An intelligent pressure maintaining control method for thin-walled products of an injection molding machine based on mold pressure feedback

Abstract

The application discloses a kind of based on mould pressing feedback's intelligent pressure maintaining control method of injection molding machine thin-walled product, comprising the following steps: (1) hardware layer construction;(2) software layer configuration;(3) cavity pressure acquisition module passes through pressure sensor and acquires original pressure data in injection molding cavity in real time;(4) data preprocessing module carries out filtering processing to the pressure signal generated by pressure data, (5) dynamic pressure maintaining control module passes through receiving real-time pressure data, product wall thickness parameter, mould structure parameter and injection molding process parameter, dynamically adjusts pressure parameter and pressure maintaining time, generates pressure maintaining control instruction and sends to injection molding machine control system, drives injection molding machine to execute pressure maintaining action;(6) dynamic pressure maintaining control module stores the pressure change curve of this pressure maintaining process, parameter adjustment record correlation in feedback learning module, carries out optimization calculation;(7) man-machine interface real-time display injection molding cavity's pressure change curve, pressure maintaining control parameter and forming state.

Description

Technical Field

[0001] This invention relates to the field of injection molding technology, and in particular to an intelligent pressure holding control method for thin-walled products in injection molding machines based on molding feedback. Background Technology

[0002] In the injection molding of thin-walled plastic products (such as electronic component housings and thin-walled food packaging parts), pressure control is a core factor determining product quality. During the pressure holding stage, continuous pressure is applied to compensate for melt shrinkage and prevent defects such as shrinkage and warping. However, current traditional injection molding machines still rely on a "fixed holding time + constant pressure setting" model, lacking dynamic adjustment capabilities linked to the actual molding state. This fails to meet the high-precision, high-efficiency molding requirements of thin-walled parts. Specific problems include:

[0003] Traditional pressure holding control does not incorporate mold cavity pressure detection and feedback mechanisms, relying solely on preset fixed parameters: the pressure holding time needs to be set manually based on experience, making it impossible to accurately identify the turning points in the pressure holding stage (such as the critical moment when the melt completely fills the mold cavity and begins to cool and shrink). If the pressure holding time is too short, the melt shrinkage is not fully compensated, and the product is prone to surface shrinkage and depressions; if the pressure holding time is too long, excess melt continues to squeeze the mold cavity, leading to increased internal stress in the product, which is prone to warping and deformation after subsequent cooling.

[0004] Constant pressure settings cannot adapt to viscosity changes during melt cooling. For example, in thin-walled parts, the mold cavity space is narrow and the melt cools quickly. Higher pressure is required later to drive the melt flow to compensate for shrinkage. However, traditional constant pressure cannot meet this dynamic requirement, further aggravating molding defects. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent pressure holding control method for thin-walled products of injection molding machines based on molding feedback. This intelligent pressure holding control method for thin-walled products of injection molding machines based on molding feedback can realize precise dynamic adjustment and optimization in the pressure holding stage, thereby improving molding quality and production efficiency.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for intelligent pressure holding control of thin-walled products in injection molding machines based on molding feedback, characterized by the following steps:

[0008] (1) Hardware layer construction: The hardware layer includes a cavity pressure acquisition module, an injection molding machine control system, a human-machine interface, a molding quality detection device and a communication module; the cavity pressure acquisition module is connected to the injection molding machine control system and the human-machine interface through the communication module; the pressure sensor is set inside the injection mold cavity;

[0009] (2) Software layer configuration: The software layer includes a data preprocessing module, a parameter input module, a dynamic pressure holding control module and a feedback learning module; the injection molding machine control system has a built-in table-driven strategy table and a regression algorithm model. The table-driven strategy table is based on the preset pressure holding parameters of the experimental data; the regression algorithm model calculates the pressure holding parameters in real time based on the multivariate linear regression formula.

[0010] (3) The cavity pressure acquisition module collects the original pressure data in the injection mold cavity in real time through the pressure sensor and transmits the collected original pressure data to the data preprocessing module; the parameter input module provides the product wall thickness parameters, mold structure parameters and injection process parameters to the table-driven strategy table and the regression algorithm model to assist the dynamic pressure holding control module in making decisions;

[0011] (4) The data preprocessing module filters the pressure signal generated by the pressure data to generate optimized real-time pressure data; at the same time, the data preprocessing module receives the product wall thickness parameters, mold structure parameters and injection molding process parameters, and sends the processing results to the dynamic pressure holding control module.

[0012] (5) The dynamic pressure holding control module receives real-time pressure data, product wall thickness parameters, mold structure parameters and injection molding process parameters, analyzes the pressure change trend of the injection mold cavity in real time based on the real-time pressure data, generates the pressure change curve of the injection mold cavity in real time, identifies the turning point of the pressure change curve to the pressure holding stage, and then dynamically adjusts the pressure parameters and pressure holding time in combination with product wall thickness parameters, mold structure parameters and injection molding process parameters.

