Extruder hydraulic flow intelligent control method and system

Abstract

The application relates to the technical field of extruders, and provides an extruder hydraulic flow intelligent control method and system, which comprises the following steps: acquiring a first coupling function according to material characteristics and real-time temperature; acquiring a second coupling function for representing the deviation degree and change trend of resistance according to real-time extrusion resistance and historical extrusion resistance; acquiring a load flow correlation function for quantifying the matching degree between the current load state and the historical flow optimal solution according to real-time working load and historical flow data; acquiring a flow correction factor by combining the first coupling function, the second coupling function and the load flow correlation function; acquiring a reference flow according to an extrusion ratio, correcting the reference flow according to the flow correction factor to obtain a target flow, and controlling the hydraulic flow of the extruder according to the target flow. The application considers and couples multiple process parameters, can adapt to the complex and changeable working conditions in the profile extrusion process, and is favorable for improving the hydraulic control precision.

Description

A method and system for intelligent control of hydraulic flow in an extruder Technical Field

[0001] This invention relates to the field of extrusion press technology, and more specifically, to an intelligent control method and system for hydraulic flow in an extrusion press. Background Technology

[0002] Extrusion presses are the main equipment for producing tubes, bars, and profiles of light alloys (aluminum alloys, copper alloys, and magnesium alloys). They mainly consist of three parts: mechanical, hydraulic, and electrical. The hydraulic part mainly consists of a master cylinder, side cylinders, locking cylinders, piercing cylinders, a large-capacity axial piston variable pump, an electro-hydraulic ratio servo valve (or electro-hydraulic proportional regulating valve), position sensors, oil pipes, an oil tank, and various hydraulic switches.

[0003] For extrusion presses, multiple process parameters are often coupled during the extrusion process, which can interfere with the control of hydraulic flow. For example, changes in the temperature of the aluminum billet affect its plasticity, thus altering the extrusion resistance. Changes in extrusion resistance, in turn, affect the pressure and flow requirements of the hydraulic system. If the control system cannot decouple these coupled parameters in real time and only controls a single parameter, the accuracy of flow control will decrease. For instance, when the aluminum billet temperature increases by 10°C, the extrusion resistance may decrease by 10%-15%. If the hydraulic flow is not adjusted in time, it will lead to excessively high extrusion speeds and out-of-tolerance profile dimensions.

[0004] Most existing hydraulic flow control algorithms for extrusion presses are based on simple PID control principles and do not couple multiple process parameters, making them difficult to adapt to the complex and ever-changing working conditions during profile extrusion and thus requiring improvement. Summary of the Invention

[0005] Therefore, in order to solve the problem that existing hydraulic flow control systems for extrusion presses do not couple multiple process parameters and are difficult to adapt to the complex and ever-changing working conditions during profile extrusion, this invention provides an intelligent hydraulic flow control method and system for extrusion presses, the specific technical solution of which is as follows:

[0006] A method for intelligent control of hydraulic flow in an extruder, comprising:

[0007] Obtain the material properties and real-time temperature of the target to be extruded, and obtain the first coupling function to characterize the coupling effect between material properties and temperature based on the material properties and real-time temperature;

[0008] Obtain real-time extrusion resistance and historical extrusion resistance, and obtain a second coupling function based on real-time extrusion resistance and historical extrusion resistance to characterize the degree of resistance deviation and its changing trend;

[0009] Obtain real-time workload and historical traffic data, and obtain a load-traffic correlation function based on the real-time workload and historical traffic data to quantify the matching degree between the current load status and the optimal historical traffic solution;

[0010] The flow correction factor is obtained by combining the first coupling function, the second coupling function, and the load flow correlation function.

[0011] The reference flow rate is obtained based on the extrusion ratio, the reference flow rate is corrected based on the flow correction factor to obtain the target flow rate, and the hydraulic flow rate of the extruder is controlled based on the target flow rate.

[0012] The intelligent hydraulic flow control method for the extruder obtains a flow correction factor by combining a first coupling function, a second coupling function, and a load flow correlation function, and then corrects the reference flow based on the flow correction factor to obtain the target force. It considers and couples multiple process parameters, can adapt to the complex and ever-changing working conditions during profile extrusion, and is conducive to improving the accuracy of hydraulic control.

[0013] Preferably, the specific method for obtaining the first coupling function includes:

[0014] Based on the Arrhenius equation, an exponential term representing the thermal activation process is obtained from the real-time temperature and deformation activation energy.

[0015] Obtain the correction term used to represent the normalization correction of the deviation between the material reference Young's modulus and hardness by referring to Young's modulus;

[0016] The product of the exponential term and the correction term is used as the first coupling function to characterize the coupling effect between material properties and temperature.

[0017] Among them, material properties include deformation activation energy, material reference Young's modulus, and hardness deviation.

[0018] Preferably, the specific method for obtaining the second coupling function includes:

[0019] The historical average resistance is obtained from the historical extrusion resistance, and the hyperbolic tangent term is obtained by using the ratio of the real-time extrusion resistance to the historical average resistance as input.

[0020] The rate of change term is obtained based on the rate of change of real-time extrusion resistance, and the sum of the hyperbolic tangent term and the rate of change term is used as a second coupling function to characterize the degree of resistance deviation and the trend of change.

[0021] Preferably, the specific method for obtaining the load-flow correlation function includes:

[0022] Obtain the load normalization term based on the real-time workload and the system's maximum design load;

[0023] Based on historical traffic data, obtain historical traffic samples and historical best traffic. Based on historical traffic samples and historical best traffic, obtain Gaussian similarity kernel functions and obtain the weighted average of Gaussian similarity kernel functions.

[0024] The product of the load normalization term and the weighted average is used as the load-flow correlation function to quantify the matching degree between the current load state and the historical optimal flow solution.

[0025] Preferably, the specific method for obtaining the flow correction factor includes:

[0026] The flow correction factor is obtained by weighting the first coupling function, the second coupling function, and the load flow correlation function.

