Intelligent control method and system for aluminum profile extrusion machine

Abstract

The application discloses an intelligent control method and system for an aluminum profile extruding machine, and belongs to the technical field of extruding machine control.The first multi-source parameter of the current working condition is used to estimate an optimal breakthrough distance based on a preset fuzzy neural network, the optimal breakthrough distance is set, the extruding quality is improved, the breakthrough pressure required is estimated according to the optimal breakthrough distance, the first multi-source parameter and the second multi-source parameter of the current working condition, the extruding speed of the extruding rod at the breakthrough node is further estimated, the breakthrough pressure and the extruding speed at the breakthrough time are used to construct an extruding speed change curve, the required flow change curve and the rotating speed control curve of the motor are further analyzed and estimated, the corresponding motor rotating speed is obtained according to the real-time position of the extruding rod on the stroke section of the optimal breakthrough distance, the motor is controlled and adjusted, the key parameters on the stroke section of the optimal breakthrough distance are predicted and controlled, the aluminum material is ensured to complete breakthrough extrusion on the optimal breakthrough distance, and the extruding quality is ensured.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of extruder control, in particular to an intelligent control method and system for an aluminum profile extruder. BACKGROUND

[0002] Aluminum profile extrusion is a common metal plastic processing technology, which applies pressure to heated aluminum billets through an extruder to form a profile with a required cross-sectional shape through a die. In the extrusion process, the breakthrough stage refers to the initial stage in which the extrusion rod pushes the aluminum billets to fill the extrusion cylinder and begins to flow out of the die opening. This stage is usually accompanied by a sharp rise in extrusion pressure and a sudden change in material flow. Therefore, the control accuracy of the breakthrough stage directly affects the forming quality of the profile, the service life of the die and the stability of the extrusion process. However, in the existing extruder control technology, the optimization of the breakthrough distance is mostly ignored. Currently, the breakthrough distance is mostly set based on operating experience or fixed values, which is difficult to adapt to different working conditions and is prone to cause large fluctuations in the extrusion process and high scrap rates. Moreover, the existing technology lacks accurate prediction of key parameters in the breakthrough stage, such as motor speed, pump flow and extrusion speed, thereby affecting the extrusion quality of the extrusion rod at the breakthrough node. SUMMARY

[0003] To solve the technical problems existing in the prior art, the application provides an intelligent control method for an aluminum profile extruder, which comprises the following steps: S1, obtaining a first multi-source parameter and a second multi-source parameter of a current working condition, and estimating an optimal breakthrough distance st2 required for this extrusion based on a preset fuzzy neural network according to the first multi-source parameter of the current working condition; the first multi-source parameter comprises: a normalized aluminum alloy grade code, an aluminum length, an aluminum diameter, an aluminum temperature and an extrusion cylinder temperature; In the preset fuzzy neural network, a plurality of fuzzy subsets corresponding to each first multi-source parameter are provided in advance, each fuzzy subset is provided with a corresponding center value and width, each fuzzy subset group is provided with a corresponding rule, the fuzzy subset group is obtained by combining one of the fuzzy subsets corresponding to each first multi-source parameter, and each rule is provided with a consequent parameter corresponding to each first multi-source parameter in advance; The optimal breakthrough distance st2 required for this extrusion is estimated based on the preset fuzzy neural network, specifically: The membership degrees of the current first multi-source parameters to the corresponding fuzzy subsets are calculated by using a Gaussian membership function; The rule fitness of each rule is analyzed by product operation according to the membership degrees, and the activation strength of each rule is obtained by normalization; The first adaptation distance is calculated based on the current first multi-source parameters and the consequent parameters of each rule, and the preferred breakthrough distance st2 is calculated based on each first adaptation distance and the activation intensity of each rule. S2. Based on the preferred breakthrough distance st2, and combined with the first and second multi-source parameters of the current working condition, estimate the breakthrough pressure required for this extrusion; the second multi-source parameters include: extrusion ratio, die working zone length, and equivalent diameter at the die working zone. S3. Estimate the extrusion speed V1 of the extrusion bar at the breakthrough node based on the breakthrough pressure required for this extrusion. S4. Construct the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance by using the preferred breakthrough distance st2 and the extrusion speed V1; S5. Based on the extrusion speed variation curve, analyze the estimated required flow rate variation curve of the hydraulic pump; S6. Based on the estimated required flow rate change curve, analyze the speed control curve of the motor corresponding to the hydraulic pump; S7. During this extrusion process, the real-time position of the extrusion rod on the stroke segment of the preferred breakthrough distance is obtained, and the corresponding motor speed is obtained from the speed control curve based on the real-time position. The motor is then controlled by the controller to run at the corresponding motor speed.

[0004] Furthermore, the expression for calculating the membership degree is: ; The expression for calculating the fitness of the rule is as follows: ; The expression for calculating the activation intensity is: ; The preferred breakthrough distance st2 is calculated as follows: ; Let be the membership degree of the current i-th type of first multi-source parameter to its corresponding j-th fuzzy subset. For the current i-th type of first multi-source parameter, The center value of the j-th fuzzy subset corresponding to the i-th type of first multi-source parameter. The width of the j-th fuzzy subset corresponding to the i-th type of first multi-source parameter. The activation strength of the nth rule, Let N be the fitness of the nth rule, and N be the total number of rules. Let I be the index of the fuzzy subset corresponding to the i-th first multi-source parameter in the n-th rule, where I is the number of first multi-source parameters. Let the first adaptive distance be obtained according to the nth rule. For the successor parameter of the current i-th first multi-source parameter in the n-th rule.

