Model-free adaptive iterative learning injection molding speed control method with attenuation factor

Abstract

The invention provides a model-free adaptive iterative learning injection molding speed control method with an attenuation factor. The method comprises the following steps: modeling an injection process as an actual output model containing measurement disturbance; constructing a control law with an attenuation factor, calculating control input, sending the control input to the injection molding system, and collecting measurement output after operation; if the sampling time is less than the maximum sampling time, updating the measurement output, and recalculating the control input; if the measurement error is greater than the error threshold value, entering the next iteration to recalculate the control input; and otherwise, taking the current control input as the control input of learning completion. The method does not need to depend on an accurate mathematical model in the injection molding process, and complex nonlinear characteristics in the injection molding process can be dealt with only through input, output and iterative learning; in addition, an attenuation factor mechanism is introduced, and the influence of measurement disturbance on control input is gradually weakened, so that the stability of system output is maintained.

Description

Technical Field

[0001] This application relates to the field of injection molding technology, specifically to a model-free adaptive iterative learning injection speed control method with a decay factor. Background Technology

[0002] Injection molding, an important plastic processing technique, involves melting solid plastic into a molten, flowing state, injecting it into a mold, and allowing it to cool and solidify to obtain plastic products with specific shapes and performance requirements. As a typical batch production process, injection molding consists of multiple stages, each completed within a limited time, resulting in the cyclical batch production of plastic products. The entire injection molding process involves the precise control of several parameters, such as barrel temperature, injection speed, nozzle pressure, and hydraulic pressure. Among these, injection speed is particularly critical, directly affecting the quality of the plastic products and production efficiency. Therefore, precise control of the injection speed is essential for improving product quality and production efficiency.

[0003] The injection molding process involves complex physical and chemical reaction mechanisms, exhibiting nonlinear characteristics and variable environmental factors, leading to uncertainties and significant challenges in controlling injection speed. In actual injection molding production, traditional PID (Proportional-Integral-Derivative) control, due to its simple structure and ease of engineering implementation, has been widely used in injection speed control. However, this traditional PID method has the following significant drawbacks:

[0004] 1) Traditional PID control is ineffective when dealing with the complex nonlinear characteristics of the injection molding process: For example, the viscosity of the melt changes nonlinearly with factors such as temperature and pressure during injection molding. PID control relies on relatively fixed control parameters and simple linear assumptions, making it difficult to accurately adapt to such complex nonlinear dynamic changes. It often fails to establish a control model that matches the actual process, resulting in low control accuracy and difficulty in meeting the quality requirements of high-precision injection molded products. 2) Traditional PID control has weak anti-interference ability when dealing with external environmental disturbances during the injection molding process: For example, there are often external disturbances such as fluctuations in workshop temperature, unstable power grid voltage, and differences in the characteristics of raw material batches during the injection molding process. The parameters of PID control are mostly based on static operating conditions, making it difficult to quickly offset the effects of these disturbances. This causes the system output to deviate from the expected injection speed trajectory, thus affecting the consistency and pass rate of injection molded products.

[0005] Therefore, how to design a model-free adaptive iterative learning injection speed control method, system, equipment, and storage medium with a decay factor that can adapt to the complex nonlinear characteristics of the injection molding process and resist external environmental disturbances is a technical problem that has not yet been solved in the existing technology. Summary of the Invention

[0006] Therefore, the technical problem to be solved by this application is to overcome the technical defects of traditional PID control in the prior art in the face of complex nonlinear characteristics and external environmental disturbances in the injection molding process, so as to provide a model-free adaptive iterative learning injection speed control method, system, device and storage medium with attenuation factor that can adapt to the complex nonlinear characteristics in the injection molding process and resist external environmental disturbances.

[0007] This application mainly includes the following aspects: In a first aspect, embodiments of this application provide a model-free adaptive iterative learning injection molding speed control method with a decay factor, the method comprising: Step 1: Obtain the maximum sampling time in a single iteration initial value of pseudo-partial derivative Initial control inputs for system operation and initial system output data ; in, To control the input, indicating the first Next iteration, The flow rate of the hydraulic system during the injection process; For system output, indicating the first... Next iteration, The injection speed during the injection process; Sampling time, ; For the number of iterations, ; Step 2: Based on the input and output data, model the injection process as a nonlinear system model with iterative characteristics; Step 3: Introduce pseudo-partial derivatives The nonlinear system model with iterative characteristics described in step 2 is linearized on the iteration axis to obtain a compact dynamic linearized model. Step 4: Based on the compact dynamic linearization model described in Step 3, transform the nonlinear system model with iterative characteristics described in Step 2 into a compact dynamic linearization output model; Step 5: Introduce measurement disturbance The compact-form dynamic linearized output model described in step 4 is then extended to an actual output model that includes measurement perturbations: ; in, For the first Next iteration, At any given time, the measurement output includes the measurement disturbance; the measurement disturbance It is a random signal with the following statistical properties: , ,in For mathematical expectation factor, The variance factor is the measurement disturbance. With the sampling time The number of iterations The measurement output The system output The control input and the pseudo-partial derivatives All are uncorrelated, measurement disturbance initial disturbance value =0; Step 6: Based on the expected output and the measurement output Calculate the measurement error ; Step 7: Construct a pseudo-partial derivative estimation algorithm and calculate the pseudo-partial derivative estimates. ; Step 8: Based on the measurement error described in Step 6 and the pseudo-partial derivative estimate in step 7 Construct a control law with a decay factor and calculate the control input. The control input The measured output is sent to the injection molding system and collected after the operation. ; The control law with the attenuation factor is as follows: ; in, Step size factor This is used to make the control law with the attenuation factor more general; As a weighting factor; This is a decay factor used to gradually reduce the measurement error as the number of iterations increases. Weighting percentage; Step 9: For the sampling time With the maximum sampling time of the single iteration If a comparison is made, Then return to step 6 and output the measurement from step 6. Updated to the measurement output acquired in step 8 after the run. Otherwise, proceed to step 10. Step 10: Collect the measurement errors at all times within this iteration, and calculate the absolute value of the maximum measurement error. With error threshold If a comparison is made, If the condition is met, return to step 6; otherwise, return to the final control input. The data is sent to the injection molding system, ending the iterative learning process.