[0013] The dynamic pressure holding control module calls the table-driven strategy table to perform initial parameter matching, and at the same time starts the regression algorithm model to perform real-time parameter calculation. Then, it combines the outputs of the table-driven strategy table and the regression algorithm model to generate pressure holding control instructions, and sends the pressure holding control instructions to the injection molding machine control system to drive the injection molding machine to perform pressure holding actions.

[0014] (6) The dynamic pressure holding control module associates and stores the pressure change curve and parameter adjustment record of this pressure holding process into the feedback learning module for subsequent optimization calculations;

[0015] (7) The human-machine interface obtains the pressure holding time and pressure parameters sent by the dynamic pressure holding control module, and displays the pressure change curve of the injection mold cavity, pressure holding control parameters and molding status in real time, so that the operator can monitor and make manual intervention.

[0016] In step (3) above, since the pressure sensor will inevitably be superimposed with high-frequency noise such as electromagnetic interference and mechanical vibration during high-speed sampling, in order to ensure the accuracy of subsequent analysis, the data preprocessing module needs to filter the pressure signal generated by the pressure data.

[0017] In step (5) above, the dynamic pressure holding control module will connect the continuously received "optimized real-time pressure data" along the time axis and convert it into a pressure change curve of the injection mold cavity that intuitively reflects the pressure change over time. This pressure change curve is not only the core basis for the module to identify the turning point of reaching the pressure holding stage (i.e., the critical moment when the pressure changes from rising to stabilizing), but also an important reference for subsequent adjustment of pressure parameters and generation of pressure holding control commands. In short, the pressure change curve of the injection mold cavity is a data form that the dynamic pressure holding control module generates synchronously when analyzing real-time pressure data for trend judgment and parameter decision-making.

[0018] The aforementioned communication module is responsible for data transmission between modules and linkage with the injection molding machine control system, ensuring the timely and accurate transmission of control commands and status data, realizing closed-loop system control, and guaranteeing smooth information sharing.

[0019] In the preferred embodiment, in step (4), the data preprocessing module filters the pressure signal generated by the pressure data using a pressure data filtering algorithm; the pressure data filtering algorithm uses a first-order low-pass filter, and its recursive formula is as follows:

[0020] ,in:

[0021] This is the raw pressure value collected at the current moment;

[0022] The filtered pressure value at the nth sampling point is expressed in MPa, where n is the sampling point number (n=1, 2, ..., n).

[0023] 'a' is the filter coefficient, 0 ≤ a ≤ 1, and the smaller 'a' is, the smoother the filter.

[0024] This is the filtered pressure value at the (n-1)th sampling point, in MPa.

[0025] The selection of the above-mentioned filter coefficient 'a' needs to be adapted to the application scenario: the value of 'a' in the filtering algorithm determines the smoothness of the signal and the response speed. When processing pressure signals with drastic fluctuations, 'a' should be appropriately reduced (e.g., 0.1-0.3) to improve the filtering capability; when the system needs to respond quickly to the adjustment of the control strategy, 'a' can be appropriately increased (e.g., 0.4-0.6) to improve the response speed.

[0026] In a further optimized scheme, step (5) employs a combination of slope detection and local maximum detection to identify the inflection point in the pressure change curve that has reached the pressure holding stage. The operation steps are as follows:

[0027] (5-1) The slope detection method is achieved by calculating the first derivative of continuous pressure data points, i.e.: ,

[0028] in: The slope of the nth sampling point relative to the (n-1)th sampling point;

[0029] This is the filtered pressure value at the nth sampling point, in MPa.

[0030] This is the filtered pressure value at the (n-1)th sampling point, in MPa.

[0031] If the slope is less than or equal to 0 at several consecutive points, then this point is determined to be the turning point of the pressure holding stage.

[0032] (5-2) The operation steps of the local maximum detection method are as follows:

[0033] (5-2.1) Set the search window length: Based on the filling time characteristics of the injection molding process, preset the search window length for local extrema;

[0034] (5-2.2) Define the rule for determining local maxima: If a certain sampling point If the condition is met that "the pressure is at its maximum value within the current search window, and the rate of change of pressure within the next three consecutive search windows is less than 10% of the current pressure value", then the point is marked as the turning point of the candidate pressure holding stage.

[0035] The formula for the rate of change of pressure is: ,

[0036] in: The pressure change rate reflects how quickly the pressure changes, and its unit is MPa / s;

[0037] The sampling period represents the time interval between two consecutive pressure samples.

[0038] i represents the number of sampling points in the time interval;

[0039] (5-2.3) Starting from the start of the injection molding process, the filtered pressure values ​​of the mold cavity are traversed by sliding a set search window. :

[0040] (5-2.3.1) When a sampling point is detected to meet the local maximum determination rule in step (5-2.2), the slope before and after the point is extracted. Perform secondary verification;

[0041] (5-2.3.2) If the slope before that point is (i=1, 2, ..., 5) greater than 0, the slope thereafter When (i=1, 2, ..., 5) rapidly decreases to less than or equal to 0, then this point is determined to be the turning point of the pressure holding phase;

[0042] (5-2.3.3) If the verification fails, adjust the search window length by increasing or decreasing the number of sampling points and repeat the local maximum detection process described above.