[0005] Furthermore, the estimated breakthrough pressure required for this extrusion is specifically as follows: Based on the current aluminum alloy grade code and aluminum temperature, the estimated static deformation resistance f is obtained through a preset database fitted with historical experimental data; Based on the current aluminum diameter D, aluminum length L1, estimated static deformation resistance f, and preferred breakthrough distance st2, analyze and estimate the frictional pressure component P1; Based on the second multi-source parameter and the estimated static deformation resistance f, analyze the estimated deformation pressure component P2 required for uniform deformation of the aluminum material. The sum of the estimated frictional pressure component P1 and the estimated deformation pressure component P2 is used as the breakthrough pressure required for this extrusion.

[0006] Furthermore, the analytical method for estimating the frictional pressure component P1 is specifically as follows: ; μ is the coefficient of friction, M is the contact surface area between the inner wall of the extrusion cylinder and the aluminum material, A is the cross-sectional area of ​​the aluminum material, and kd is the dynamic friction correction factor. For preset adjustment coefficients, The preset attenuation coefficient is used, st0 is the benchmark breakthrough distance, which is calculated by averaging the historical breakthrough distances of the current aluminum profile extrusion press, and e is the natural constant. The analysis method for estimating the deformation pressure component P2 required for uniform deformation of the current aluminum material is as follows: ; λ is the current extrusion ratio, L2 is the length of the die working zone, and D2 is the equivalent diameter at the die working zone.

[0007] Furthermore, the step of estimating the extrusion speed V1 of the extrusion rod at the breakthrough node based on the breakthrough pressure required for this extrusion is specifically achieved by using a pre-trained first deep learning model, taking the first multi-source parameters of the current working condition and the breakthrough pressure required for this extrusion as inputs, and outputting the extrusion speed V1 of the extrusion rod at the breakthrough node.

[0008] Furthermore, the step of analyzing the estimated required flow rate change curve of the hydraulic pump based on the extrusion speed change curve is specifically as follows: The theoretical required flow rate of the hydraulic pump is obtained from the extrusion speed change curve: Q1 = A2 × V, where Q1 is the theoretical required flow rate, A2 is the effective working area of ​​the extrusion cylinder, and V is the extrusion speed corresponding to the extrusion speed change curve. The extrusion speed variation curve is converted into a corresponding acceleration variation curve. The coordinate system of the acceleration variation curve is with acceleration as the vertical axis and extrusion distance of the extrusion rod as the horizontal axis. The compensation coefficient variation curve is obtained based on the acceleration variation curve: Kc=1+ka×a, where Kc is the compensation coefficient, ka is the preset acceleration compensation factor, and a is the acceleration corresponding to the acceleration variation curve. Based on the theoretical required flow rate change curve and the compensation coefficient change curve, obtain the corresponding estimated required flow rate change curve: Q2=Q1×Kc, where Q2 is the estimated required flow rate.

[0009] Furthermore, based on the estimated required flow rate change curve, the speed control curve of the motor corresponding to the hydraulic pump is analyzed. Specifically, through a pre-trained second deep learning model, each estimated required flow rate on the estimated required flow rate change curve is used as input, and the corresponding motor speed is output. The speed control curve is formed based on the output corresponding motor speed.

[0010] The present invention also provides an intelligent control system for an aluminum profile extrusion press, which applies any of the above-described intelligent control methods for aluminum profile extrusion presses, including: The first analysis module acquires the first and second multi-source parameters of the current working condition. Based on the first multi-source parameters of the current working condition, it estimates the preferred breakthrough distance st2 required for this extrusion based on a preset fuzzy neural network. The first multi-source parameters include: normalized aluminum alloy grade code, aluminum length, aluminum diameter, aluminum temperature, and extrusion cylinder temperature. In the preset fuzzy neural network, multiple fuzzy subsets are pre-set for each first multi-source parameter, and each fuzzy subset has a corresponding center value and width; each fuzzy subset group has a corresponding rule; the fuzzy subset group is obtained by combining one of the fuzzy subsets corresponding to each first multi-source parameter; each rule corresponds to a preset consequent parameter for each first multi-source parameter. The optimal breakthrough distance st2 required for this compression is estimated based on a preset fuzzy neural network, specifically as follows: The membership degree of each first multi-source parameter to its corresponding fuzzy subset is calculated using a Gaussian membership function. The fitness of each rule is analyzed by multiplication based on the membership degree, and then normalized to obtain the activation strength of each rule. The first adaptation distance is calculated based on the current first multi-source parameters and the consequent parameters of each rule, and the preferred breakthrough distance st2 is calculated based on each first adaptation distance and the activation intensity of each rule. The second analysis module estimates the breakthrough pressure required for this extrusion based on the preferred breakthrough distance st2 and the first and second multi-source parameters of the current working conditions. The second multi-source parameters include: extrusion ratio, die working zone length, and equivalent diameter at the die working zone. The third analysis module estimates the extrusion speed V1 of the extrusion rod at the breakthrough node based on the breakthrough pressure required for this extrusion. The first curve analysis module constructs the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance by using the preferred breakthrough distance st2 and the extrusion speed V1; The second curve analysis module analyzes the estimated required flow rate change curve of the hydraulic pump based on the extrusion speed change curve. The third curve analysis module analyzes the speed control curve of the motor corresponding to the hydraulic pump based on the estimated required flow rate change curve. During the extrusion process, the control module acquires the real-time position of the extrusion rod on the stroke segment of the preferred breakthrough distance, obtains the corresponding motor speed from the speed control curve based on the real-time position, and controls the motor to run at the corresponding motor speed through the controller.