[0008] According to one embodiment of the present invention, the nonlinear system model with iterative characteristics modeled in step 2 includes: ; in, To output the order of the variable, The order of the input variable.

[0009] According to one embodiment of the present invention, the compact scheme dynamic linearization model established in step 3 includes: ; ; ; in, and These represent the changes in control input and system output respectively during adjacent batches of system operation; for any sampling time... and the number of iterations All satisfy ,in, It is a constant; all pseudo-partial derivatives All symbols remain consistent, that is ,in, It is a constant.

[0010] According to one embodiment of the present invention, the compact-format dynamic linearization output model derived in step 4 includes: .

[0011] According to one embodiment of the present invention, the measurement error is calculated in step 6. The methods include: ; in, The desired output represents the known optimal injection speed; For the first Next iteration, The measured output at time [time].

[0012] According to one embodiment of the present invention, in step 7, a pseudo-partial derivative estimation algorithm is constructed, and pseudo-partial derivative estimates are calculated. The steps include: A pseudo-partial derivative estimation function is established, and a quantitative index for the estimation error is determined; the pseudo-partial derivative estimation function includes: ;in, The objective function is... These are the weighting coefficients; This is an estimate of the pseudo-partial derivative; To measure the change in output, the calculation formula is: ; To control the amount of input variation, the calculation formula is as follows: ; Based on the pseudo-partial derivative estimation function and optimal conditions The pseudo-partial derivative estimation algorithm is obtained, and the pseudo-partial derivative estimates are calculated. The pseudo-partial derivative estimation algorithm includes: ;in, Step size factor This is used to easily adjust the pseudo-partial derivatives and enhance the flexibility of the algorithm; These are the weighting coefficients; A reset algorithm is established to ensure that the pseudo-partial derivative estimation algorithm has good tracking performance; the reset algorithm includes: ; like or or ; in, The initial value of the pseudo-partial derivative obtained in step 1 ; To reset the threshold; It is a symbolic function.

[0013] According to one embodiment of the present invention, in step 8, the measurement error is based on that in step 6. and the pseudo-partial derivative estimate in step 7 The steps to construct a control law with a decay factor include: Establish the objective function for controlling the input: ; in, The objective function is... As a weighting factor; Based on the control input objective function and optimal conditions Thus, the control law is obtained: ; in, Step size factor This is used to make the control law more general; As a weighting factor; Based on the control law and the pseudo-partial derivative estimate in step 7. and attenuation factor Establish the control law with the attenuation factor to calculate the control input. .

[0014] Secondly, embodiments of this application also provide a model-free adaptive iterative learning injection molding speed control system with a decay factor, the system comprising: Acquisition module: Used to obtain the maximum sampling time in a single iteration. initial value of pseudo-partial derivative Initial control input Initial system output data Expected output and measurement output after operation ; Storage module: Used to store relevant data; Calculation module: used to calculate measurement error Pseudo-partial derivative estimates Control input ; Comparison module: used for sampling time With the maximum sampling time of a single iteration Compare, and the absolute value of the maximum measurement error. With error threshold Compare; Output module: Used to output the final control input. Send to the injection molding system.

[0015] Thirdly, embodiments of this application also provide a computer device, the computer device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the steps of the method described above when executing the instructions.

[0016] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described above.

[0017] The technical solution provided in this application has the following advantages: 1. The model-free adaptive iterative learning control strategy adopted in this application does not rely on a precise mathematical model of the injection molding process. It can dynamically optimize the control strategy and adapt to the complex nonlinear characteristics of the injection molding process by relying only on real-time system input and output data and iterative learning mechanism, thereby accurately controlling the injection speed.

[0018] 2. This application incorporates the measured disturbance into the model, enabling the model to more realistically reflect the injection molding system. This ensures that the control law designed based on the model can specifically address the impact of disturbances, guaranteeing the effectiveness of the control strategy in actual working conditions and avoiding error expansion or system instability caused by not considering disturbances.

[0019] 3. This application introduces an attenuation factor mechanism, which can gradually reduce the weight ratio of measurement error during the iterative update of the control algorithm, thereby gradually weakening the impact of measurement disturbance on the control input, reducing the cumulative effect of noise signal in the iteration process, thus maintaining the stability of the system output and improving the control strategy's ability to suppress disturbances.

[0020] 4. This application introduces an attenuation factor to moderately attenuate the information affected by disturbances in past iterations, reduce the interference of invalid data on the current control decision, make the iterative learning process more focused on effective information, and gradually optimize the control quantity, thereby improving the dynamic response characteristics of the system to a certain extent and enhancing the consistency of control performance.