[0027] Preferably, the specific method for obtaining the baseline flow rate includes:

[0028] The reference flow rate is obtained by multiplying the extrusion ratio, the reference speed of the extrusion rod, and the cross-sectional area of ​​the billet.

[0029] An intelligent control system for hydraulic flow in an extruder, used to implement the control method described above, includes:

[0030] The first coupling function acquisition module is used to acquire the material properties and real-time temperature of the target to be extruded, and to acquire the first coupling function to characterize the coupling effect between material properties and temperature based on the material properties and real-time temperature.

[0031] The second coupling function acquisition module is used to acquire real-time extrusion resistance and historical extrusion resistance, and to acquire a second coupling function based on real-time extrusion resistance and historical extrusion resistance to characterize the degree of resistance deviation and the trend of change.

[0032] The load-flow correlation function acquisition module is used to acquire real-time workload and historical traffic data, and to acquire a load-flow correlation function based on the real-time workload and historical traffic data to quantify the matching degree between the current load state and the optimal historical traffic solution.

[0033] The traffic correction factor acquisition module is used to obtain the traffic correction factor by combining the first coupling function, the second coupling function, and the load-traffic correlation function.

[0034] The hydraulic flow control module is used to obtain a reference flow rate based on the extrusion ratio, correct the reference flow rate based on the flow correction factor, obtain the target flow rate, and control the hydraulic flow rate of the extruder based on the target flow rate.

[0035] Preferably, the first coupling function acquisition module includes:

[0036] The exponent term acquisition unit is used to acquire an exponent term representing the thermal activation process based on the Arrhenius equation, real-time temperature, and deformation activation energy.

[0037] The correction term acquisition unit is used to acquire correction terms that represent the normalization correction of the deviation between the material reference Young's modulus and hardness by referring to Young's modulus.

[0038] The first coupling function acquisition unit is used to obtain the product of the exponential term and the correction term as the first coupling function for characterizing the coupling effect between material properties and temperature.

[0039] Among them, material properties include deformation activation energy, material reference Young's modulus, and hardness deviation.

[0040] Preferably, the second coupling function acquisition module includes:

[0041] The hyperbolic tangent term acquisition unit is used to obtain the historical average resistance based on the historical extrusion resistance, and to obtain the hyperbolic tangent term by using the ratio of the real-time extrusion resistance to the historical average resistance as input.

[0042] The second coupling function acquisition unit is used to obtain the rate of change term based on the rate of change of the real-time extrusion resistance, and uses the sum of the hyperbolic tangent term and the rate of change term as the second coupling function to characterize the degree of resistance deviation and the trend of change.

[0043] Preferably, the load traffic correlation function acquisition module includes:

[0044] The load normalization term acquisition unit is used to acquire the load normalization term based on the real-time workload and the maximum design load of the system.

[0045] The weighted average value acquisition unit is used to obtain historical traffic samples and historical best traffic based on historical traffic data, obtain Gaussian similarity kernel functions based on historical traffic samples and historical best traffic, and obtain the weighted average value of Gaussian similarity kernel functions.

[0046] The load-flow correlation function acquisition unit is used to obtain the load-flow correlation function by multiplying the load normalization term and the weighted average value, which is used to quantify the matching degree between the current load state and the historical optimal solution of flow. Attached Figure Description

[0047] The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0048] Figure 1 is a schematic flowchart of an intelligent control method for hydraulic flow of an extruder according to an embodiment of the present invention.

[0049] Figure 2 is a flowchart illustrating a specific method for obtaining the first coupling function in one embodiment of the present invention;

[0050] Figure 3 is a flowchart illustrating a specific method for obtaining the second coupling function in one embodiment of the present invention;

[0051] Figure 4 is a flowchart illustrating a specific method for obtaining the load flow correlation function in one embodiment of the present invention;

[0052] Figure 5 is a schematic diagram of the overall structure of an intelligent hydraulic flow control system for an extruder according to an embodiment of the present invention.

[0053] Figure 6 is a schematic diagram of the functional module structure of the first coupling function acquisition module in an embodiment of the present invention;

[0054] Figure 7 is a schematic diagram of the functional module structure of the second coupling function acquisition module in one embodiment of the present invention;

[0055] Figure 8 is a schematic diagram of the functional module structure of the load flow correlation function acquisition module in one embodiment of the present invention;

[0056] Figure 9 is a schematic diagram of the process for oscillation detection and triggering based on the second derivative of real-time extrusion resistance in one embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to its embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of the invention.

[0058] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0059] 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 invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0060] In this invention, "first" and "second" do not represent a specific quantity or order, but are merely used to distinguish names.

[0061] As shown in Figure 1, an embodiment of the present invention provides a method for intelligent control of hydraulic flow in an extruder, which includes the following steps:

[0062] S1, obtain the material properties and real-time temperature of the target to be extruded, and obtain the first coupling function to characterize the coupling effect between material properties and temperature based on the material properties and real-time temperature.

[0063] Specifically, the elastic modulus of an alloy is closely related to the types and amounts of its constituent elements. Generally, the addition of lightweight elements (such as aluminum and magnesium) reduces the alloy's density, but also decreases its elastic modulus to some extent. Conversely, the addition of heavy metal elements (such as tungsten and molybdenum) significantly increases the elastic modulus, but correspondingly increases the material's density and cost. Furthermore, certain alloying elements can indirectly affect the elastic modulus by altering the alloy's crystal structure, lattice constant, and other microscopic properties. The content of alloying elements is also a crucial factor influencing the elastic modulus. Within a certain range, the elastic modulus of the metal increases with the increase of alloying element content. However, excessively high alloying element content may lead to phase transformations and grain coarsening within the metal, thereby reducing the elastic modulus. Therefore, rationally controlling the content of alloying elements is key to optimizing the material's elastic modulus.