[0043] In step (5-2.1) above, if the filling phase lasts approximately 1 second and the pressure sampling frequency is... If the value is 0.01s, then the search window is set to a continuous i=20 sampling points.

[0044] The pressure change rate formula above reflects the magnitude of pressure change per unit time, i.e., the speed of change.

[0045] By combining slope detection and local maximum detection, misjudgments caused by pressure fluctuations in single slope monitoring can be effectively avoided, improving the stability and accuracy of inflection point identification during the pressure holding phase.

[0046] In the preferred embodiment, in step (5), the table-driven strategy table receives the optimized holding pressure parameters from the feedback learning module and outputs the preset parameter matching results to the dynamic holding pressure control module; the regression algorithm model receives the corrected coefficients from the feedback learning module and outputs the real-time parameter calculation results to the dynamic holding pressure control module. Regardless of whether the table-driven strategy table or the regression algorithm model is used, the generated holding pressure time and pressure must meet the physical and logical limitations of the injection molding machine control system. For example, the maximum holding pressure should not exceed the rated oil pressure of the system; the holding pressure time should not be less than the minimum value before the material is cooled (e.g., at least 0.8s for ABS). The switching threshold between the table-driven strategy table and the regression algorithm model needs to be set according to the product precision requirements: for general products with low precision requirements, the table-driven strategy table can be used first to improve the response speed; for high-precision thin-walled products, the regression algorithm model is automatically switched to ensure control precision. The above feedback learning module outputs the optimized holding pressure parameters to the table-driven strategy table and outputs the corrected coefficients to the regression algorithm model, forming a closed-loop optimization of the dual strategies. By using a dual-strategy collaborative decision-making approach combining a table-driven strategy table and a regression algorithm model, the holding time and pressure are adjusted in real time based on changes in injection mold cavity pressure, balancing control efficiency and accuracy while avoiding defects such as shrinkage and warping caused by traditional fixed parameters. This enables adaptive selection of the optimal control strategy under different working conditions, broadening the system's application scope.

[0047] In a further preferred embodiment, in step (5), the table entries of the table-driven strategy table include material, wall thickness, die stamping time, recommended holding time and holding pressure. The injection molding machine control system automatically matches the corresponding table entries and executes them based on the current sampling parameters.

[0048] In a further optimized scheme, in step (5), the multivariate linear regression formula in the regression algorithm model is as follows:

[0049] Holding pressure formula: Hold_Pressure = a1 × Thickness + a2 × Material_Code + a3 × T_fill + b;

[0050] The formula for holding time is: Hold_Time = c1 × Thickness + c2 × Pressure_Max + d;

[0051] a1: Regression coefficient for cavity wall thickness;

[0052] a2: Regression coefficient of Material Code;

[0053] a3: Regression coefficient for fill time (T_fill);

[0054] b: Basic adjustment value, a correction item to ensure that the formula calculation results are more in line with actual production;

[0055] c1: Regression coefficient for cavity wall thickness;

[0056] c2: Regression coefficient of maximum filling pressure (Pressure_Max);

[0057] d: Basic adjustment value, a correction item to ensure that the formula calculation results are more in line with actual production.

[0058] In the preferred embodiment, in step (6), after the pressure holding process is completed, the molded product is subjected to quality inspection to obtain key indicator data of the molded product, and the detected key indicator data is transmitted to the feedback learning module. The feedback learning module manages and stores the key indicator data of the molded product, and optimizes and adaptively adjusts it in combination with the pressure change curve and parameter adjustment record of this pressure holding process, and sends the optimized pressure holding control parameters to the dynamic pressure holding control module. The aforementioned key indicator data includes the shrinkage rate, warpage, and internal stress value of the molded product.

[0059] In a further preferred embodiment, in step (6), after the pressure holding process is completed, the molded product is inspected using a molding quality inspection device to obtain key indicator data of the molded product. The molding quality inspection device includes a dimensional measuring instrument and a stress measuring instrument. The dimensional measuring instrument is used to detect the shrinkage rate and warpage of the molded product; the stress measuring instrument is used to detect the internal stress value of the molded product.

[0060] In a further preferred embodiment, in step (6), the injection molding machine control system first presets a threshold for the deviation coefficient, and then performs a multivariate regression analysis on the holding time, pressure and key indicator data through the feedback learning module to calculate the deviation coefficient; when the deviation coefficient exceeds the preset threshold, the coefficient of the regression algorithm model is automatically modified, and the holding parameters after the table-driven strategy table is updated.

[0061] In the preferred embodiment, in step (7), the injection molding machine control system presets the number of molding cycles. After each preset number of molding cycles is completed, the feedback learning module generates a parameter optimization report and prompts the operator through the human-machine interface to confirm whether to enable the new parameters. If the pressure in the injection mold cavity suddenly rises to more than 120% or the pressure sensor communication is interrupted during the pressure holding process, the dynamic pressure holding control module immediately triggers the emergency mechanism: suspends the pressure holding action, issues an alarm signal through the human-machine interface, and automatically calls the historical optimal parameters for a reset attempt. After the pressure holding abnormality is resolved, the injection molding machine is re-driven to perform the pressure holding action.