[0021] 5. This application can effectively suppress the adverse effects of measurement disturbances on injection speed, effectively reduce tracking deviation, and enable the injection speed to closely and stably follow the preset trajectory, thereby ensuring that the injection molding process can maintain a high-precision operating state under complex interference, providing a strong guarantee for the high quality and stability of injection molding production.

[0022] 6. The model-free adaptive iterative learning control strategy adopted in this application does not rely on the precise mathematical model of a specific system. Therefore, it is not only applicable to the injection speed control of the injection molding process, but can also be extended to other dynamic systems with nonlinear and repetitive characteristics, and has good versatility.

[0023] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is the overall flowchart of the model-free adaptive iterative learning injection speed control method with decay factor in this application.

[0026] Figure 2 This is the iterative operation framework diagram of this application.

[0027] Figure 3 This is the system architecture diagram of this application.

[0028] Figure 4 This is a schematic diagram of the basic structure of an existing hydraulic screw injection molding machine.

[0029] Figure 5 This is a schematic diagram of the existing optimal injection speed curve.

[0030] Figure 6 This is a schematic diagram comparing the maximum tracking error curves of this application and a general model-free adaptive iterative learning control scheme in the iterative domain.

[0031] Figure 7 This is a comparative schematic diagram of the maximum tracking error curves in the iterative domain under different measurement noise variance conditions.

[0032] Figure 8 This is a schematic diagram of injection speed tracking curves under different control schemes.

[0033] Figure 9 This is a schematic diagram illustrating the injection speed tracking error of this application and the PID control scheme. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0035] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0036] The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0037] Example

[0038] like Figure 1 and Figure 2 As shown, this embodiment provides a model-free adaptive iterative learning injection molding speed control method with a decay factor, including the following steps: Step 1: Obtain the maximum sampling time in a single iteration initial value of pseudo-partial derivative Initial control inputs for system operation and initial system output data ; in, To control the input, indicating the first Next iteration, The flow rate of the hydraulic system during the injection process; For system output, indicating the first... Next iteration, The injection speed during the injection process; Sampling time, ; For the number of iterations, ; Step 2: Based on the input and output data, model the injection process as a nonlinear system model with iterative characteristics; Step 3: Introduce pseudo-partial derivatives The nonlinear system model with iterative characteristics from step 2 is linearized on the iteration axis to obtain a compact dynamic linearized model. Step 4: Based on the compact form dynamic linearization model in Step 3, transform the nonlinear system model with iterative characteristics in Step 2 into a compact form dynamic linearization output model. Step 5: Introduce measurement disturbance The compact-form dynamic linearized output model from step 4 is extended to a real output model containing measurement perturbations: ; in, For the first Next iteration, At any given time, the measurement output includes the measurement disturbance; the measurement disturbance It is a random signal with the following statistical properties: , ,in For mathematical expectation factor, The variance factor is the measurement disturbance. With sampling time Number of iterations Measurement output System output Control input and pseudo-partial derivatives All are uncorrelated, measurement disturbance initial disturbance value =0; Step 6: Based on the expected output and measurement output Calculate the measurement error ; Step 7: Construct a pseudo-partial derivative estimation algorithm and calculate the pseudo-partial derivative estimates. ; Step 8: Measurement error based on step 6 And the pseudo-partial derivative estimate from step 7 Construct a control law with a decay factor and calculate the control input. Control input Send to the injection molding system; The control law with the attenuation factor is: ; in, Step size factor This is used to make control laws with attenuation factors more general; As a weighting factor; This is the attenuation factor, used to gradually reduce the measurement error as the number of iterations increases. Weighting percentage; Step 9: Sampling time With the maximum sampling time of a single iteration If a comparison is made, Then return to step 6 and output the measurement from step 6. Updated to the measurement output acquired in step 8 after the run. Otherwise, proceed to step 10.

[0039] Step 10: Collect the measurement errors at all times within this iteration, and calculate the absolute value of the maximum measurement error. With error threshold If a comparison is made, If yes, return to step 6; otherwise, input the final control. The data is sent to the injection molding system, ending the iterative learning process.

[0040] The following details each step of the model-free adaptive iterative learning injection speed control method with a decay factor: Step 1: Obtain the maximum sampling time in a single iteration initial value of pseudo-partial derivative Initial control inputs for system operation and initial system output data ; in, To control the input, indicating the first Next iteration, The flow rate of the hydraulic system during the injection process; For system output, indicating the first... Next iteration, The injection speed during the injection process; Sampling time, ; For the number of iterations, ; Here, the maximum sampling time in a single iteration The initial value of the pseudo-partial derivative depends on the specific system conditions. The initial control input is 10 during the first iteration. The initial system output data is 0. It is 0.

[0041] It should be noted that, as Figure 4 As shown, the basic structure of a hydraulic screw injection molding machine includes a mold clamping unit, an injection unit, a power unit, and a control unit. The injection unit includes a plasticizing device, a screw drive device, an injection device, a measuring device, an injection seat, a frame, and a barrel feeding device. The injection device includes a hydraulic cylinder, a screw, and a nozzle. During injection, hydraulic oil is injected into the hydraulic cylinder, generating pressure to push the screw to pressurize the melt at the front of the barrel, causing the melt to enter the mold cavity through the nozzle.