[0064] Furthermore, the distribution of alloying elements also affects the elastic modulus. Uniform distribution of alloying elements within the metal matrix is ​​beneficial for improving the material's elastic modulus, while segregation or aggregation can lead to a decrease in the mechanical properties of localized areas, thus affecting the overall elastic modulus. Therefore, during alloy preparation, it is necessary to control the distribution of alloying elements through processes such as finishing and forging.

[0065] For aluminum billet materials, their elemental composition and crystal structure directly influence the key parameters in the first coupling function. First, elements such as Mg and Si form the Mg2Si strengthening phase, which can increase Young's modulus. For example, in 6061 aluminum alloy, the Mg content is 0.8-1.2% and the Si content is 0.4-0.8%, and this ratio directly affects the material's baseline Young's modulus value. The Fe / Si ratio (recommended 1.0-1.5) determines the content of the hard and brittle phases, affecting the hardness deviation (the difference between the measured hardness value and the theoretical hardness value). Excessive Fe content will lead to increased billet brittleness and a larger hardness deviation. Second, material purity and impurity content (such as Na and Ca) affect the diffusion barrier: residual Na2O (2-3%) in electrolytic aluminum billets will reduce the hot deformation activation energy Q value, accelerating high-temperature deformation. Microalloying elements (such as Mn and Cr) can increase the dynamic recrystallization threshold and increase the hot deformation activation energy value by pinning grain boundaries. Third, fine equiaxed grains homogenize impurity distribution, reducing local stress concentration and enhancing deformation uniformity.

[0066] As a preferred technical solution, as shown in Figure 2, the specific method for obtaining the first coupling function includes:

[0067] S11, based on the Arrhenius equation, uses an exponential term to represent the thermal activation process, obtained from the real-time temperature and deformation activation energy.

[0068] Plastic deformation of metals is a thermally activated process. As temperature increases, the atomic diffusion barrier of the metal decreases, and the effective value of the deformation activation energy Q decreases, thus softening the material. The Arrhenius equation is the core formula in chemical kinetics describing the relationship between the reaction rate constant k and temperature T. It applies to elementary reactions, non-elementary reactions, and some heterogeneous reactions. Its core assumption is that the activation energy is constant over a temperature range, quantifying the accelerating effect of temperature on the reaction rate through an exponential relationship. The Arrhenius equation shows that the rate of change of the natural logarithm of the reaction rate constant with temperature T is proportional to the activation energy Ea. In other words, the higher the activation energy, the faster the reaction rate increases with increasing temperature; that is, the higher the activation energy, the more sensitive the reaction rate is to temperature. If several reactions exist simultaneously, higher temperatures favor reactions with higher activation energies, while lower temperatures favor reactions with lower activation energies. In production, this principle is often used to select an appropriate temperature to accelerate the main reaction and suppress side reactions.

[0069] For example, the exponential term is represented as Where e represents the natural constant, Q represents the deformation activation energy, and R and T represent the gas constant and real-time temperature, respectively. The deformation activation energy can be understood as the energy threshold required for a material to undergo plastic deformation, which is usually obtained by fitting a stress-strain curve through thermal simulation experiments. The gas constant is a constant and can be acquired online using infrared thermocouples / embedded thermocouples to collect the real-time temperature of the target to be extruded, such as an aluminum billet.

[0070] S12, obtain the correction term used to represent the normalization correction of the deviation between the material reference Young's modulus and hardness by referring to Young's modulus.

[0071] For example, the correction term is represented as Among them, E M0 ΔH, E ref These represent the reference Young's modulus, hardness deviation, and material-based Young's modulus, respectively. The material-based Young's modulus is generally determined through tensile testing or ultrasonic resonance. The hardness deviation is the difference between the measured hardness value and the theoretical hardness value. The reference Young's modulus is a fixed constant. For example, for pure aluminum 1060, its reference Young's modulus at room temperature is 69 GPa. The material-based Young's modulus is used for normalization, converting the stiffness of different materials into a ratio relative to the material-based Young's modulus (e.g., pure aluminum), thus eliminating the influence of dimensions.

[0072] k MThe material hardening factor represents the influence of alloying elements on the modulus. It is dimensionless and primarily used to compensate for hardness deviations, addressing modulus shifts caused by impurities / grain size anomalies. This material hardening factor is typically obtained by constructing a composition regression model. This involves determining the component weights of each alloying element by fitting a function curve to the calibration model, and then calculating the weighted average of these weights to obtain the material hardening factor. For example, in recycled aluminum containing iron impurities, the hardness deviation can sometimes reach 20 HB, requiring modulus correction using the material hardening factor.

[0073] S13 uses the product of the exponential term and the correction term as the first coupling function to characterize the coupling effect between material properties and temperature.

[0074] For example, the first coupling function is represented as The material properties include deformation activation energy, material baseline Young's modulus, and hardness deviation. In this first coupling function, a nonlinear coupling between real-time temperature and material properties is achieved. As the real-time temperature increases, the exponential term dominates; however, some alloys experience increased precipitate formation at high temperatures, potentially through k... M • ΔH compensates for hardness deviations to dynamically balance the interaction between temperature and material properties.

[0075] Specifically, this first coupling function breaks through the limitations of traditional single empirical formulas by separating and modeling the thermodynamic process (exponential term) and the microstructure evolution (linear term). It quantifies the synergistic influence of the material properties to be extruded and real-time temperature on the extrusion flow rate, solving the defect of independent processing of the two in the hydraulic flow control of traditional extruders. It can achieve the following functions: 1. Temperature sensitivity compensation: dynamic cancellation of high-temperature softening effect and low-temperature embrittlement; 2. Material heterogeneity adaptation: modeling based on the difference in deformation resistance of different alloying elements; 3. Process stability improvement: suppressing the jump in flow rate setpoint caused by material fluctuations.