[0011] Furthermore, the estimated breakthrough pressure required for this extrusion is specifically as follows: Based on the current aluminum alloy grade code and aluminum temperature, the estimated static deformation resistance f is obtained through a preset database fitted with historical experimental data; Based on the current aluminum diameter D, aluminum length L1, estimated static deformation resistance f, and preferred breakthrough distance st2, analyze and estimate the frictional pressure component P1; Based on the second multi-source parameter and the estimated static deformation resistance f, analyze the estimated deformation pressure component P2 required for uniform deformation of the aluminum material. The sum of the estimated frictional pressure component P1 and the estimated deformation pressure component P2 is used as the breakthrough pressure required for this extrusion.

[0012] Furthermore, the analytical method for estimating the frictional pressure component P1 is specifically as follows: ; μ is the coefficient of friction, M is the contact surface area between the inner wall of the extrusion cylinder and the aluminum material, A is the cross-sectional area of ​​the aluminum material, and kd is the dynamic friction correction factor. For preset adjustment coefficients, The preset attenuation coefficient is used, st0 is the benchmark breakthrough distance, which is calculated by averaging the historical breakthrough distances of the current aluminum profile extrusion press, and e is the natural constant. The analysis method for estimating the deformation pressure component P2 required for uniform deformation of the current aluminum material is as follows: ; λ is the current extrusion ratio, L2 is the length of the die working zone, and D2 is the equivalent diameter at the die working zone.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention estimates the optimal breakthrough distance based on a preset fuzzy neural network using the first multi-source parameters of the current working condition, thereby optimizing the breakthrough distance and improving extrusion quality. Simultaneously, based on the optimal breakthrough distance and the first and second multi-source parameters of the current working condition, it estimates the required breakthrough pressure, and then estimates the extrusion speed of the extrusion rod at the breakthrough point. Based on the breakthrough pressure and the extrusion speed at breakthrough, it constructs an extrusion speed variation curve, and then analyzes and estimates the required flow rate variation curve and the motor speed control curve. Combined with the real-time position of the extrusion rod in the stroke segment of the optimal breakthrough distance, it obtains the corresponding motor speed and controls and adjusts the motor, completing the prediction and control of key parameters in the stroke segment of the optimal breakthrough distance, ensuring that the aluminum material completes the breakthrough extrusion at the optimal breakthrough distance, and guaranteeing extrusion quality. Attached Figure Description

[0014] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of an intelligent control method for an aluminum profile extrusion press according to the present invention; Figure 2 This is a structural block diagram of an intelligent control system for an aluminum profile extrusion machine according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0019] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0020] Example 1 See Figure 1 As shown, the present invention provides an intelligent control method for an aluminum profile extrusion press, which specifically includes the following steps: S1. Obtain the first and second multi-source parameters of the current working condition. Based on the first multi-source parameters of the current working condition, estimate the preferred breakthrough distance st2 required for this extrusion based on the preset fuzzy neural network. S2. Based on the preferred breakthrough distance, and combined with the first and second multi-source parameters of the current working conditions, estimate the breakthrough pressure required for this extrusion. S3. Estimate the extrusion speed V1 of the extrusion bar at the breakthrough node based on the breakthrough pressure required for this extrusion. S4. Construct the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance by using the preferred breakthrough distance st2 and the extrusion speed V1; S5. Based on the extrusion speed variation curve, analyze the estimated required flow rate variation curve of the hydraulic pump; S6. Based on the estimated required flow rate change curve, analyze the speed control curve of the motor corresponding to the hydraulic pump; S7. During this extrusion process, the real-time position of the extrusion rod on the stroke segment of the preferred breakthrough distance is obtained, and the corresponding motor speed is obtained from the speed control curve based on the real-time position. The motor is then controlled by the controller to run at the corresponding motor speed.

[0021] The following provides a detailed explanation of each step: S1. Based on the first multi-source parameters of the current working condition, estimate the preferred breakthrough distance st2 required for this extrusion based on the preset fuzzy neural network.

[0022] In step S1, the first multi-source parameters include: normalized aluminum alloy grade code, aluminum length, aluminum diameter, aluminum temperature, and extrusion cylinder temperature.

[0023] The aluminum alloy grade code determines the rheological stress of the aluminum ingot at high temperatures, that is, the material's ability to resist plastic deformation. Different grades of aluminum alloys have different chemical compositions and strengthening mechanisms, resulting in significant differences in yield strength at the same temperature and strain rate. For example, soft alloys have lower deformation resistance, require less extrusion pressure during upsetting, and are easier to fill the extrusion cylinder, so the required breakthrough distance is relatively short. Hard alloys have higher deformation resistance, requiring greater upsetting force to fully deform them and fill the gaps in the cylinder wall, so a longer breakthrough distance is needed to build up sufficient pressure.