[0042] Based on the structure of the hydraulic screw injection molding machine described above, the control input of this application embodiment is the flow rate of the hydraulic system during the injection process, specifically the flow rate of the hydraulic oil injected into the cylinder; correspondingly, the system output of this application embodiment is the injection speed during the injection process, specifically the speed at which the melt is injected into the mold cavity through the nozzle, which can be measured by a speed sensor.

[0043] Step 2: Based on the input and output data, model the injection process as a nonlinear system model with iterative characteristics; In one possible implementation, the nonlinear system model with iterative characteristics modeled in step 2 includes: ; in, To output the order of the variable, The order of the input variable.

[0044] It should be noted that at the start of the injection molding process, the mold is first closed, causing the moving mold and the fixed mold to close together. Simultaneously, the injection stage moves forward, bringing the nozzle at the front of the barrel into contact with the main sprue of the mold. Once the mold is closed and the nozzle is pressed against the main sprue, the injection process begins: the hydraulic system supplies hydraulic oil to the injection cylinder, pushing the screw forward. The molten polymer at the screw head, propelled by the screw, passes through the nozzle, runner, gate, and finally enters the mold cavity. After the injection process, the holding pressure stage begins: this stage is to prevent the molten material in the mold cavity from flowing back due to the adverse pressure gradient, and to compensate for the shrinkage caused by the cooling and solidification of the plastic material. During the injection process, the injection nozzle must maintain a certain pressure to propel the melt into the mold cavity. Once the melt in the mold cavity cools to the point where it cannot flow back, the cooling stage begins, allowing the plastic product to further cool and solidify within the mold cavity. Simultaneously, the system enters the plasticizing stage: the hydraulic motor drives the screw to rotate, and the material is conveyed forward along the screw groove under the action of friction. At the same time, the material begins to plasticize and melt under the action of the barrel heating coil, transforming into a viscous fluid stored in the storage chamber at the front of the screw, preparing for the next injection. After the plasticizing and cooling stages are completed, the mold is opened, the ejector device is activated to release the product, and the injection molding process ends.

[0045] The injection phase can be represented by a model constructed from the following set of equations: ; ; ; ; in, Let be the cross-sectional area of ​​the injection cylinder. Let be the cross-sectional area of ​​the barrel. The volume of the injection tube. The volume of material in the barrel. For injection speed, For the flow rate of the hydraulic system, The bulk modulus of the hydraulic fluid. This represents the bulk modulus of the material in the nozzle. For the injection cylinder pressure, For nozzle pressure, Where is the nozzle radius, For screw mass, Injection site, This represents the power-law exponent of the material's melting. This is the initial length of the screw. This represents the ratio of the screw's radius to the nozzle's radius. Shear rate The corresponding melt viscosity, This represents the average flow rate of the material.

[0046] Due to differences in working environments and measurement conditions, modeling in real-world environments is more complex, and constructing a mathematical model that accurately reflects the actual situation is more difficult. Therefore, the following nonlinear system model is constructed to characterize the above-mentioned injection stage: ; in, To output the order of the variable, The order of the input variable; It is a non-linear function.

[0047] Because the injection molding process is repetitive, an iteration index can be introduced. The number of iterations of the system is used to further model the above nonlinear system model into a nonlinear system model with iterative characteristics: .

[0048] The model-free adaptive iterative learning control strategy adopted in this application does not rely on a precise mathematical model of the injection molding process. It can dynamically optimize the control strategy and adapt to the complex nonlinear characteristics of the injection molding process by relying only on real-time system input and output data and iterative learning mechanism, thereby accurately controlling the injection speed.

[0049] Similarly, since the embodiments of this application do not rely on the precise mathematical model of a specific system, they are not only applicable to the injection speed control of the injection molding process, but can also be extended to other dynamic systems with nonlinear and repeatable characteristics, thus having good versatility.

[0050] Step 3: Introduce pseudo-partial derivatives The nonlinear system model with iterative characteristics from step 2 is linearized on the iteration axis to obtain a compact dynamic linearized model. In one possible implementation, the compact-format dynamic linearization model established in step 3 includes: ; ; ; in, and These represent the changes in control input and system output respectively during adjacent batches of system operation; for any sampling time... and number of iterations All satisfy ,in, It is a constant; all pseudo-partial derivatives All symbols remain consistent, that is ,in, It is a constant.

[0051] Here, the derivation logic of the compact-format dynamic linearization model established in step 3 is as follows:

[0052] Nonlinear system models with iterative characteristics satisfy the assumptions of the compact scheme dynamic linearization method: Assumption 1: Regarding control input The partial derivatives are continuous, that is... , It is a positive number.

[0053] Assumption 2: Nonlinear system models with iterative characteristics satisfy the generalized Lipschitz condition on the iteration axis, that is, for... , and have: ; in, ,and It is a constant; ; .

[0054] When a nonlinear system model with iterative characteristics satisfies Assumption 1 and Assumption 2, if Then there exists a pseudo-partial derivative (PPD) that makes the injection process system represent a compact-form dynamic linearized model: ; Step 4: Based on the compact form dynamic linearization model in Step 3, transform the nonlinear system model with iterative characteristics in Step 2 into a compact form dynamic linearization output model. In one possible implementation, the compact-form dynamic linearization model from step 3 can transform the nonlinear system model with iterative characteristics from step 2 into a compact-form dynamic linearization output model: .

[0055] Specifically: Substitute You can then obtain: .