[0076] S2, obtain the real-time extrusion resistance and historical extrusion resistance, and obtain the second coupling function based on the real-time extrusion resistance and historical extrusion resistance to characterize the degree of resistance deviation and the trend of change.

[0077] Specifically, an imbalance in the iron-silicon ratio can induce microcracks, leading to a sudden increase in real-time extrusion resistance. Residual aluminum ash can form hard spots on the billet surface, increasing the rate of change in real-time extrusion resistance. In this case, dynamic compensation can be applied to compensate for the rate of change in real-time extrusion resistance.

[0078] As a preferred technical solution, as shown in Figure 3, the specific method for obtaining the second coupling function includes:

[0079] S21, obtain the historical average resistance based on the historical extrusion resistance, and obtain the hyperbolic tangent term by using the ratio of the real-time extrusion resistance to the historical average resistance as input.

[0080] Real-time extrusion resistance can be understood as the current thrust of the hydraulic cylinder of the extruder, which can be collected by a pressure sensor. Historical average resistance is the average historical extrusion resistance of extruded products of the same specification. For example, the hyperbolic tangent term is represented as tanh(real-time extrusion resistance F). r Historical average resistance F h When the real-time extrusion resistance equals the historical average resistance, the output of the hyperbolic tangent term is 0, which conforms to the baseline balance principle. When the result of real-time extrusion resistance / historical average resistance approaches zero (extrusion resistance drops sharply), the output of the hyperbolic tangent term is approximately -1, indicating a possible rod breakage fault, requiring an immediate reduction in hydraulic flow. When the result of real-time extrusion resistance / historical average resistance approaches infinity (extrusion resistance increases dramatically), the output of the hyperbolic tangent term is approximately +1, indicating a possible die blockage fault, requiring an immediate increase in hydraulic flow for stamping. Generally, the result of real-time extrusion resistance / historical average resistance should be between 0.8 and 1.2 to reduce vibration during the extrusion process.

[0081] S22, the rate of change term is obtained based on the rate of change of real-time extrusion resistance, and the sum of the hyperbolic tangent term and the rate of change term is used as the second coupling function to characterize the degree of resistance deviation and the trend of change.

[0082] This rate of change can be understood as the time-domain derivative of the real-time extrusion resistance. For example, the rate of change term is expressed as: resistance change sensitivity factor λ × rate of change of real-time extrusion resistance. This rate-of-change term is similar to the derivative term in PID control, but it's an improvement specifically for the extrusion process. Specifically, when the rate of change of real-time extrusion resistance increases positively, it can be interpreted as the extruder being about to overload; in this case, the flow rate can be increased in advance to offset the resistance. When the rate of change of real-time extrusion resistance drops negatively, it can be interpreted as the extruder material becoming unstable; in this case, the hydraulic flow rate should be actively reduced to prevent breakage. The resistance change sensitivity factor is generally an empirical value, for example, set to 0.05.

[0083] Preferably, the method for obtaining the resistance change sensitivity factor is as follows: first, obtain the measured oscillation frequency, a preset critical frequency reference, and a basic damping coefficient; then, determine whether the measured oscillation frequency is greater than the preset critical frequency reference, for example, 10Hz. If so, the basic damping coefficient is linearly increased according to the frequency ratio of the measured oscillation frequency to the critical frequency reference, and the resistance change sensitivity factor is dynamically obtained; otherwise, the basic damping coefficient is used as the resistance change sensitivity factor. For example, when the measured oscillation frequency is greater than the preset critical frequency reference, the resistance change sensitivity factor = basic damping coefficient × (1 + measured oscillation frequency / critical frequency reference); when the measured oscillation frequency is less than or equal to the preset critical frequency reference, the resistance change sensitivity factor = basic damping coefficient.

[0084] The basic damping coefficient can be understood as the structural damping ratio of the system in a static state, and its range is generally set to 0.04-0.09, for example, 0.05 or 0.06. The critical frequency reference can be understood as the starting frequency at which mechanical resonance (standing wave resonance) occurs in the hydraulic pipeline of the hydraulic system. For example, assuming that 10-15Hz is a high-incidence area of ​​mechanical resonance in an extruder hydraulic system, the critical frequency reference can be set to 10Hz. Here, by linearly increasing the basic damping coefficient based on the ratio of the measured oscillation frequency to the critical frequency reference, the resistance change sensitivity factor is obtained. This allows for targeted suppression of the high-frequency resonance band (high-incidence area of ​​mechanical resonance) in the extruder hydraulic system, automatically increasing the value of the resistance change sensitivity factor to enhance damping.

[0085] For example, the second coupling function In this second coupling function, the hyperbolic tangent term is mainly used to cope with sudden changes in resistance and achieve rapid amplitude limiting response, while the rate of change term is mainly used to cope with gradual resistance drift and achieve prediction of extrusion resistance trends. In summary, this second coupling function dynamically quantifies the deviation between real-time extrusion resistance and the historical average resistance, and captures resistance change trends. It combines resistance state normalization with forward-looking control of the rate of change, achieving a leap from passive response to active defense in the extrusion process. It can not only provide early warning of sudden resistance changes and detect abnormal conditions such as broken rods and foreign object jamming, but also achieve gradual resistance compensation to adapt to slow-variable disturbances such as die wear and temperature drift. Furthermore, it adjusts hydraulic flow in advance based on the rate of change of resistance to suppress oscillations, providing core algorithmic support for high-precision intelligent control of extruder hydraulic flow.

[0086] S3, obtain real-time workload and historical traffic data, and obtain a load-traffic correlation function based on the real-time workload and historical traffic data to quantify the matching degree between the current load state and the optimal solution of historical traffic.

[0087] Specifically, real-time workload is generally equivalent to the hydraulic thrust of an extruder. Historical flow data includes, but is not limited to, historical flow samples and historical optimal flow rates. Historical flow samples can be extracted from the process database or server in the factory workshop. The historical optimal flow rate can be understood as the quality target value under the same operating conditions, which can be set based on experience or obtained based on multi-objective optimization methods (such as the Pareto solution of quality + energy consumption + stability).