[0024] The length of the aluminum billet is the most important parameter affecting the extrusion stroke. The longer the length, the more stroke is needed to compress the entire aluminum billet and make it fill the extrusion cylinder evenly.

[0025] Temperature is the most active factor affecting the flow stress of aluminum. When the temperature is high, aluminum softens and its yield strength decreases. It can undergo plastic deformation under relatively low pressure, and the stroke required to fill the extrusion cylinder is shorter.

[0026] The extrusion cylinder temperature determines the interface conditions when the aluminum material comes into contact with the cylinder wall. A high cylinder temperature and a small temperature difference between the aluminum material result in slow heat conduction. As a result, the aluminum material experiences less temperature drop during the upsetting process, maintains lower deformation resistance, and the metal flows more easily, leading to a shorter breakthrough distance.

[0027] In step S1, in the preset fuzzy neural network, multiple fuzzy subsets are provided in advance for each first multi-source parameter, and each fuzzy subset has a corresponding center value and width; each fuzzy subset group has a corresponding rule; the fuzzy subset group is obtained by combining one of the fuzzy subsets corresponding to each first multi-source parameter; each rule corresponds to a preset consequent parameter for each first multi-source parameter.

[0028] For example, taking only two types of first multi-source parameters as examples: for the length of aluminum material, there are three fuzzy subsets, namely low length subset, medium length subset and high length subset; for the diameter of aluminum material, there are three fuzzy subsets, namely low diameter subset, medium diameter subset and high diameter subset; among them, the low length subset and the low diameter subset can be combined into a fuzzy subset group, and there are a total of 9 fuzzy subset groups in this example.

[0029] In the fuzzy neural network of this scheme, the consequent parameter refers to the parameter in the "THEN" part of the fuzzy rule. The rule records the consequent parameter that matches each first multi-source parameter of the corresponding fuzzy subset group. For example, taking only the two first multi-source parameters of aluminum material length and aluminum material diameter as examples, one set of fuzzy subset groups is the low length subset and the low diameter subset. The corresponding rule is "the consequent parameter of aluminum material length is a, and the consequent parameter of aluminum material diameter is b".

[0030] In step S1, the estimation of the preferred breakthrough distance st2 required for this squeeze based on a preset fuzzy neural network specifically involves: S11. Calculate the membership degree of each of the current first multi-source parameters to their corresponding fuzzy subsets using a Gaussian membership function: ; Let be the membership degree of the current i-th type of first multi-source parameter to its corresponding j-th fuzzy subset. For the current i-th type of first multi-source parameter, The center value of the j-th fuzzy subset corresponding to the i-th type of first multi-source parameter. The width of the j-th fuzzy subset corresponding to the i-th type of first multi-source parameter; S12. Analyze the fitness of each rule based on its membership degree through product operations, and then normalize it to obtain the activation strength of each rule: ; The activation strength of the nth rule, Let N be the fitness of the nth rule, and N be the total number of rules. Let I be the index of the fuzzy subset corresponding to the i-th first multi-source parameter in the n-th rule, where I is the number of first multi-source parameters. S13. Calculate the corresponding first adaptation distance based on the current first multi-source parameters and the consequent parameters of each rule, and then calculate the preferred breakthrough distance st2 based on each first adaptation distance and the activation intensity of each rule: ; Let the first adaptive distance be obtained according to the nth rule. For the successor parameter of the current i-th first multi-source parameter in the n-th rule.

[0031] Breakthrough distance refers to the distance the extrusion rod travels during the initial stage of the extrusion cycle, from the moment the extrusion rod contacts the aluminum material and applies pressure until the metal flows out uniformly from the die orifice (extrusion through the die orifice). This parameter is crucial to the stability of the extrusion process, the surface quality of the product, and the lifespan of the die. Currently, the setting of the breakthrough distance often relies on operational experience or fixed values, making it difficult to adapt to different working conditions. This can easily lead to large fluctuations in the extrusion process and a high scrap rate. For example, if the breakthrough distance is too short (i.e., the metal flows out before it has fully filled the angle between the extrusion cylinder and the die), it can easily lead to loose texture, shrinkage, or dimensional instability in the product. This embodiment estimates the optimal breakthrough distance required for this extrusion based on the first multi-source parameters of the current working conditions, reaching an optimal stroke range that ensures smooth material output, product quality, and equipment safety.

[0032] S2. Based on the preferred breakthrough distance, and combined with the first and second multi-source parameters of the current working conditions, estimate the breakthrough pressure required for this extrusion.

[0033] In step S2, the second multi-source parameters include: extrusion ratio, die working strip length, and equivalent diameter at the die working strip.

[0034] In step S2, estimating the breakthrough pressure required for this extrusion specifically involves: S21. Based on the current aluminum alloy grade code and aluminum temperature, obtain the estimated static deformation resistance f through a preset database fitted with historical experimental data; The preset database, which is based on fitting historical experimental data, records the estimated static deformation resistance corresponding to different aluminum alloy grade codes and aluminum temperatures.

[0035] Static deformation resistance refers to the ability of a material to resist plastic deformation; it is the force per unit area required to cause a metal to begin plastic flow at a certain temperature.