[0056] Step 5: Introduce measurement disturbance The compact-form dynamic linearized output model from step 4 is extended to a real output model containing measurement perturbations: ; in, For the first Next iteration, At any given time, the measurement output includes the measurement disturbance; the measurement disturbance It is a random signal with the following statistical properties: , ,in For mathematical expectation factor, The variance factor is the measurement disturbance. With sampling time Number of iterations Measurement output System output Control input and pseudo-partial derivatives All are uncorrelated, measurement disturbance initial disturbance value =0; Here, the measurement disturbances are various external disturbances, such as the measurement error of the speed sensor (inherent sensor noise, detection deviation caused by installation gaps, and signal distortion caused by electromagnetic interference), fluctuations in workshop temperature, instability of power grid voltage, and differences in characteristics between batches of raw materials. These disturbances are not simulated but real and can affect the sensor measurement results. The embodiments of this application take into account the impact of these measurement disturbances on the measurement results, and therefore incorporate measurement disturbances into the compact form dynamic linearized output model. Thus, the actual output model containing measurement perturbations was obtained: .

[0057] This application incorporates measured disturbances into the model, enabling the model to more realistically reflect the injection molding system. This ensures that the control law designed based on the model can specifically address the impact of disturbances, guaranteeing the effectiveness of the control strategy in actual working conditions and avoiding error expansion or system instability caused by not considering disturbances.

[0058] Step 6: Based on the expected output and measurement output Calculate the measurement error ; In one possible implementation, the measurement error is calculated in step 6. The methods include: ; in, The desired output represents the known optimal injection speed; For the first Next iteration, The output of the time measurement.

[0059] It should be noted that the desired output of this embodiment is the optimal injection rate curve that dynamically changes over time (e.g., Figure 5As shown in the figure, the control objective of this embodiment is to accurately "track" the optimal injection speed curve, focusing on dynamically following the preset speed trajectory rather than simply adjusting the injection speed to a fixed value.

[0060] Here, the optimal injection speed curve is further explained: Based on general industry understanding and practical experience, when molten material can be injected into the mold cavity at a stable and uniform rate at a specific injection speed, the benefits are multi-dimensional and significant. On the one hand, it ensures that the injection-molded products fully meet the stringent standards set for injection molding production in terms of dimensional accuracy, color consistency, and shape integrity, greatly reducing the proportion of defective products. On the other hand, it can effectively promote the overall production efficiency of the injection molding process, maximizing resource utilization and capacity release. The injection speed corresponding to this is defined in the industry as the optimal injection speed, which is also equivalent to the desired injection speed.

[0061] In the injection molding process, the screw rotates and propels the melt continuously into the mold cavity, a process that can be divided into four stages. The first stage is the high-speed push phase, where the screw pushes the melt at a relatively high speed to reduce heat loss, maintain good melt flowability, minimize heat loss during transport, and ensure the melt maintains suitable temperature and rheological properties for smooth entry into the mold cavity. The second stage is the mold cavity deceleration phase, where the screw speed needs to be reduced in time when the melt front reaches the mold cavity opening to avoid splashing when the melt impacts the mold cavity opening at high speed, prevent material waste and uneven pressure inside the mold cavity, and ensure a smooth injection molding process. The third stage is the accelerated filling phase, where the screw needs to accelerate its push speed to achieve rapid and complete filling of the mold cavity and prevent the melt from cooling and affecting its flowability. This overcomes the problem of decreased flowability caused by melt cooling and ensures that the melt fills every corner of the mold cavity in a short time, ensuring complete product molding. The fourth stage is the secondary deceleration phase, where the screw needs to reduce its push speed again. If the injection speed is too fast, the melt injection volume will exceed the expectation, which may lead to product defects, such as flash affecting dimensional accuracy and appearance quality, and may also cause internal structural defects that reduce mechanical properties. Reducing the speed allows for precise control of the melt injection volume, ensuring that the product quality meets the standards.

[0062] Step 7: Construct a pseudo-partial derivative estimation algorithm and calculate the pseudo-partial derivative estimates. ; In one possible implementation, step 7 involves constructing a pseudo-partial derivative estimation algorithm and calculating the pseudo-partial derivative estimates. The steps include the following: S701: Establish pseudo-partial derivative estimation functions and determine the quantification index of estimation error; the pseudo-partial derivative estimation functions include: ;in, The objective function is... These are the weighting coefficients; This is an estimate of the pseudo-partial derivative; To measure the change in output, the calculation formula is: ; To control the amount of input variation, the calculation formula is as follows: ; S702: Estimation of Function Based on Pseudopartial Derivative and Optimal Conditions The pseudo-partial derivative estimation algorithm is obtained, and the pseudo-partial derivative estimates are calculated. Pseudo-partial derivative estimation algorithms include: ;in, Step size factor This is used to easily adjust the pseudo-partial derivatives and enhance the flexibility of the algorithm; These are the weighting coefficients; S703: Establish a reset algorithm to ensure good tracking performance of the pseudo-partial derivative estimation algorithm; the reset algorithm includes: ; like or or ; in, The initial value of the pseudo-partial derivative obtained in step 1 ; To reset the threshold, specifically: ; It is a symbolic function.