[0088] As a preferred technical solution, as shown in Figure 4, the specific method for obtaining the load flow correlation function includes:

[0089] S31, obtain the load normalization term based on the real-time workload L and the system's maximum design load.

[0090] Here, the load normalization term is used to convert the load of extruders of different tonnages into a dimensionless parameter to dynamically adjust the flow rate reference. Assuming a linear relationship between the hydraulic flow rate and pressure characteristics of the extruder's hydraulic system, to simplify calculations, the load normalization term can be exemplarily expressed as: real-time working load / system maximum design load. The system maximum design load can be obtained from the extruder's nameplate parameters.

[0091] In the actual operation of an extruder, the hydraulic flow and pressure characteristics of the extruder's hydraulic system often exhibit a non-linear saturation relationship. When the real-time working load exceeds a certain proportion of the system's maximum design load, the extrusion pressure at the same flow rate rises sharply. Therefore, dynamic compensation is required for the result of real-time working load / system maximum design load. Specifically, when the actual working load is less than or equal to a certain multiple of the system's maximum design load, such as when the actual working load is ≤0.6 times or ≤0.8 times the system's maximum design load, the load normalization term is expressed as: real-time working load / system maximum design load. Otherwise, it is first calculated according to the formula... Obtain the compensation coefficient, where mx represents the ratio of the real-time working load to the maximum design load of the system corresponding to the inflection point of the hydraulic flow-pressure characteristic curve, which is generally taken as 0.6 or 0.8, and η represents the compensation coefficient adjustment factor, which is generally taken as 1.5. After obtaining the compensation coefficient, obtain the load normalization term based on the compensation coefficient. The final result can be expressed as the load normalization term: compensation coefficient × real-time working load / maximum design load of the system.

[0092] S32, Obtain historical traffic samples V based on historical traffic data. i And the historical best traffic V, obtain the Gaussian similarity kernel function based on the historical traffic sample and the historical best traffic, and obtain the weighted average of the Gaussian similarity kernel function.

[0093] For example, the Gaussian similarity kernel function is represented as: Here, exp() represents an exponential function with the natural constant e as the base, and δ represents the similarity bandwidth, i.e., the allowable fluctuation range. This Gaussian similarity kernel function can be understood as a normal distribution model constructed with the historical optimal flow as the expected value and the similarity bandwidth δ as the standard deviation. The similarity bandwidth can be determined based on the formula... The calculation yields the following result: k' and N' represent the die wear coefficient and the number of extrusions, respectively. For new dies, based on experience, the die wear coefficient can be set to 0 or 0.1, etc. For older dies, such as those used for one or two years, the die wear coefficient can be set to 0.8 or 0.9, etc. Specifically, the die wear coefficient = a × (number of used dies / number of designed dies) + b. Here, a and b are calibration parameters, which can be determined using the least squares method.

[0094] S33 uses the product of the load normalization term and the weighted average as the load-flow correlation function to quantify the matching degree between the current load state and the historical optimal flow solution.

[0095] For example, the load-flow correlation function F” = load normalization term × weighted average. This load-flow correlation function, by quantifying the matching degree between the current load state and the historical optimal flow solution, can achieve: 1. Dynamically adjusting the flow benchmark to adapt to the nonlinear characteristics of the hydraulic system; 2. Transforming historical high-quality process parameters into real-time control rules; 3. Improving robustness by filtering noisy data through a Gaussian kernel function. In summary, this load-flow correlation function solves the problems of strong experience dependence and weak adaptability in traditional extrusion press hydraulic control systems.

[0096] S4, combine the first coupling function, the second coupling function, and the load-flow correlation function to obtain the flow correction factor.

[0097] The specific method for obtaining the traffic correction factor includes: obtaining the traffic correction factor based on the weighted values ​​of the first coupling function, the second coupling function, and the load-flow correlation function. For example, the traffic correction factor κ can be expressed as κ=α·F+β·F'+γ·F". Here, α, β, and γ represent the weight coefficients of the first coupling function, the second coupling function, and the load-flow correlation function, respectively. Generally, α, β, and γ can be set to 0.5, 0.3, and 0.2, respectively.

[0098] Preferably, when the temperature rises by 20 degrees Celsius within a preset time period, such as 60 seconds, the weighting coefficient of the first coupling function can be increased from 0.5 to 0.65 to enhance temperature compensation; when the rate of change of real-time extrusion resistance is greater than 10% / s, the weighting coefficient of the second coupling function can be increased from 0.3 to 0.45 to preferentially suppress abnormal extrusion resistance.

[0099] This flow correction factor integrates multi-dimensional parameters including temperature, material properties, real-time workload, real-time extrusion resistance, historical extrusion resistance, and historical flow data. It can achieve multi-physics coupling perception, provide advanced correction signals for the target flow control of the extruder, and overcome the lag problem of the hydraulic system of the extruder.

[0100] S5. Obtain the reference flow rate based on the extrusion ratio, correct the reference flow rate according to the flow rate correction factor, obtain the target flow rate, and control the hydraulic flow rate of the extruder according to the target flow rate. Generally, the specific method for obtaining the reference flow rate includes obtaining the reference flow rate based on the product of the extrusion ratio, the reference speed of the extrusion rod, and the cross-sectional area of ​​the billet. That is, the reference flow rate Q = extrusion ratio × billet cross-sectional area × extrusion rod reference speed. For example, the target flow rate = reference flow rate × (1 + flow rate correction factor).

[0101] As a preferred technical solution, the reference flow rate can be corrected based on the PID control principle to obtain the target flow rate. The target flow rate Q' can be expressed as... Among them, K p T d These represent the proportional gain and integral gain, respectively. The proportional gain is typically between 0.1 and 0.8, with a default value of 0.25. The integral gain is used for suppression. The hydraulic flow oscillation caused by sudden changes is generally between 0.02 and 0.05, with a default value of 0.03.