[0036] In another embodiment, the estimated static deformation resistance in step S21 can be obtained by a pre-trained third deep learning model, using the current aluminum alloy grade code and aluminum temperature as input; the first deep learning model is trained using a large number of different aluminum alloy grade codes, aluminum temperatures, and corresponding experimental static deformation resistances as training samples, and the experimental static deformation resistance is obtained by actual experimental analysis of aluminum with the corresponding aluminum alloy grade code and aluminum temperature.

[0037] S22. Based on the current aluminum diameter D, aluminum length L1, estimated static deformation resistance f, and preferred breakthrough distance st2, analyze and estimate the frictional pressure component P1: ; μ is the coefficient of friction, M is the contact surface area between the inner wall of the extrusion cylinder and the aluminum material, A is the cross-sectional area of the aluminum material, kd is the dynamic friction correction factor, is a preset adjustment coefficient, is a preset attenuation coefficient, st0 is the reference breakthrough distance, which is obtained by calculating the average value of the historical breakthrough distances of the current aluminum profile extruder, and e is the natural constant.

[0038] The preset attenuation coefficient β determines the influence degree of the change of the preferred breakthrough distance st2 on the dynamic friction correction factor kd. The larger the β value, the less sensitive the pressure is to the change of the preferred breakthrough distance st2. When st0 < st2, kd < 1, which means that it is hoped to build pressure slowly, and a lower peak impact is obtained by extending the pressure building time. When st0 > st2, kd > 1, which means that it is hoped to break through quickly, which means that the pressure needs to be doubled to achieve a short-distance breakthrough.

[0039] The frictional force is distributed on the side surface, but the pressure on the side surface cannot be directly measured, and only the end face of the extrusion rod can be controlled. Therefore, the frictional effect scattered on the side surface is converted to the end face of the extrusion rod through the ratio M / A.

[0040] S23. Analyze the estimated deformation pressure component P2 required for the current aluminum material to deform uniformly according to the second multi-source parameter and the estimated static deformation resistance f: ; λ is the current extrusion ratio, L2 is the length of the die working belt, and D2 is the equivalent diameter at the die working belt.

[0041] The first term represents the unit deformation force required to change the shape of the aluminum material only under ideal conditions; The second term represents the unit deformation force that must be additionally consumed to overcome the friction of the working belt wall when the aluminum material passes through the die sizing area.

[0042] S24. Take the sum of the estimated frictional pressure component P1 and the estimated deformation pressure component P2 as the breakthrough pressure required for this extrusion.

[0043] In step S3, estimate the extrusion speed V1 of the extrusion rod at the breakthrough node according to the breakthrough pressure required for this extrusion. Specifically, through a pre-trained first deep learning model, with the first multi-source parameter of the current working condition and the breakthrough pressure required for this extrusion as inputs, the extrusion speed V1 of the extrusion rod at the breakthrough node is output.

[0044] The first deep learning model is trained with a large number of different historical actual breakthrough pressures, historical first multi-source parameters and corresponding historical extrusion speeds as training samples.

[0045] In step S4, the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance is constructed by using the preferred breakthrough distance st2 and the extrusion speed V1, specifically as follows: A coordinate system is constructed with the extrusion speed as the vertical axis and the extrusion distance of the extrusion rod as the horizontal axis. The extrusion speed V1 and the preferred breakthrough distance st2 form the end coordinate point. The end coordinate point is connected to the origin of the coordinate system to form the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance.

[0046] S5. Based on the extrusion speed variation curve, analyze the estimated required flow rate variation curve of the hydraulic pump, specifically as follows: S51. Obtain the theoretical required flow rate change curve of the hydraulic pump based on the extrusion speed change curve: Q1=A2×V, where Q1 is the theoretical required flow rate, A2 is the effective working area of ​​the extrusion cylinder, and V is the extrusion speed corresponding to the extrusion speed change curve. S52. Convert the extrusion speed change curve into the corresponding acceleration change curve. The coordinate system of the acceleration change curve is with acceleration as the vertical axis and extrusion distance of the extrusion rod as the horizontal axis. Obtain the compensation coefficient change curve based on the acceleration change curve: Kc=1+ka×a, where Kc is the compensation coefficient, ka is the preset acceleration compensation factor, and a is the acceleration corresponding to the acceleration change curve. S53. Obtain the corresponding estimated required flow rate change curve based on the theoretical required flow rate change curve and the compensation coefficient change curve: Q2=Q1×Kc, where Q2 is the estimated required flow rate.

[0047] In this scheme, the hydraulic pump flow rate refers to the volume of oil output by the hydraulic pump per unit time. In the hydraulic system of the extruder, the high-pressure oil output by the pump enters the extrusion cylinder, pushes the extrusion rod forward, and thus extrudes the aluminum rod into the die to form a profile.

[0048] In step S6, based on the estimated required flow rate change curve, the speed control curve of the motor corresponding to the hydraulic pump is analyzed. Specifically, through a pre-trained second deep learning model, each estimated required flow rate on the estimated required flow rate change curve is used as input, and the corresponding motor speed is output. The speed control curve is formed based on the output corresponding motor speed.