[0063] Step 8: Measurement error based on step 6 And the pseudo-partial derivative estimate from step 7 Construct a control law with a decay factor and calculate the control input. Control input Send to the injection molding system and collect the measurement output after operation. ; The control law with the attenuation factor is: ; in, Step size factor This is used to make control laws with attenuation factors more general; As a weighting factor; This is the attenuation factor, used to gradually reduce the measurement error as the number of iterations increases. Weighting percentage; In one possible implementation, step 8 is based on the measurement error from step 6. And the pseudo-partial derivative estimate from step 7 The steps to construct a control law with a decay factor include the following: S801: Establish the objective function for the control input: ; in, The objective function is... As a weighting factor; S802: Based on control input objective function and optimal conditions Thus, the control law is obtained: ; in, Step size factor This is used to make the control law more general; As a weighting factor; S803: Based on the control law, the pseudo-partial derivative estimate in step 7 and attenuation factor Establish a control law with a decay factor to calculate the control input. .

[0064] The attenuation factor in this application embodiment can gradually reduce the weight ratio of measurement error during the iterative update process of the control algorithm, thereby gradually weakening the impact of measurement disturbance on the control input, reducing the cumulative effect of noise signal in the iteration process, thus maintaining the stability of the system output and improving the control strategy's ability to suppress disturbances; furthermore, the attenuation factor in this application embodiment can moderately attenuate the information affected by disturbances in past iterations, reduce the interference of invalid data on the current control decision, make the iterative learning process more focused on effective information, gradually optimize the control quantity, thereby improving the dynamic response characteristics of the system to a certain extent and enhancing the consistency of control performance.

[0065] Step 9: Sampling time With the maximum sampling time of a single iteration If a comparison is made, Then return to step 6 and output the measurement from step 6. Updated to the measurement output acquired in step 8 after the run. Otherwise, proceed to step 10. Step 10: Collect the measurement errors at all times within this iteration, and calculate the absolute value of the maximum measurement error. With error threshold If a comparison is made, If yes, return to step 6; otherwise, input the final control. The data is sent to the injection molding system, ending the iterative learning process.

[0066] Among them, the absolute value of the maximum measurement error This value reflects the worst control accuracy in this iteration; if this value is less than or equal to the error threshold... This indicates that the deviation at all times within this iteration is within an acceptable range, avoiding product defects caused by acceptable average error but excessive deviation at local times; here, the error threshold... Specifically .

[0067] like Figure 3 As shown, based on the same application concept, this application embodiment also provides a model-free adaptive iterative learning injection molding speed control system with a decay factor, including an acquisition module, a storage module, a calculation module, an output module, and a comparison module; wherein the acquisition module is used to acquire the maximum sampling time of a single iteration. initial value of pseudo-partial derivative Initial control input Initial system output data Expected output and measurement output after operation The storage module stores relevant data; the calculation module calculates measurement errors. Pseudo-partial derivative estimates Control input The comparison module is used to compare the sampling time. With the maximum sampling time of a single iteration Compare, and the absolute value of the maximum measurement error. With error threshold The comparison is performed; the output module is used to convert the final control input. Send to the injection molding system.

[0068] Based on the same concept, embodiments of this application also provide a computer device, including at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by a processor, the instructions being executed by the at least one processor to cause the at least one processor to implement the above-described method when executing the instructions.

[0069] Based on the same concept, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method.

[0070] The convergence analysis is performed below: 1. Prove that the error must converge: Define the pseudo-partial derivative estimation error as Subtract both sides of the pseudo-partial derivative estimation algorithm simultaneously Formula (1) is obtained: ; in ; Depend on achievable ; by measurement disturbance Bounded attainable It is bounded, making , where M is a positive constant.

[0071] Taking the absolute value of both sides of formula (1), we get formula (2): ; Depend on and Formula (3) can be obtained: ; According to the basic inequality, by scaling formula (3), we can obtain formulas (4) and (5): ; ; There exists a constant We obtain formula (6): ; From formulas (4) and (5), we can obtain formula (7): ; From formula (7), we can see that Bounded.

[0072] Tracking error The calculation formula is: ; Based on the compact-form dynamic linearized output model and the control law with a decay factor, we substitute these into the tracking error. Formula (8) can be obtained from the calculation formula: ; make Formula (8) can be written as formula (9): ; Due to measurement disturbance It is bounded, therefore it has positive constants. Thus, we obtain formula (10): ; make ,for There exists a constant This makes the following inequality formula (11) hold: ; Therefore, positive numbers exist. , Thus, we obtain formula (12): ; Assume there exists a constant Number of iterations If it is a positive integer, then we get formula (13): ; Taking the absolute value of both sides of formula (9), and combining it with formulas (12) and (13), we can obtain formula (14): ; According to formula (14), when the number of iterations approaches infinity, we can obtain formula (15): ; Equation (15) shows that within a finite time interval, when the number of iterations... When approaching infinity, the tracking error The iteration converges to zero point by point along the iteration axis; therefore, the embodiments of this application have strong robustness to the effects of measurement disturbances, and the injection molding process achieves better output performance.

[0073] The above proof process, from the perspective of algebraic convergence, uses the recursive inequality scaling method to prove that the embodiments of this application can make the tracking error of the injection molding process containing measurement disturbance converge to zero point by point along the iteration axis, that is, to prove that "the error must converge". 2. Prove that convergence is not violated by random perturbations: Taking the limit on both sides of formula (9) Formula (16) can be obtained: ; Therefore, we obtain formula (17): ; Using measurement disturbance Based on the statistical properties, taking the expected values ​​of both sides of formula (17) yields formula (18): ; Squaring both sides of formula (9) yields formula (19): ; Using measurement disturbance Based on the statistical properties, taking the expected values ​​of both sides of formula (19) yields formula (20): ; because ,and Substitute into formula (20) and simultaneously apply to both sides Taking the limit, we get formula (21): ; From formula (21), we can obtain formula (22): ; In summary, the MFAILC algorithm based on the attenuation factor guarantees that the mean of the output tracking error is equal to 0, and the error variance is equal to 0. Convergence analysis complete.