[0102] In the target flow function In this context, term 1 is used to maintain the basic stability of the system and prevent the additive structure in the target flow function from failing in the low flow region; term κ·K p Used to implement feedback compensation for process status; Item This is used to achieve perturbation feedforward suppression, i.e., suppression. Hydraulic flow oscillations caused by sudden changes. That is to say, the target flow function has a three-term coupling mechanism, which can achieve: 1. Process state adaptation: real-time correction of hydraulic flow based on the flow correction factor; 2. Sudden disturbance suppression: utilizing differential terms... Predicting system instability trends and suppressing them Fluctuations in hydraulic flow caused by sudden changes.

[0103] Traditional PID control relies on error signals. The target flow function of this invention deeply couples the flow correction factor with the proportional gain and integral gain. It integrates multi-dimensional parameters such as material characteristics and historical flow data, which can improve the intelligence and accuracy of hydraulic flow control in extruders.

[0104] In summary, the intelligent hydraulic flow control method for extruders obtains a flow correction factor by combining a first coupling function, a second coupling function, and a load flow correlation function, and then corrects the reference flow based on the flow correction factor to obtain the target force. It considers and couples multiple process parameters, can adapt to the complex and ever-changing working conditions during profile extrusion, and is conducive to improving the accuracy of hydraulic control.

[0105] As a preferred technical solution, the control method further includes the following steps: obtaining the second derivative of the real-time extrusion resistance; if the second derivative is greater than a preset oscillation trigger threshold, the system can be determined to have entered a high-risk oscillation state. At this time, the target flow rate is compensated and corrected based on the first derivative of the real-time extrusion resistance (i.e., the rate of change of the real-time extrusion resistance) to obtain the final flow rate, and the hydraulic flow rate of the extruder is controlled based on the final flow rate. For example, if the second derivative of the real-time extrusion resistance... The final flow rate Q is represented as The oscillation trigger threshold ζ can be understood as the critical value of drag acceleration, reflecting the system's inertial disturbance rejection capability. For example, for 7075 aluminum alloy, the corresponding oscillation trigger threshold can be set to 8.6 kN / s. 2 μ is the adaptive attenuation coefficient, typically between 0.06 and 0.12, with a default value of 0.08. For example, for an 800T thin-walled profile extrusion press, the adaptive attenuation coefficient can be set to 0.06; for a 2500T automotive structural component extrusion press, it can be set to 0.10. `sign()` represents the sign function. When the rate of change of real-time extrusion resistance is greater than 0, the final flow rate decreases to suppress the upward trend of real-time extrusion resistance; when the rate of change of real-time extrusion resistance is less than 0, the final flow rate increases to prevent a precipitous drop in real-time extrusion resistance. Here, based on the second derivative of the real-time extrusion resistance, the initial oscillation disturbance can be captured, enabling advanced prediction of hydraulic flow control; the sign function can identify the oscillation phase, achieving precise control of the hydraulic flow compensation direction.

[0106] Preferably, the control method described in this embodiment further includes: based on the measured oscillation frequency f of the extruder. osc The adaptive attenuation coefficient is dynamically adjusted based on the system's natural frequency f0 to dynamically match the oscillation intensity, and the attenuation strength increases with the measured oscillation frequency. Here, the system's natural frequency refers to the undamped natural frequency of the hydraulic actuator (cylinder + piston + mold), and the measured oscillation frequency refers to the actual disturbance frequency derived from the extrusion resistance signal. For example, the adaptive attenuation coefficient... Wherein, 0.4 represents the frequency response gain, an empirical value that can be adjusted according to actual conditions. μ0 represents the basic attenuation coefficient, which can be understood as the minimum effective attenuation intensity of the system at the reference frequency. For example, for an 800T thin-walled profile extruder, the basic attenuation coefficient can be set to 0.06, and for a 2500T automotive structural component extruder, the basic attenuation coefficient can be set to 0.10. The adaptive attenuation coefficient function increases the attenuation intensity as the measured oscillation frequency increases, avoiding overcompensation (increased energy consumption) or undercompensation (continued oscillation) caused by traditional fixed gain, thus extending the life of the extruder's hydraulic system and optimizing energy consumption.

[0107] Figure 9 shows a flowchart of the oscillation detection and triggering based on the second derivative of real-time extrusion resistance in this embodiment. Specifically, the real-time extrusion resistance is collected, and it is determined whether the second derivative of the real-time extrusion resistance is greater than the oscillation trigger threshold. If so, oscillation suppression is initiated, and the target flow rate is compensated and corrected to obtain the final flow rate; otherwise, the target flow rate output is maintained.

[0108] Preferably, the oscillation trigger threshold can be determined according to the formula... Among them, T op This indicates the continuous operating time, which is the continuous working time of the hydraulic system since the last maintenance, in hours. If the hydraulic system is forcibly cooled every 8 hours, the continuous operating time is reset to 0. This can be understood as the time decay factor, which to some extent can be explained by the stiffness degradation effect of the hydraulic system over time. When the hydraulic system is first started, the continuous operating time is 0, the time decay factor is 1, and the stiffness of the hydraulic system can be considered as 100%. As the continuous operating time increases, the time decay factor decreases rapidly. This can be understood as a material hardness correction ratio, which is used to quantify the stiffness difference between the current material and the equipment design benchmark. When the material hardness correction ratio is greater than 1, it means that the current material is harder than the equipment design benchmark, and the oscillation trigger threshold needs to be increased to allow for greater oscillation disturbances. When the material hardness correction ratio is less than 1, it means that the current material is softer than the equipment design benchmark, and the high oscillation trigger threshold needs to be decreased to perform more sensitive detection and improve system sensitivity.

