed based on different aluminum profile material parameters and extrusion process parameters. From each set of die flow simulation results, the maximum thickness of the oxide layer in the chamfer area, the maximum stress in the die cavity, the standard deviation of the velocity at the profile exit section, and the decrease in peak force during the extrusion process are extracted. The maximum thickness of the oxide layer in the chamfered area, the maximum stress in the die cavity, the standard deviation of the velocity at the profile exit section, and the decrease in peak force during extrusion are input into the preset defect rate fitting model to obtain the finished surface defect rate of the aluminum profile extrusion die under the corresponding combination of chamfer parameters and aluminum profile material parameters and extrusion process parameters.

4. The method for optimizing the chamfering of aluminum profile extrusion dies as described in claim 1, characterized in that, The aluminum profile characteristic sensing layer senses the thermoplastic and oxidation states of the aluminum profile based on the target aluminum profile material parameters and target extrusion process parameters, generating thermoplastic feature vectors and oxidation feature vectors, including: The thermoplastic strength coefficient is calculated based on the extrusion temperature, yield strength, tensile strength, and melting point temperature of the target aluminum profile; the flow resistance coefficient is calculated based on the friction coefficient, extrusion pressure, extrusion speed, and cross-sectional area of ​​the profile; the softening risk coefficient is determined based on the extrusion temperature, melting point temperature, and recrystallization temperature of the target aluminum profile; and a thermoplastic characteristic vector is constructed based on the thermoplastic strength coefficient, flow resistance coefficient, and softening risk coefficient. The oxidation rate coefficient is calculated based on the oxidation reaction activation energy, gas constant, extrusion temperature, and frequency factor; the oxide layer thickness is generated based on the oxidation rate coefficient; and an oxidation feature vector is constructed based on the oxidation rate coefficient and the oxide layer thickness.

5. The method for optimizing the chamfering of aluminum profile extrusion dies as described in claim 4, characterized in that, The process involves fusing the thermoplastic feature vector, oxidation risk feature vector, and the working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters, based on the feature extraction layer, to generate fused features, including: The thermoplastic strength coefficient is correlated with the extrusion temperature and recrystallization temperature to construct a first correlation feature between temperature and thermoplastic strength; the flow resistance coefficient is correlated with the extrusion ratio and friction coefficient to construct a second correlation feature between process and flow resistance; the softening risk coefficient is correlated with the extrusion speed and extrusion temperature to construct a third correlation feature between softening risk coefficients. The oxidation rate coefficient is correlated with extrusion temperature and time to construct a fourth correlation feature to characterize the effect of extrusion temperature and time on the oxidation rate; the oxidation rate coefficient is correlated with extrusion pressure and extrusion speed to construct a fifth correlation feature to characterize the regulatory effect of the process on the oxidation rate; the oxide layer thickness is correlated with extrusion time and extrusion temperature to construct a sixth correlation feature to quantify the oxide thickness corresponding to a unit extrusion temperature. Based on the first association feature, the second association feature, the third association feature, the fourth association feature, the fifth association feature, and the sixth association feature, an association feature vector is generated; The thermoplastic feature vector, oxidation feature vector, working condition feature vector, and associated feature vector are concatenated to obtain an initial high-dimensional feature matrix; The features in the initial high-dimensional feature matrix are weighted and fused according to the attention weights of each feature to generate the fused feature vector.

6. The method for optimizing the chamfering of aluminum profile extrusion dies as described in claim 5, characterized in that, The feature parameter mapping layer, based on a preset aluminum profile characteristic association weight matrix, performs mapping calculations on the thermoplastic feature vector, oxidation risk feature vector, and preliminary chamfer parameter combination to fine-tune the preliminary parameters for characteristic adaptation, generating a fine-tuned chamfer parameter combination, including: Based on the aluminum profile characteristic correlation weight matrix, thermoplasticity eigenvector, and oxidation risk eigenvector, calculate the characteristic influence coefficient of the chamfering parameter combination; The parameter adjustment amount is calculated based on the characteristic influence coefficient and the initial chamfer parameter combination; The initial chamfer parameter combination is fine-tuned according to the parameter adjustment amount to generate the fine-tuned chamfer parameter combination.

7. The method for optimizing the chamfering of aluminum profile extrusion dies as described in claim 6, characterized in that, The process of inputting the fine-tuned chamfer parameter combination into the physical geometry constraint branch layer allows the physical geometry constraint branch layer to determine whether the fine-tuned chamfer parameter combination satisfies the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction, including: Determine whether each chamfer parameter in the fine-tuned chamfer parameter combination is within the corresponding physical geometric constraint range, and determine the number of parameters that are not within the corresponding physical geometric constraint range. Based on the number of parameters that are not within the range of the corresponding physical geometric constraints, the standard physical constraint satisfaction is linearly reduced to generate the corresponding physical constraint satisfaction.

8. A chamfering optimization device for aluminum profile extrusion dies, characterized in that, include: The parameter acquisition module is used to acquire the target aluminum profile material parameters and the target extrusion process parameters; The model parameter input module is used to input the target aluminum profile material parameters and target extrusion process parameters into the chamfer parameter prediction model; wherein, the chamfer parameter prediction model includes: an aluminum profile characteristic perception layer, a feature extraction layer, an initial chamfer parameter generation layer, a feature parameter mapping layer, a physical geometric constraint branch layer, and a fusion layer; The characteristic sensing module is used to sense the thermoplastic state and oxidation state of the aluminum profile based on the aluminum profile characteristic sensing layer, according to the target aluminum profile material parameters and the target extrusion process parameters, and generate thermoplastic feature vectors and oxidation feature vectors. The feature extraction module is used to fuse the thermoplastic feature vector, oxidation risk feature vector, and working condition features corresponding to the target aluminum profile material parameters and target extrusion process parameters based on the feature extraction layer to generate fused features; The preliminary chamfer generation module is used to transmit the fused features to the initial chamfer parameter generation layer to generate a preliminary chamfer parameter combination; The chamfer parameter adjustment module is used to perform mapping calculations on the thermoplastic feature vector, oxidation risk feature vector and preliminary chamfer parameter combination based on the feature parameter mapping layer and the preset aluminum profile characteristic association weight matrix, so as to fine-tune the preliminary parameters for characteristic adaptation and generate the fine-tuned chamfer parameter combination. The physical constraint module is used to input the fine-tuned chamfer parameter combination into the physical geometric constraint branch layer, so that the physical geometric constraint branch layer can determine whether the fine-tuned chamfer parameter combination meets the preset chamfer parameter constraints and generate the corresponding physical constraint satisfaction. The chamfer secondary adjustment module is used to perform secondary adjustments on the fine-tuned chamfer parameter combination based on the fusion layer and the physical constraint satisfaction, to generate the final chamfer parameter combination. The optimization module is used to optimize the chamfer of the aluminum profile extrusion die based on the final chamfer parameter combination.

9. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the aluminum profile extrusion die chamfering optimization method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the aluminum profile extrusion die chamfering optimization method as described in any one of claims 1-7.

Images