[0074] From formula (18) and formula (22) It can be concluded that even with measurement disturbances, the embodiments of this application can still make the mathematical expectation and variance of the error converge to zero, ensuring statistical optimality and disturbance robustness, that is, proving that "convergence is not destroyed by random disturbances".

[0075] It should be noted that in practical applications, due to the combined effects of various factors such as component precision limitations, environmental interference, and circuit noise, a certain degree of filter error is inevitable, although the magnitude of this error is relatively small. Therefore, the output error and variance will not be equal to zero in a strictly theoretical sense, but will approach an extremely small positive number. However, even with the aforementioned actual situation, compared with general model-free adaptive iterative learning control algorithms, the embodiments of this application still have a significant improvement in output performance.

[0076] This concludes the convergence analysis. The simulation verification will proceed below: It should be noted that, to simplify the model during the injection phase, the following state variables are introduced: , , , , The injection phase model is simplified to the following state equation: ; Since the injection phase is a nonlinear system, the injection phase model provided in this application embodiment is only used to generate input and output data. The remaining parameters in the model are all known system parameters, and their magnitudes and physical meanings are detailed in Table 1 below:

[0077] Table 1

[0078] The following simulation experiment was conducted using the MATLAB platform to verify the effectiveness of the embodiments of this application in tracking injection speed trajectory. According to the state equation, the injection speed system of the injection molding machine is a single-input single-output system; based on this, a mathematical model of the injection speed was constructed in Simulink. Furthermore, to simulate measurement disturbances in the actual environment, white noise with a frequency range of 500-600Hz was introduced into the system to verify the robustness of the embodiments of this application.

[0079] The controller parameters involved in the MATLAB simulation are set as follows: , , , , , Initial control input Initial system output data Maximum sampling time in a single iteration Sampling time is .

[0080] The simulation results are as follows: Figure 6 This paper presents a comparison of the maximum tracking error (maximum learning error) curves in the iterative domain between an embodiment of this application (MFAILC scheme with a decay factor) and a general model-free adaptive iterative learning control scheme (MFAILC scheme); Figure 6 It can be seen that after the 30th iteration of learning, the tracking error of this embodiment of the application is significantly reduced, while the iterative index tracking error is also reduced. The convergence speed decreases rapidly along the iteration axis and is faster than that of a typical model-free adaptive iterative learning control scheme. Therefore, the convergence performance of the embodiments in this application is better than that of a typical model-free adaptive iterative learning control scheme.

[0081] Figure 7 This paper presents a comparison of the maximum tracking error curves in the iterative domain under different measurement noise variance conditions, based on embodiments of this application; Figure 7 It can be seen that the smaller the measurement noise variance, the more significant the improvement in convergence performance, and even if the measurement noise variance is large, the tracking error can still be ensured to converge on the iteration axis.

[0082] Figure 8 The injection speed tracking curves under different control schemes are shown. In injection molding systems, PID control is a common control method, and the PID controller structure is as follows: ,in , , As shown in Figure 8, after 50 iterations of learning, the embodiment of this application has a significant advantage over the general MFAILC scheme and PID control scheme in terms of injection speed tracking accuracy.

[0083] Figure 9 This paper demonstrates the injection speed tracking error of embodiments of this application and a PID control scheme; by Figure 9 As can be calculated, at the 50th iteration of learning, the average tracking error value of the embodiment of this application is The average error value of the tracking trajectory error of the injection molding system under PID control is Therefore, it can be seen that after 50 learning cycles, the average tracking error of the embodiment of this application is significantly smaller than the average tracking error under PID control, which fully demonstrates that the embodiment of this application has excellent tracking performance.

[0084] The above experimental results clearly demonstrate that the embodiments of this application can effectively suppress the adverse effects of measurement disturbances on injection speed, effectively reduce tracking deviation, and enable the injection speed to closely and stably follow the preset trajectory. This ensures that the injection molding process can maintain a high-precision operating state under complex interference, providing a strong guarantee for the high quality and stability of injection molding production.

[0085] In this embodiment, the computer program, when run by the processor, can also execute other machine-readable instructions to perform other methods as described in the embodiments. For details on the specific execution steps and principles, please refer to the description of the embodiments, which will not be repeated here.

[0086] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0087] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0088] In addition, the functional units in the embodiments provided in 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.