[0062] In the preferred embodiment, in step (7), the human-machine interface includes a real-time pressure curve display area, a pressure holding control parameter setting area, an alarm and status display area, and a control operation button area, which facilitates the operator to perform visual monitoring and parameter adjustment of the injection molding process, thereby improving the system's intelligence and ease of operation.

[0063] In the preferred embodiment, the pressure change curve of the injection mold cavity in step (7) refers to the curve showing the change of pressure within the injection mold cavity over time. The pressure change curve includes the filling stage, the holding stage, and the cooling stage. Typically, the pressure change curve of the injection mold cavity marks key time points such as the injection start point T0, the filling completion point T1, the inflection point of the holding stage T2, the holding end point T3, and the cooling end point T4, to illustrate the dynamic identification basis of the holding pressure control parameters. The filling stage refers to the time period from the injection start point T0 to the filling completion point T1. The holding stage refers to the time period from the holding stage inflection point T2 to the holding end point T3. The cooling stage refers to the time period from the holding end point T3 to the cooling end point T4.

[0064] Compared with the prior art, the present invention has the following advantages:

[0065] (1) This invention uses a cavity pressure sensor to collect pressure change curves in real time during the injection molding process of thin-walled plastic products. Combined with the product wall thickness and mold structure parameters, the holding time and holding pressure are adjusted in real time through a dynamic feedback control module. The fluctuation trend and turning point of the cavity pressure are analyzed, and the start and end times of the holding stage are accurately determined. This enables adaptive adjustment of the holding parameters, thereby optimizing the injection molding process, improving molding quality and production efficiency, and realizing intelligent and efficient control of the injection molding process.

[0066] (2) By realizing real-time intelligent adjustment of the holding pressure parameters, the present invention can improve the first pass rate of injection molded products, reduce scrap and rework caused by over-holding or under-holding pressure, reduce material and energy costs, and reduce manual intervention and debugging time, thereby improving production efficiency and creating considerable economic benefits.

[0067] (3) This invention fills the technical gap in intelligent control of the pressure holding stage of injection molding, has wide industrial adaptability, is suitable for injection molding scenarios with high requirements for molding accuracy and product quality, especially suitable for the production of thin-walled products, optical components, precision structural parts, etc., and can be widely used in medical devices, electronic products, automotive parts and intelligent injection molding production lines. It conforms to the development trend of intelligent manufacturing and precision injection molding, and has good market prospects and industrialization potential. Attached Figure Description

[0068] Figure 1 This is a general structural block diagram of a specific embodiment of the present invention;

[0069] Figure 2 This is a flowchart of the dynamic pressure holding control according to a specific embodiment of the present invention;

[0070] Figure 3 This is a pressure change curve of the injection mold cavity according to a specific embodiment of the present invention. Detailed Implementation

[0071] The following description, in conjunction with the accompanying drawings and preferred embodiments of the present invention, will provide further details.

[0072] like Figure 1-3 As shown in this embodiment, the intelligent pressure holding control method for thin-walled injection molded products based on molding feedback includes the following steps:

[0073] (1) Hardware layer construction: The hardware layer includes a cavity pressure acquisition module, an injection molding machine control system, a human-machine interface, a molding quality detection device and a communication module; the cavity pressure acquisition module is connected to the injection molding machine control system and the human-machine interface through the communication module; the pressure sensor is set inside the injection mold cavity;

[0074] (2) Software layer configuration: The software layer includes a data preprocessing module, a parameter input module, a dynamic pressure holding control module and a feedback learning module; the injection molding machine control system has a built-in table-driven strategy table and a regression algorithm model. The table-driven strategy table is based on the preset pressure holding parameters of the experimental data; the regression algorithm model calculates the pressure holding parameters in real time based on the multivariate linear regression formula.

[0075] (3) The cavity pressure acquisition module collects the original pressure data in the injection mold cavity in real time through the pressure sensor and transmits the collected original pressure data to the data preprocessing module; the parameter input module provides the product wall thickness parameters, mold structure parameters and injection process parameters to the table-driven strategy table and the regression algorithm model to assist the dynamic pressure holding control module in making decisions;

[0076] (4) The data preprocessing module filters the pressure signal generated by the pressure data to generate optimized real-time pressure data; at the same time, the data preprocessing module receives the product wall thickness parameters, mold structure parameters and injection molding process parameters, and sends the processing results to the dynamic pressure holding control module.

[0077] (5) The dynamic pressure holding control module receives real-time pressure data, product wall thickness parameters, mold structure parameters and injection molding process parameters, analyzes the pressure change trend of the injection mold cavity in real time based on the real-time pressure data, generates the pressure change curve of the injection mold cavity in real time, identifies the turning point of the pressure change curve to the pressure holding stage, and then dynamically adjusts the pressure parameters and pressure holding time in combination with product wall thickness parameters, mold structure parameters and injection molding process parameters.

[0078] The dynamic pressure holding control module calls the table-driven strategy table to perform initial parameter matching, and at the same time starts the regression algorithm model to perform real-time parameter calculation. Then, it combines the outputs of the table-driven strategy table and the regression algorithm model to generate pressure holding control instructions, and sends the pressure holding control instructions to the injection molding machine control system to drive the injection molding machine to perform pressure holding actions.