[0109] In the oscillation trigger threshold formula, λ' is the material coefficient, which is calculated based on the contribution weight of material stiffness to the system's inertia. Generally, the material coefficient = (measured critical acceleration / theoretical maximum acceleration) × safety margin. For 6061 aluminum alloy or pure aluminum 1060, the material coefficient can be set to 0.15.

[0110] In summary, this oscillation trigger threshold, through the dual coupling of the time-varying effect of material stiffness and the cumulative effect of equipment fatigue, can reduce the failure rate and extend the extruder life.

[0111] As shown in Figure 5, an embodiment of the present invention also provides an intelligent control system for hydraulic flow of an extruder, used to implement the control method described above, which includes a first coupling function acquisition module, a second coupling function acquisition module, a load flow correlation function acquisition module, a flow correction factor acquisition module, and a hydraulic flow control module.

[0112] The first coupling function acquisition module is used to acquire the material properties and real-time temperature of the target to be extruded, and to acquire a first coupling function to characterize the coupling effect between material properties and temperature based on the material properties and real-time temperature. Specifically, as shown in Figure 6, the first coupling function acquisition module includes an exponential term acquisition unit, a correction term acquisition unit, and a first coupling function acquisition unit.

[0113] The exponent term acquisition unit is used to acquire an exponent term representing the thermal activation process based on the Arrhenius equation, real-time temperature, and deformation activation energy; the correction term acquisition unit is used to acquire a correction term representing the normalization correction of the material reference Young's modulus and hardness deviation by referencing Young's modulus; the first coupling function acquisition unit is used to use the product of the exponent term and the correction term as the first coupling function characterizing the coupling effect between material properties and temperature; wherein, the material properties include deformation activation energy, material reference Young's modulus, and hardness deviation.

[0114] This first coupling function breaks through the limitations of traditional single empirical formulas by separating the thermodynamic process (exponential term) and microstructure evolution (linear term) into separate models. It quantifies the synergistic effect of the material properties of the target material to be extruded and the real-time temperature on the extrusion flow rate, and solves the defect of independent processing of the two in the hydraulic flow control of traditional extruders.

[0115] The second coupling function acquisition module is used to acquire real-time extrusion resistance and historical extrusion resistance, and to acquire a second coupling function based on the real-time extrusion resistance and historical extrusion resistance to characterize the degree of resistance deviation and its changing trend. Specifically, as shown in Figure 7, the second coupling function acquisition module includes a hyperbolic tangent term acquisition unit and a second coupling function acquisition unit.

[0116] The hyperbolic tangent term acquisition unit is used to obtain the historical average resistance based on the historical extrusion resistance, and uses the ratio of the real-time extrusion resistance to the historical average resistance as input to obtain the hyperbolic tangent term; the second coupling function acquisition unit is used to obtain the rate of change term based on the rate of change of the real-time extrusion resistance, and uses the sum of the hyperbolic tangent term and the rate of change term as the second coupling function used to characterize the degree of resistance deviation and the trend of change.

[0117] This second coupling function dynamically quantifies the deviation between real-time extrusion resistance and historical average resistance, and captures the trend of resistance change. It combines resistance state normalization and forward-looking control of the rate of change, realizing a leap from passive response to active defense in the extrusion process. It can not only provide early warning of sudden resistance changes and detect abnormal working conditions such as broken bars and foreign object jamming, but also achieve gradual resistance compensation to adapt to slow variable disturbances such as die wear and temperature drift. Furthermore, it can adjust the hydraulic flow in advance based on the rate of change of resistance to suppress oscillations, providing core algorithm support for high-precision intelligent control of hydraulic flow in extruders.

[0118] The load-flow correlation function acquisition module is used to acquire real-time workload and historical traffic data, and to acquire a load-flow correlation function based on the real-time workload and historical traffic data to quantify the matching degree between the current load state and the optimal historical traffic solution. Specifically, as shown in Figure 8, the load-flow correlation function acquisition module includes a load normalization term acquisition unit, a weighted average value acquisition unit, and a load-flow correlation function acquisition unit.

[0119] The load normalization term acquisition unit is used to obtain the load normalization term based on the real-time workload and the system's maximum design load; the weighted average value acquisition unit is used to obtain historical traffic samples and historical optimal traffic based on historical traffic data, obtain the Gaussian similarity kernel function based on the historical traffic samples and historical optimal traffic, and obtain the weighted average value of the Gaussian similarity kernel function; the load traffic correlation function acquisition unit is used to use the product of the load normalization term and the weighted average value as the load traffic correlation function used to quantify the matching degree between the current load state and the historical optimal traffic solution.

[0120] This load-flow correlation function solves the problems of strong experience dependence and weak adaptability in traditional hydraulic control systems by quantifying the matching degree between the current load state and the historical optimal flow solution.

[0121] The traffic correction factor acquisition module is used to obtain the traffic correction factor by combining the first coupling function, the second coupling function, and the load-traffic correlation function. Specifically, the traffic correction factor acquisition module obtains the traffic correction factor based on the weighted value of the first coupling function, the second coupling function, and the load-traffic correlation function.

[0122] The hydraulic flow control module is used to obtain a reference flow rate based on the extrusion ratio, correct the reference flow rate according to a flow correction factor, obtain a target flow rate, and control the hydraulic flow rate of the extruder according to the target flow rate. For example, the reference flow rate Q = extrusion ratio × billet cross-sectional area × extrusion bar reference speed. For example, the target flow rate = reference flow rate × (1 + flow correction factor).

[0123] In summary, the intelligent hydraulic flow control system for the extruder obtains the flow correction factor by combining the first coupling function, the second coupling function, and the load flow correlation function, and then corrects the reference flow based on the flow correction factor to obtain the target force. It considers and couples multiple process parameters, can adapt to the complex and ever-changing working conditions during profile extrusion, and is conducive to improving the accuracy of hydraulic control.