[0089] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 several 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 described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0090] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0091] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A model-free adaptive iterative learning injection molding speed control method with a decay factor, characterized in that, include: Step 1: Obtain the maximum sampling time in a single iteration initial value of pseudo-partial derivative Initial control inputs for system operation and initial system output data ; in, To control the input, indicating the first Next iteration, The flow rate of the hydraulic system during the injection process; For system output, indicating the first... Next iteration, The injection speed during the injection process; Sampling time, ; For the number of iterations, ; Step 2: Based on the input and output data, model the injection process as a nonlinear system model with iterative characteristics; Step 3: Introduce pseudo-partial derivatives The nonlinear system model with iterative characteristics described in step 2 is linearized on the iteration axis to obtain a compact dynamic linearized model. Step 4: Based on the compact dynamic linearization model described in Step 3, transform the nonlinear system model with iterative characteristics described in Step 2 into a compact dynamic linearization output model; Step 5: Introduce measurement disturbance The compact-form dynamic linearized output model described in step 4 is then extended to an actual output model that includes measurement perturbations: ； in, For the first Next iteration, At any given time, the measurement output includes the measurement disturbance; the measurement disturbance It is a random signal with the following statistical properties: , ,in For mathematical expectation factor, The variance factor is the measurement disturbance. With the sampling time The number of iterations The measurement output The system output The control input and the pseudo-partial derivatives All are uncorrelated, measurement disturbance initial disturbance value =0; Step 6: Based on the expected output and the measurement output Calculate the measurement error ; Step 7: Construct a pseudo-partial derivative estimation algorithm and calculate the pseudo-partial derivative estimates. ; Step 8: Based on the measurement error described in Step 6 and the pseudo-partial derivative estimate in step 7 Construct a control law with a decay factor and calculate the control input. The control input The measured output is sent to the injection molding system and collected after the operation. ; The control law with the attenuation factor is as follows: ； in, Step size factor This is used to make the control law with the attenuation factor more general; As a weighting factor; This is a decay factor used to gradually reduce the measurement error as the number of iterations increases. Weighting percentage; Step 9: For the sampling time With the maximum sampling time of the single iteration If a comparison is made, Then return to step 6 and output the measurement from step 6. Updated to the measurement output acquired in step 8 after the run. Otherwise, proceed to step 10. Step 10: Collect the measurement errors at all times within this iteration, and calculate the absolute value of the maximum measurement error. With error threshold If a comparison is made, If so, return to step 6; otherwise, return to the final control input. The data is sent to the injection molding system, ending the iterative learning process.

2. The model-free adaptive iterative learning injection speed control method with attenuation factor according to claim 1, characterized in that, The nonlinear system models with iterative characteristics modeled in step 2 include: ； in, To output the order of the variable, The order of the input variable.

3. The model-free adaptive iterative learning injection speed control method with attenuation factor according to claim 1, characterized in that, The compact-format dynamic linearization model established in step 3 includes: ； ； ； in, and These represent the changes in control input and system output respectively during adjacent batches of system operation; for any sampling time... and the number of iterations All satisfy ,in, It is a constant; all pseudo-partial derivatives All symbols remain consistent, that is ,in, It is a constant.

4. The model-free adaptive iterative learning injection speed control method with attenuation factor according to claim 1, characterized in that, The compact-format dynamic linearized output model derived in step 4 includes: 。 5. The model-free adaptive iterative learning injection speed control method with attenuation factor according to claim 1, characterized in that, Step 6 involves calculating the measurement error. The methods include: ； in, The desired output represents the known optimal injection speed; For the first Next iteration, The measured output at time [time].

6. The model-free adaptive iterative learning injection speed control method with attenuation factor according to claim 1, characterized in that, In step 7, a pseudo-partial derivative estimation algorithm is constructed, and the pseudo-partial derivative estimates are calculated. The steps include: A pseudo-partial derivative estimation function is established, and a quantitative index for the estimation error is determined; the pseudo-partial derivative estimation function includes: ;in, The objective function is... These are the weighting coefficients; This is an estimate of the pseudo-partial derivative; To measure the change in output, the calculation formula is as follows: ; To control the amount of input variation, the calculation formula is as follows: ; Based on the pseudo-partial derivative estimation function and optimal conditions The pseudo-partial derivative estimation algorithm is obtained, and the pseudo-partial derivative estimates are calculated. The pseudo-partial derivative estimation algorithm includes: ;in, Step size factor This is used to easily adjust the pseudo-partial derivatives and enhance the flexibility of the algorithm; These are the weighting coefficients; A reset algorithm is established to ensure that the pseudo-partial derivative estimation algorithm has good tracking performance; the reset algorithm includes: ； like or or ; in, The initial value of the pseudo-partial derivative obtained in step 1 ; To reset the threshold; It is a symbolic function.

7. The model-free adaptive iterative learning injection speed control method with a decay factor according to claim 1, characterized in that, In step 8, the measurement error is based on that in step 6. and the pseudo-partial derivative estimate in step 7 The steps to construct a control law with a decay factor include: Establish the objective function for controlling the input: ； in, The objective function is... As a weighting factor; Based on the control input objective function and optimal conditions Thus, the control law is obtained: ; in, Step size factor This is used to make the control law more general; As a weighting factor; Based on the control law and the pseudo-partial derivative estimate in step 7. and attenuation factor Establish the control law with the attenuation factor to calculate the control input. .

8. A model-free adaptive iterative learning injection molding speed control system with a decay factor, characterized in that, include: Acquisition module: Used to obtain the maximum sampling time in a single iteration. initial value of pseudo-partial derivative Initial control input Initial system output data Expected output and measurement output after operation ; Storage module: Used to store relevant data; Calculation module: used to calculate measurement error Pseudo-partial derivative estimates Control input ; Comparison module: used for sampling time Maximum sampling time in a single iteration Compare, and the absolute value of the maximum measurement error. With error threshold Compare; Output module: Used to output the final control input. Send to the injection molding system.

9. A computer device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the steps of the method according to any one of claims 1 to 7 when executing the instructions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images