[0079] (6) The dynamic pressure holding control module associates and stores the pressure change curve and parameter adjustment record of this pressure holding process into the feedback learning module for subsequent optimization calculations;

[0080] (7) The human-machine interface obtains the pressure holding time and pressure parameters sent by the dynamic pressure holding control module, and displays the pressure change curve of the injection mold cavity, pressure holding control parameters and molding status in real time, so that the operator can monitor and make manual intervention.

[0081] In step (3) above, since the pressure sensor will inevitably be superimposed with high-frequency noise such as electromagnetic interference and mechanical vibration during high-speed sampling, in order to ensure the accuracy of subsequent analysis, the data preprocessing module needs to filter the pressure signal generated by the pressure data.

[0082] In step (5) above, the dynamic pressure holding control module will connect the continuously received "optimized real-time pressure data" along the time axis and convert it into a pressure change curve of the injection mold cavity that intuitively reflects the pressure change over time. This pressure change curve is not only the core basis for the module to identify the turning point of reaching the pressure holding stage (i.e., the critical moment when the pressure changes from rising to stabilizing), but also an important reference for subsequent adjustment of pressure parameters and generation of pressure holding control commands. In short, the pressure change curve of the injection mold cavity is a data form that the dynamic pressure holding control module generates synchronously when analyzing real-time pressure data for trend judgment and parameter decision-making.

[0083] The aforementioned communication module is responsible for data transmission between modules and linkage with the injection molding machine control system, ensuring the timely and accurate transmission of control commands and status data, realizing closed-loop system control, and guaranteeing smooth information sharing.

[0084] In step (4), the data preprocessing module filters the pressure signal generated by the pressure data using a pressure data filtering algorithm; the pressure data filtering algorithm uses a first-order low-pass filter, and its recursive formula is as follows:

[0085] ,in:

[0086] This is the raw pressure value collected at the current moment;

[0087] The filtered pressure value at the nth sampling point is expressed in MPa, where n is the sampling point number (n=1, 2, ..., n).

[0088] 'a' is the filter coefficient, 0 ≤ a ≤ 1, and the smaller 'a' is, the smoother the filter.

[0089] This is the filtered pressure value at the (n-1)th sampling point, in MPa.

[0090] The selection of the above-mentioned filter coefficient 'a' needs to be adapted to the application scenario: the value of 'a' in the filtering algorithm determines the smoothness of the signal and the response speed. When processing pressure signals with drastic fluctuations, 'a' should be appropriately reduced (e.g., 0.1-0.3) to improve the filtering capability; when the system needs to respond quickly to the adjustment of the control strategy, 'a' can be appropriately increased (e.g., 0.4-0.6) to improve the response speed.

[0091] In a further optimized scheme, step (5) employs a combination of slope detection and local maximum detection to identify the inflection point in the pressure change curve that has reached the pressure holding stage. The operation steps are as follows:

[0092] (5-1) The slope detection method is achieved by calculating the first derivative of continuous pressure data points, i.e.: ,

[0093] in: The slope of the nth sampling point relative to the (n-1)th sampling point;

[0094] This is the filtered pressure value at the nth sampling point, in MPa.

[0095] This is the filtered pressure value at the (n-1)th sampling point, in MPa.

[0096] If the slope is less than or equal to 0 at several consecutive points, then this point is determined to be the turning point of the pressure holding stage.

[0097] (5-2) The operation steps of the local maximum detection method are as follows:

[0098] (5-2.1) Set the search window length: Based on the filling time characteristics of the injection molding process, preset the search window length for local extrema;

[0099] (5-2.2) Define the rule for determining local maxima: If a certain sampling point If the condition is met that "the pressure is at its maximum value within the current search window, and the rate of change of pressure within the next three consecutive search windows is less than 10% of the current pressure value", then the point is marked as the turning point of the candidate pressure holding stage.

[0100] The formula for the rate of change of pressure is: ,

[0101] in: The pressure change rate reflects how quickly the pressure changes, and its unit is MPa / s;

[0102] The sampling period represents the time interval between two consecutive pressure samples.

[0103] i represents the number of sampling points in the time interval;

[0104] (5-2.3) Starting from the start of the injection molding process, the filtered pressure values ​​of the mold cavity are traversed by sliding a set search window. :

[0105] (5-2.3.1) When a sampling point is detected to meet the local maximum determination rule in step (5-2.2), the slope before and after the point is extracted. Perform secondary verification;

[0106] (5-2.3.2) If the slope before that point is (i=1, 2, ..., 5) greater than 0, the slope thereafter When (i=1, 2, ..., 5) rapidly decreases to less than or equal to 0, then this point is determined to be the turning point of the pressure holding phase;

[0107] (5-2.3.3) If the verification fails, adjust the search window length by increasing or decreasing the number of sampling points and repeat the local maximum detection process described above.

[0108] In step (5-2.1) above, if the filling phase lasts approximately 1 second and the pressure sampling frequency is... If the value is 0.01s, then the search window is set to a continuous i=20 sampling points.