[0049] The speed control curve is generated based on the corresponding motor speeds output. Specifically, a first coordinate system is constructed with the motor speed as the vertical axis and the extrusion distance of the extrusion rod as the horizontal axis. The coordinate points obtained by the motor speed and the corresponding extrusion distance output by the second deep learning model are mapped onto the first coordinate system and then smoothly connected to form the speed control curve.

[0050] The second deep learning model is trained using a large number of different historical hydraulic pump flow rates and corresponding historical motor speeds as training samples, specifically: A single-point mapping mode is adopted (i.e., inputting a single hydraulic pump flow value and outputting the corresponding motor speed). Each historical hydraulic pump flow-historical motor speed pair is used to form a training sample. All training samples are randomly divided into training set, validation set and test set. Initialize the model parameters of the second deep learning model; Perform forward propagation: Input the training set into the network and calculate the predicted rotational speed; Calculate the loss: Calculate the error between the predicted speed and the actual value (i.e., the corresponding historical motor speed) using the loss function; Perform backpropagation: Calculate the gradient using automatic differentiation and update the model parameters using the Adam optimizer; The loss of the second deep learning model after each update of the model parameters is calculated using the validation set, and the model parameters corresponding to the smallest loss are retained as the model parameters of the final second deep learning model. The final second deep learning model is tested using a test set, and the required evaluation metrics (such as root mean square error) are calculated. If the evaluation metrics fail, the training samples are updated and the model is retrained.

[0051] Example 2 See Figure 2 As shown, the present invention also provides an intelligent control system for an aluminum profile extrusion press, specifically comprising: The first analysis module acquires the first and second multi-source parameters of the current working condition. Based on the first multi-source parameters of the current working condition, it estimates the preferred breakthrough distance st2 required for this extrusion based on a preset fuzzy neural network. The first multi-source parameters include: normalized aluminum alloy grade code, aluminum length, aluminum diameter, aluminum temperature, and extrusion cylinder temperature. In the preset fuzzy neural network, multiple fuzzy subsets are pre-set for each first multi-source parameter, and each fuzzy subset has a corresponding center value and width; each fuzzy subset group has a corresponding rule; the fuzzy subset group is obtained by combining one of the fuzzy subsets corresponding to each first multi-source parameter; each rule corresponds to a preset consequent parameter for each first multi-source parameter. The optimal breakthrough distance st2 required for this compression is estimated based on a preset fuzzy neural network, specifically as follows: The membership degree of each first multi-source parameter to its corresponding fuzzy subset is calculated using a Gaussian membership function. The fitness of each rule is analyzed by multiplication based on the membership degree, and then normalized to obtain the activation strength of each rule. The first adaptation distance is calculated based on the current first multi-source parameters and the consequent parameters of each rule, and the preferred breakthrough distance st2 is calculated based on each first adaptation distance and the activation intensity of each rule. The second analysis module estimates the breakthrough pressure required for this extrusion based on the preferred breakthrough distance st2 and the first and second multi-source parameters of the current working conditions. The second multi-source parameters include: extrusion ratio, die working zone length, and equivalent diameter at the die working zone. The third analysis module estimates the extrusion speed V1 of the extrusion rod at the breakthrough node based on the breakthrough pressure required for this extrusion. The first curve analysis module constructs the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance by using the preferred breakthrough distance st2 and the extrusion speed V1; The second curve analysis module analyzes the estimated required flow rate change curve of the hydraulic pump based on the extrusion speed change curve. The third curve analysis module analyzes the speed control curve of the motor corresponding to the hydraulic pump based on the estimated required flow rate change curve. During the extrusion process, the control module acquires the real-time position of the extrusion rod on the stroke segment of the preferred breakthrough distance, obtains the corresponding motor speed from the speed control curve based on the real-time position, and controls the motor to run at the corresponding motor speed through the controller.

[0052] Example 3 The present invention also provides an electronic device, including: a processor, a transmitting device, an input device, an output device, and a memory. The processor may be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory may be implemented using a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), and is used to store computer program code. The computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any of the above possible implementation methods.

[0053] Example 4 The present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform a method as described in any of the above possible implementations.

[0054] The beneficial effects of this invention are as follows: This invention estimates the optimal breakthrough distance based on a preset fuzzy neural network using the first multi-source parameters of the current working condition, thereby optimizing the breakthrough distance and improving extrusion quality. Simultaneously, based on the optimal breakthrough distance and the first and second multi-source parameters of the current working condition, it estimates the required breakthrough pressure, and then estimates the extrusion speed of the extrusion rod at the breakthrough point. Based on the breakthrough pressure and the extrusion speed at breakthrough, it constructs an extrusion speed variation curve, and then analyzes and estimates the required flow rate variation curve and the motor speed control curve. Combined with the real-time position of the extrusion rod in the stroke segment of the optimal breakthrough distance, it obtains the corresponding motor speed and controls and adjusts the motor, completing the prediction and control of key parameters in the stroke segment of the optimal breakthrough distance, ensuring that the aluminum material completes the breakthrough extrusion at the optimal breakthrough distance, and guaranteeing extrusion quality.