[0124] The technical features of the embodiments described can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for intelligent control of hydraulic flow in an extruder, characterized in that, The control method includes: acquiring the material properties and real-time temperature of the target to be extruded; acquiring a first coupling function to characterize the coupling effect between material properties and temperature based on the material properties and real-time temperature; acquiring real-time extrusion resistance and historical extrusion resistance; acquiring a second coupling function to characterize the degree of resistance deviation and its changing trend based on the real-time extrusion resistance and historical extrusion resistance; acquiring real-time workload and historical flow data; acquiring a load-flow correlation function to quantify the matching degree between the current load state and the optimal historical flow solution based on the real-time workload and historical flow data; acquiring a flow correction factor by combining the first coupling function, the second coupling function, and the load-flow correlation function; acquiring a reference flow rate based on the extrusion ratio; correcting the reference flow rate based on the flow correction factor to acquire the target flow rate; and controlling the hydraulic flow rate of the extruder based on the target flow rate. The specific method for obtaining the first coupling function includes: based on the Arrhenius equation, obtaining an exponential term representing the thermal activation process according to the real-time temperature and deformation activation energy; obtaining a correction term representing the normalization correction of the material reference Young's modulus and hardness deviation by referencing Young's modulus; and using the product of the exponential term and the correction term as the first coupling function characterizing the coupling effect between material properties and temperature; wherein, material properties include deformation activation energy, material reference Young's modulus, and hardness deviation. The specific method for obtaining the second coupling function includes: obtaining the historical average resistance based on the historical extrusion resistance, obtaining a hyperbolic tangent term using the ratio of the real-time extrusion resistance to the historical average resistance as input; obtaining a rate of change term based on the rate of change of the real-time extrusion resistance, and using the sum of the hyperbolic tangent term and the rate of change term as the second coupling function characterizing the degree and trend of resistance deviation.

2. The intelligent control method for hydraulic flow of an extruder as described in claim 1, characterized in that, The specific methods for obtaining the load-flow correlation function include: obtaining a load normalization term based on the real-time workload and the system's maximum design load; obtaining historical traffic samples and historical optimal traffic based on historical traffic data; obtaining a Gaussian similarity kernel function based on the historical traffic samples and historical optimal traffic; and obtaining a weighted average of the Gaussian similarity kernel function; using the product of the load normalization term and the weighted average as the load-flow correlation function to quantify the matching degree between the current load state and the historical optimal traffic solution.

3. The intelligent control method for hydraulic flow of an extruder as described in claim 1, characterized in that, The specific methods for obtaining the traffic correction factor include: obtaining the traffic correction factor based on the weighted value of the first coupling function, the second coupling function, and the load-traffic correlation function.

4. The intelligent control method for hydraulic flow of an extruder as described in claim 1, characterized in that, Specific methods for obtaining the reference flow rate include: obtaining the reference flow rate based on the product of the extrusion ratio, the reference speed of the extrusion rod, and the cross-sectional area of ​​the billet.

5. An intelligent control system for hydraulic flow of an extruder, used to implement the control method as described in any one of claims 1-4, characterized in that, The control system includes: a first coupling function acquisition module, used to acquire the material properties and real-time temperature of the target to be extruded, and acquire a first coupling function to characterize the coupling effect between material properties and temperature based on the material properties and real-time temperature; a second coupling function acquisition module, used to acquire real-time extrusion resistance and historical extrusion resistance, and acquire a second coupling function to characterize the degree of resistance deviation and its changing trend based on the real-time extrusion resistance and historical extrusion resistance; a load-flow correlation function acquisition module, used to acquire real-time working load and historical flow data, and acquire a load-flow correlation function to quantify the matching degree between the current load state and the optimal solution of historical flow based on the real-time working load and historical flow data; a flow correction factor acquisition module, used to acquire a flow correction factor by combining the first coupling function, the second coupling function, and the load-flow correlation function; and a hydraulic flow control module, used to acquire a reference flow rate based on the extrusion ratio, correct the reference flow rate based on the flow correction factor, acquire a target flow rate, and control the hydraulic flow rate of the extruder based on the target flow rate.

6. The intelligent control system for hydraulic flow of an extruder as described in claim 5, characterized in that, The first coupling function acquisition module includes: an exponent term acquisition unit, used to acquire an exponent term representing the thermal activation process based on the Arrhenius equation, real-time temperature, and deformation activation energy; a correction term acquisition unit, used to acquire a correction term representing the normalization correction of the material reference Young's modulus and hardness deviation by referencing Young's modulus; and a first coupling function acquisition unit, used to use the product of the exponent term and the correction term as the first coupling function characterizing the coupling effect between material properties and temperature; wherein, the material properties include deformation activation energy, material reference Young's modulus, and hardness deviation.

7. The intelligent control system for hydraulic flow of an extruder as described in claim 5, characterized in that, The second coupling function acquisition module includes: a hyperbolic tangent term acquisition unit, used to acquire historical average resistance based on historical extrusion resistance, and to acquire the hyperbolic tangent term using the ratio of real-time extrusion resistance to historical average resistance as input; and a second coupling function acquisition unit, used to acquire the rate of change term based on the rate of change of real-time extrusion resistance, and to use the sum of the hyperbolic tangent term and the rate of change term as the second coupling function used to characterize the degree of resistance deviation and the trend of change.

8. The intelligent control system for hydraulic flow of an extruder as described in claim 5, characterized in that, The load-flow correlation function acquisition module includes: a load normalization term acquisition unit, used to acquire a load normalization term based on real-time workload and the system's maximum design load; a weighted average value acquisition unit, used to acquire historical traffic samples and historical optimal traffic based on historical traffic data, acquire a Gaussian similarity kernel function based on the historical traffic samples and historical optimal traffic, and acquire a weighted average value of the Gaussian similarity kernel function; and a load-flow correlation function acquisition unit, used to use the product of the load normalization term and the weighted average value as the load-flow correlation function to quantify the matching degree between the current load state and the historical optimal traffic solution.

Images