[0109] The pressure change rate formula above reflects the magnitude of pressure change per unit time, i.e., the speed of change.

[0110] By combining slope detection and local maximum detection, misjudgments caused by pressure fluctuations in single slope monitoring can be effectively avoided, improving the stability and accuracy of inflection point identification during the pressure holding phase.

[0111] In step (5), the table-driven strategy table receives the optimized holding pressure parameters from the feedback learning module and outputs the preset parameter matching results to the dynamic holding pressure control module; the regression algorithm model receives the corrected coefficients from the feedback learning module and outputs the real-time parameter calculation results to the dynamic holding pressure control module. Regardless of whether the table-driven strategy table or the regression algorithm model is used, the generated holding pressure time and pressure must meet the physical and logical limitations of the injection molding machine control system. For example, the maximum holding pressure should not exceed the rated oil pressure of the system; the holding pressure time should not be less than the minimum value before the material is cooled (e.g., at least 0.8s for ABS). The switching threshold between the table-driven strategy table and the regression algorithm model needs to be set according to the product precision requirements: for general products with low precision requirements, the table-driven strategy table can be used first to improve the response speed; for high-precision thin-walled products, the automatic switch to the regression algorithm model is used to ensure control precision. The above feedback learning module outputs the optimized holding pressure parameters to the table-driven strategy table and the corrected coefficients to the regression algorithm model, forming a closed-loop optimization of the dual strategies. By using a dual-strategy collaborative decision-making approach combining a table-driven strategy table and a regression algorithm model, the holding time and pressure are adjusted in real time based on changes in injection mold cavity pressure, balancing control efficiency and accuracy while avoiding defects such as shrinkage and warping caused by traditional fixed parameters. This enables adaptive selection of the optimal control strategy under different working conditions, broadening the system's application scope.

[0112] In step (5), the table entries of the table-driven strategy table include material, wall thickness, die time, recommended holding time and holding pressure. The injection molding machine control system automatically matches the corresponding table entries and executes them based on the current sampled parameters.

[0113] A pre-defined table-driven strategy is used for common operating conditions, as shown in the table below:

[0114] Material Wall thickness (mm) Die stamping time (ms) Recommended holding time (s) Holding pressure (MPa) PP 0.6 ＜400 1.2 38 ABS 1.0 ＜600 1.8 45

[0115] In step (5), the multivariate linear regression formula in the regression algorithm model is as follows:

[0116] Holding pressure formula: Hold_Pressure = a1 × Thickness + a2 × Material_Code + a3 × T_fill + b;

[0117] The formula for holding time is: Hold_Time = c1 × Thickness + c2 × Pressure_Max + d;

[0118] a1: Regression coefficient for cavity wall thickness;

[0119] a2: Regression coefficient of Material Code;

[0120] a3: Regression coefficient for fill time (T_fill);

[0121] b: Basic adjustment value, a correction item to ensure that the formula calculation results are more in line with actual production;

[0122] c1: Regression coefficient for cavity wall thickness;

[0123] c2: Regression coefficient of maximum filling pressure (Pressure_Max);

[0124] d: Basic adjustment value, a correction item to ensure that the formula calculation results are more in line with actual production.

[0125] In step (6), after the pressure holding process is completed, the molded product is subjected to quality inspection to obtain key indicator data. The detected key indicator data is then transmitted to the feedback learning module. The feedback learning module manages and stores the key indicator data of the molded product, and optimizes and adaptively adjusts it by combining the pressure change curve and parameter adjustment records of this pressure holding process. The optimized pressure holding control parameters are then sent to the dynamic pressure holding control module. The aforementioned key indicator data includes the shrinkage rate, warpage, and internal stress value of the molded product.

[0126] In step (6), after the pressure holding process is completed, the molded product is inspected using molding quality testing equipment to obtain key indicator data of the molded product. The molding quality testing equipment includes a dimensional measuring instrument and a stress measuring instrument. The dimensional measuring instrument is used to detect the shrinkage rate and warpage of the molded product; the stress measuring instrument is used to detect the internal stress value of the molded product.

[0127] In step (6), the injection molding machine control system first presets the threshold of the deviation coefficient, and then performs multiple regression analysis on the holding time, pressure and key indicator data through the feedback learning module to calculate the deviation coefficient. When the deviation coefficient exceeds the preset threshold, the coefficient of the regression algorithm model is automatically modified and the holding parameters after the table-driven strategy table is updated.

[0128] In step (7), the injection molding machine control system presets the number of molding cycles. After each preset number of molding cycles is completed, the feedback learning module generates a parameter optimization report and prompts the operator through the human-machine interface to confirm whether to enable the new parameters. If the pressure in the injection mold cavity suddenly rises to more than 120% or the pressure sensor communication is interrupted during the pressure holding process, the dynamic pressure holding control module immediately triggers the emergency mechanism: suspends the pressure holding action, issues an alarm signal through the human-machine interface, and automatically calls the historical optimal parameters for a reset attempt. After the pressure holding abnormality is resolved, the injection molding machine is re-driven to perform the pressure holding action.