[0055] In the description of this specification, the references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0056] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0057] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for intelligent control of an aluminum profile extrusion press, characterized in that, Includes the following steps: S1. Obtain the first and second multi-source parameters of the current working condition. Based on the first multi-source parameters of the current working condition, estimate the preferred breakthrough distance st2 required for this extrusion based on the preset fuzzy neural network. The first multi-source parameters include: normalized aluminum alloy grade code, aluminum length, aluminum diameter, aluminum temperature, and extrusion cylinder temperature; In the preset fuzzy neural network, multiple fuzzy subsets are pre-set for each first multi-source parameter, and each fuzzy subset has a corresponding center value and width; each fuzzy subset group has a corresponding rule; the fuzzy subset group is obtained by combining one of the fuzzy subsets corresponding to each first multi-source parameter; each rule corresponds to a preset consequent parameter for each first multi-source parameter. The optimal breakthrough distance st2 required for this compression is estimated based on a preset fuzzy neural network, specifically as follows: The membership degree of each first multi-source parameter to its corresponding fuzzy subset is calculated using a Gaussian membership function. The fitness of each rule is analyzed by multiplication based on the membership degree, and then normalized to obtain the activation strength of each rule. The first adaptation distance is calculated based on the current first multi-source parameters and the consequent parameters of each rule, and the preferred breakthrough distance st2 is calculated based on each first adaptation distance and the activation intensity of each rule. S2. Based on the preferred breakthrough distance st2, and combined with the first and second multi-source parameters of the current working condition, estimate the breakthrough pressure required for this extrusion; the second multi-source parameters include: extrusion ratio, die working zone length, and equivalent diameter at the die working zone. S3. Estimate the extrusion speed V1 of the extrusion bar at the breakthrough node based on the breakthrough pressure required for this extrusion. S4. Construct the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance by using the preferred breakthrough distance st2 and the extrusion speed V1; S5. Based on the extrusion speed variation curve, analyze the estimated required flow rate variation curve of the hydraulic pump; S6. Based on the estimated required flow rate change curve, analyze the speed control curve of the motor corresponding to the hydraulic pump; S7. During this extrusion process, the real-time position of the extrusion rod on the stroke segment of the preferred breakthrough distance is obtained, and the corresponding motor speed is obtained from the speed control curve based on the real-time position. The motor is then controlled by the controller to run at the corresponding motor speed.

2. The intelligent control method for aluminum profile extrusion press according to claim 1, characterized in that, The expression for calculating the membership degree is: ； The expression for calculating the fitness of the rule is as follows: ； The expression for calculating the activation intensity is: ； The preferred breakthrough distance st2 is calculated as follows: ； ； Let be the membership degree of the current i-th type of first multi-source parameter to its corresponding j-th fuzzy subset. For the current i-th type of first multi-source parameter, The center value of the j-th fuzzy subset corresponding to the i-th type of first multi-source parameter. The width of the j-th fuzzy subset corresponding to the i-th type of first multi-source parameter. The activation strength of the nth rule, Let N be the fitness of the nth rule, and N be the total number of rules. Let I be the index of the fuzzy subset corresponding to the i-th first multi-source parameter in the n-th rule, where I is the number of first multi-source parameters. Let the first adaptive distance be obtained according to the nth rule. For the successor parameter of the current i-th first multi-source parameter in the n-th rule.

3. The intelligent control method for an aluminum profile extrusion press according to claim 1, characterized in that, The estimated breakthrough pressure required for this extrusion is specifically as follows: Based on the current aluminum alloy grade code and aluminum temperature, the estimated static deformation resistance f is obtained through a preset database fitted with historical experimental data; Based on the current aluminum diameter D, aluminum length L1, estimated static deformation resistance f, and preferred breakthrough distance st2, analyze and estimate the frictional pressure component P1; Based on the second multi-source parameter and the estimated static deformation resistance f, analyze the estimated deformation pressure component P2 required for uniform deformation of the aluminum material. The sum of the estimated frictional pressure component P1 and the estimated deformation pressure component P2 is used as the breakthrough pressure required for this extrusion.

4. The intelligent control method for an aluminum profile extrusion press according to claim 3, characterized in that, The analytical method for estimating the frictional pressure component P1 is as follows: ； ； ； μ is the coefficient of friction, M is the contact surface area between the inner wall of the extrusion cylinder and the aluminum material, A is the cross-sectional area of ​​the aluminum material, and kd is the dynamic friction correction factor. For preset adjustment coefficients, The preset attenuation coefficient is used, st0 is the benchmark breakthrough distance, which is calculated by averaging the historical breakthrough distances of the current aluminum profile extrusion press, and e is the natural constant. The analysis method for estimating the deformation pressure component P2 required for uniform deformation of the current aluminum material is as follows: ； λ is the current extrusion ratio, L2 is the length of the die working zone, and D2 is the equivalent diameter at the die working zone.

5. The intelligent control method for an aluminum profile extrusion press according to claim 1, characterized in that, The method of estimating the extrusion speed V1 of the extrusion rod at the breakthrough node based on the breakthrough pressure required for this extrusion is specifically achieved by using a pre-trained first deep learning model, taking the first multi-source parameters of the current working condition and the breakthrough pressure required for this extrusion as inputs, and outputting the extrusion speed V1 of the extrusion rod at the breakthrough node.