[0129] In step (7), the human-machine interface includes a real-time pressure curve display area, a pressure holding control parameter setting area, an alarm and status display area, and a control operation button area, which facilitates the operator to visually monitor and adjust parameters of the injection molding process, thereby improving the system's intelligence and ease of operation.

[0130] The pressure change curve of the injection mold cavity in step (7) refers to the curve showing the change of pressure within the injection mold cavity over time. This pressure change curve includes the filling stage, the holding stage, and the cooling stage. Typically, the pressure change curve of the injection mold cavity marks key time points such as the injection start point T0, the filling completion point T1, the inflection point of the holding stage T2, the holding stage end point T3, and the cooling stage end point T4, which are used to illustrate the dynamic identification basis of the holding pressure control parameters. The filling stage refers to the time period from the injection start point T0 to the filling completion point T1. The holding stage refers to the time period from the holding stage inflection point T2 to the holding stage end point T3. The cooling stage refers to the time period from the holding stage end point T3 to the cooling stage end point T4.

[0131] Furthermore, it should be noted that the names of the various parts of the specific embodiments described in this specification may differ. All equivalent or simple variations made to the structure, features, and principles described in this invention are included within the scope of protection of this invention. Those skilled in the art can make various modifications or additions to the described specific embodiments or use similar methods to replace them, as long as they do not deviate from the structure of this invention or exceed the scope defined in these claims, all of which should fall within the scope of protection of this invention.

Claims

1. A method for intelligent pressure holding control of thin-walled products in injection molding machines based on molding feedback, characterized in that... Includes the following steps: (1) Hardware layer construction: The hardware layer includes a cavity pressure acquisition module, an injection molding machine control system, a human-machine interface, a molding quality detection device and a communication module; the cavity pressure acquisition module is connected to the injection molding machine control system and the human-machine interface through the communication module; the pressure sensor is set inside the injection mold cavity; (2) Software layer configuration: The software layer includes a data preprocessing module, a parameter input module, a dynamic pressure holding control module and a feedback learning module; the injection molding machine control system has a built-in table-driven strategy table and a regression algorithm model, and the table-driven strategy table is based on the preset pressure holding parameters of the experimental data; The regression algorithm model calculates the pressure holding parameters in real time based on the multivariate linear regression formula; (3) The cavity pressure acquisition module collects the original pressure data in the injection mold cavity in real time through the pressure sensor and transmits the collected original pressure data to the data preprocessing module. The parameter input module provides the product wall thickness parameters, mold structure parameters, and injection molding process parameters to the table-driven strategy table and the regression algorithm model to assist the dynamic pressure holding control module in making decisions. (4) The data preprocessing module filters the pressure signal generated by the pressure data to generate optimized real-time pressure data; at the same time, the data preprocessing module receives the product wall thickness parameters, mold structure parameters and injection molding process parameters, and sends the processing results to the dynamic pressure holding control module. (5) The dynamic pressure holding control module receives real-time pressure data, product wall thickness parameters, mold structure parameters and injection molding process parameters, analyzes the pressure change trend of the injection mold cavity in real time based on the real-time pressure data, generates the pressure change curve of the injection mold cavity in real time, identifies the turning point of the pressure change curve to the pressure holding stage, and then dynamically adjusts the pressure parameters and pressure holding time in combination with product wall thickness parameters, mold structure parameters and injection molding process parameters. The dynamic pressure holding control module calls the table-driven strategy table to perform initial parameter matching, and at the same time starts the regression algorithm model to perform real-time parameter calculation. Then, it combines the outputs of the table-driven strategy table and the regression algorithm model to generate pressure holding control instructions, and sends the pressure holding control instructions to the injection molding machine control system to drive the injection molding machine to perform pressure holding actions. (6) The dynamic pressure holding control module associates and stores the pressure change curve and parameter adjustment record of this pressure holding process into the feedback learning module for subsequent optimization calculations; (7) The human-machine interface obtains the pressure holding time and pressure parameters sent by the dynamic pressure holding control module, and displays the pressure change curve of the injection mold cavity, pressure holding control parameters and molding status in real time, so that the operator can monitor and make manual intervention.

2. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 1, characterized in that: In step (4), the data preprocessing module filters the pressure signal generated by the pressure data using a pressure data filtering algorithm; the pressure data filtering algorithm uses a first-order low-pass filter, and its recursive formula is as follows: ,in: This is the raw pressure value collected at the current moment; The filtered pressure value at the nth sampling point is expressed in MPa, where n is the sampling point number (n=1, 2, ..., n). 'a' is the filter coefficient, 0 ≤ a ≤ 1, and the smaller 'a' is, the smoother the filter. This is the filtered pressure value at the (n-1)th sampling point, in MPa.

3. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 2, characterized in that: In step (5), the slope detection method and the local maximum detection method are used in combination to identify the inflection point in the pressure change curve that reaches the pressure holding stage. The operation steps are as follows: (5-1) The slope detection method is achieved by calculating the first derivative of continuous pressure data points, i.e.: , in: The slope of the nth sampling point relative to the (n-1)th sampling point; This is the filtered pressure value at the nth sampling point, in MPa. This is the filtered pressure value at the (n-1)th sampling point, in MPa. If the slope is less than or equal to 0 at several consecutive points, then this point is determined to be the turning point of the pressure holding stage. (5-2) The operation steps of the local maximum detection method are as follows: (5-2.1) Set the search window length: Based on the filling time characteristics of the injection molding process, preset the search window length for local extrema; (5-2.2) Define the rule for determining local maxima: If a certain sampling point If the condition is met that "the pressure is at its maximum value within the current search window, and the rate of change of pressure within the next three consecutive search windows is less than 10% of the current pressure value", then the point is marked as the turning point of the candidate pressure holding stage. The formula for the rate of change of pressure is: , in: The pressure change rate reflects how quickly the pressure changes, and its unit is MPa / s; The sampling period represents the time interval between two consecutive pressure samples. i represents the number of sampling points in the time interval; (5-2.3) Starting from the start of the injection molding process, the filtered pressure values ​​of the mold cavity are traversed by sliding a set search window. : (5-2.3.1) When a sampling point is detected to meet the local maximum determination rule in step (5-2.2), the slope before and after the point is extracted. Perform secondary verification; (5-2.3.2) If the slope before that point is (i=1, 2, ..., 5) greater than 0, the slope thereafter When (i=1, 2, ..., 5) rapidly decreases to less than or equal to 0, then this point is determined to be the turning point of the pressure holding phase; (5-2.3.3) If the verification fails, adjust the search window length by increasing or decreasing the number of sampling points and repeat the local maximum detection process described above.

4. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 1, characterized in that: In step (5), the table-driven strategy table receives the pressure holding parameters optimized by the feedback learning module and outputs the preset parameter matching result to the dynamic pressure holding control module. The regression algorithm model receives the corrected coefficients from the feedback learning module and outputs the real-time parameter calculation results to the dynamic pressure holding control module.

5. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 4, characterized in that: In step (5), the table entries of the table-driven strategy table include material, wall thickness, die time, recommended holding time and holding pressure. The injection molding machine control system automatically matches the corresponding table entries and executes them based on the current sampling parameters.

6. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 4, characterized in that: In step (5), the multivariate linear regression formula in the regression algorithm model is as follows: Holding pressure formula: Hold_Pressure = a1 × Thickness + a2 × Material_Code + a3 × T_fill + b; The formula for holding time is: Hold_Time = c1 × Thickness + c2 × Pressure_Max + d; a1: Regression coefficient for cavity wall thickness; a2: Regression coefficient of Material Code; a3: Regression coefficient for fill time (T_fill); b: Basic adjustment value, a correction item to ensure that the formula calculation results are more in line with actual production; c1: Regression coefficient for cavity wall thickness; c2: Regression coefficient of maximum filling pressure (Pressure_Max); d: Basic adjustment value, a correction item to ensure that the formula calculation results are more in line with actual production.

7. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 1, characterized in that: In step (6), after the pressure holding process is completed, the quality inspection of the molded product is carried out to obtain the key index data of the molded product, and the detected key index data is transmitted to the feedback learning module. The feedback learning module manages and stores the key index data of the molded product, and optimizes and adaptively adjusts it in combination with the pressure change curve and parameter adjustment record of this pressure holding process, and sends the optimized pressure holding control parameters to the dynamic pressure holding control module.

8. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 7, characterized in that: In step (6), after the pressure holding process is completed, the molded product is inspected by the molding quality inspection equipment to obtain the key indicator data of the molded product.

9. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 8, characterized in that: In step (6), the injection molding machine control system first presets the threshold of the deviation coefficient, and then performs multiple regression analysis on the holding time, pressure and key indicator data through the feedback learning module to calculate the deviation coefficient. When the deviation coefficient exceeds the preset threshold, the coefficient of the regression algorithm model is automatically modified and the holding parameters after the table-driven strategy table is updated.

10. The intelligent pressure holding control method for thin-walled injection molded products based on molding feedback as described in claim 1, characterized in that: In step (7), the injection molding machine control system presets the number of molding cycles. After each preset number of molding cycles is completed, the feedback learning module generates a parameter optimization report and prompts the operator through the human-machine interface to confirm whether to enable the new parameters. If the pressure in the injection mold cavity suddenly rises to more than 120% or the pressure sensor communication is interrupted during the pressure holding process, the dynamic pressure holding control module immediately triggers the emergency mechanism: suspends the pressure holding action, issues an alarm signal through the human-machine interface, and automatically calls the historical optimal parameters for a reset attempt. After the pressure holding abnormality is resolved, the injection molding machine is re-driven to perform the pressure holding action. In step (7), the human-machine interface includes a real-time pressure curve display area, a pressure holding control parameter setting area, an alarm and status display area, and a control operation button area, which allows for visual monitoring and parameter adjustment of the injection molding process. The pressure change curve of the injection mold cavity in step (7) refers to the curve of pressure change in the injection mold cavity over time; the pressure change curve of the injection mold cavity includes the filling stage, the holding stage and the cooling stage.

Images