6. The intelligent control method for an aluminum profile extrusion press according to claim 1, characterized in that, The process of analyzing the estimated required flow rate curve of the hydraulic pump based on the extrusion speed variation curve is as follows: The theoretical required flow rate of the hydraulic pump is obtained from the extrusion speed change curve: Q1 = A2 × V, where Q1 is the theoretical required flow rate, A2 is the effective working area of ​​the extrusion cylinder, and V is the extrusion speed corresponding to the extrusion speed change curve. The extrusion speed variation curve is converted into a corresponding acceleration variation curve. The coordinate system of the acceleration variation curve is with acceleration as the vertical axis and extrusion distance of the extrusion rod as the horizontal axis. The compensation coefficient variation curve is obtained based on the acceleration variation curve: Kc=1+ka×a, where Kc is the compensation coefficient, ka is the preset acceleration compensation factor, and a is the acceleration corresponding to the acceleration variation curve. Based on the theoretical required flow rate change curve and the compensation coefficient change curve, obtain the corresponding estimated required flow rate change curve: Q2=Q1×Kc, where Q2 is the estimated required flow rate.

7. The intelligent control method for an aluminum profile extrusion press according to claim 1, characterized in that, The process involves analyzing the speed control curve of the motor corresponding to the hydraulic pump based on the estimated required flow rate change curve. Specifically, a pre-trained second deep learning model takes each estimated required flow rate on the estimated required flow rate change curve as input and outputs the corresponding motor speed. The speed control curve is then formed based on the output corresponding motor speeds.

8. An intelligent control system for an aluminum profile extrusion press, characterized in that, include: The first analysis module obtains the first and second multi-source parameters of the current working condition, and estimates the preferred breakthrough distance st2 required for this extrusion based on the first multi-source parameters of the current working condition and a preset fuzzy neural network. The first multi-source parameters include: normalized aluminum alloy grade code, aluminum length, aluminum diameter, aluminum temperature, and extrusion cylinder temperature; In the preset fuzzy neural network, multiple fuzzy subsets are pre-set for each first multi-source parameter, and each fuzzy subset has a corresponding center value and width; each fuzzy subset group has a corresponding rule; the fuzzy subset group is obtained by combining one of the fuzzy subsets corresponding to each first multi-source parameter; each rule corresponds to a preset consequent parameter for each first multi-source parameter. The optimal breakthrough distance st2 required for this compression is estimated based on a preset fuzzy neural network, specifically as follows: The membership degree of each first multi-source parameter to its corresponding fuzzy subset is calculated using a Gaussian membership function. The fitness of each rule is analyzed by multiplication based on the membership degree, and then normalized to obtain the activation strength of each rule. The first adaptation distance is calculated based on the current first multi-source parameters and the consequent parameters of each rule, and the preferred breakthrough distance st2 is calculated based on each first adaptation distance and the activation intensity of each rule. The second analysis module estimates the breakthrough pressure required for this extrusion based on the preferred breakthrough distance st2 and the first and second multi-source parameters of the current working conditions. The second multi-source parameters include: extrusion ratio, die working zone length, and equivalent diameter at the die working zone. The third analysis module estimates the extrusion speed V1 of the extrusion rod at the breakthrough node based on the breakthrough pressure required for this extrusion. The first curve analysis module constructs the extrusion speed variation curve of the extrusion rod in the stroke segment of the preferred breakthrough distance by using the preferred breakthrough distance st2 and the extrusion speed V1; The second curve analysis module analyzes the estimated required flow rate change curve of the hydraulic pump based on the extrusion speed change curve. The third curve analysis module analyzes the speed control curve of the motor corresponding to the hydraulic pump based on the estimated required flow rate change curve. During the extrusion process, the control module acquires the real-time position of the extrusion rod on the stroke segment of the preferred breakthrough distance, obtains the corresponding motor speed from the speed control curve based on the real-time position, and controls the motor to run at the corresponding motor speed through the controller.

9. The intelligent control system for the aluminum profile extrusion press according to claim 8, characterized in that, The estimated breakthrough pressure required for this extrusion is specifically as follows: Based on the current aluminum alloy grade code and aluminum temperature, the estimated static deformation resistance f is obtained through a preset database fitted with historical experimental data; Based on the current aluminum diameter D, aluminum length L1, estimated static deformation resistance f, and preferred breakthrough distance st2, analyze and estimate the frictional pressure component P1; Based on the second multi-source parameter and the estimated static deformation resistance f, analyze the estimated deformation pressure component P2 required for uniform deformation of the aluminum material. The sum of the estimated frictional pressure component P1 and the estimated deformation pressure component P2 is used as the breakthrough pressure required for this extrusion.

10. The intelligent control system for the aluminum profile extrusion press according to claim 9, characterized in that, The analytical method for estimating the frictional pressure component P1 is as follows: ； μ is the coefficient of friction, M is the contact surface area between the inner wall of the extrusion cylinder and the aluminum material, A is the cross-sectional area of ​​the aluminum material, and kd is the dynamic friction correction factor. For preset adjustment coefficients, The preset attenuation coefficient is used, st0 is the benchmark breakthrough distance, which is calculated by averaging the historical breakthrough distances of the current aluminum profile extrusion press, and e is the natural constant. The analysis method for estimating the deformation pressure component P2 required for uniform deformation of the current aluminum material is as follows: ； λ is the current extrusion ratio, L2 is the length of the die working zone, and D2 is the equivalent diameter at the die working zone.

Images