A method and system for managing the manufacturing process of an electric control box assembly

Abstract

The application relates to the technical field of injection molding manufacturing, and discloses an electric control box assembly manufacturing process management method and system, which comprises the following steps: extracting pressure characteristic values of a current production mode based on pressure time sequence data, and calling pressure characteristic values of corresponding historical production modes; performing applicable boundary judgment and deviation source judgment on the current production mode according to the pressure characteristic values of the current production mode and the pressure characteristic values of the historical production modes; when the current production mode is within the applicable boundary and the deviation source is material property change or injection molding mold failure, a first risk result is generated; when the current production mode is not within the applicable boundary or the deviation source cannot be determined, a second risk result is generated; and a control action for the injection molding process is selected according to the first risk result or the second risk result, a characteristic deviation is identified through a double judgment mechanism, risk underestimation caused by model misjudgment on out-of-distribution data is avoided, and self-adaptive quality management and control of the injection molding process are realized.

Description

Technical Field

[0001] This application relates to the field of injection molding manufacturing technology, and more specifically, to a method and system for managing the manufacturing process of an electrical control box assembly. Background Technology

[0002] In the continuous injection molding production of electrical control box components, to achieve real-time quality control, distributed sensor arrays are typically used to collect pressure distribution and melt flow status. This data is then combined with a machine learning model trained on a historical process database to predict the probability of defects such as shrinkage marks. A process window optimization algorithm is then used to dynamically adjust the injection rate and mold opening / closing sequence. However, in actual production, injection molding plants often switch raw material batches due to supply adjustments. When the melt index of the new batch of raw material is significantly higher than that of historically commonly used batches, resulting in a significant decrease in melt viscosity, the pressure feature values ​​extracted by edge computing shift downwards, falling into the long-tailed, low-density region of the historical training sample distribution. Because the prediction function of the fixed model in this sparse region has significant uncertainty, it tends to map low pressure features to low shrinkage marks, underestimating the risk of shrinkage marks. Upon receiving this underestimated risk probability, the constraint penalty term is reduced, resulting in ineffective constraint of the injection rate and insufficient holding pressure. This leads to insufficient melt feeding in thick-walled areas such as the roots of reinforcing ribs during the cooling stage, ultimately causing a concentration of shrinkage mark defects. Existing technologies lack adaptive sensing and calibration mechanisms for input distribution shifts, making it difficult to identify the decrease in model confidence when features shift, leading to the failure of process parameter adjustments.

[0003] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application provides a method and system for managing the manufacturing process of electrical control box components. This method and system can solve the technical problem that pressure characteristic shifts caused by raw material batch switching lead to underestimated shrinkage risk and ultimately cause the failure of process parameter adjustments.

[0005] In a first aspect, this application provides a method for managing the manufacturing process of an electronic control box assembly, including: Obtain the pressure timing data of the current production cycle in the injection molding process of the electrical control box assembly, extract the pressure feature value of the current production cycle based on the pressure timing data, and retrieve the pressure feature value of the corresponding historical production cycle from the historical process database. Based on the pressure characteristic value of the current production module and the pressure characteristic value of the historical production modules, the current production module is subjected to application boundary judgment and offset source judgment. The application boundary judgment is used to determine whether the pressure characteristic value of the current production module deviates from the benchmark characteristic range corresponding to the pressure characteristic value of the historical production module. The offset source judgment is used to determine the offset source of the characteristic change of the pressure characteristic value of the current production module relative to the pressure characteristic value of the historical production module. When the current production cycle is within the applicable boundary and the source of the deviation is a change in the material properties or a failure of the injection molding die, a first risk result is generated; when the current production cycle is not within the applicable boundary or the source of the deviation cannot be determined, a second risk result is generated. Select control actions for the injection molding process based on the first or second risk outcome.

[0006] Optionally, the step of acquiring pressure timing data during the injection molding process of the electrical control box assembly in the current production cycle, and extracting pressure feature values ​​for the current production cycle based on the pressure timing data, includes: Pressure timing data during the holding phase is acquired by a distributed pressure sensor array deployed within the injection molding mold cavity. The pressure maintenance rate during the pressure holding phase is calculated based on pressure time series data and used as the pressure characteristic value of the current production cycle. The pressure maintenance rate is the ratio of the pressure value at the end of the pressure holding phase to the pressure value at the beginning of the pressure holding phase.

[0007] Optionally, the step of retrieving the pressure characteristic values ​​of the corresponding historical production modules from the historical process database includes: Obtain the injection mold number and the raw material grade information for the current production batch; Based on the mold number information and raw material grade information, load the historical production cycles and pressure maintenance rates of the historical production cycles that match the mold number information and raw material grade information from the historical process database.

[0008] Optionally, the applicable boundary determination can be performed through the following steps: Obtain the baseline characteristic range corresponding to the pressure characteristic values ​​of historical production cycles, where the baseline characteristic range is the numerical range of the pressure maintenance rate of historical production cycles. If the pressure maintenance rate of the current production cycle is within the baseline characteristic range, then the current production cycle is determined to be within the applicable boundary; otherwise, the current production cycle is determined not to be within the applicable boundary.

[0009] Optionally, the offset source determination can be performed through the following steps: Obtain the pressure difference between the gate area and the root area of ​​the reinforcing rib in the injection mold cavity during the current production cycle; The pressure difference and pressure maintenance rate are compared with the baseline pressure difference and baseline pressure maintenance rate of historical production cycles, respectively. When the scaling ratio of the pressure difference relative to the reference pressure difference does not exceed the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate does not exceed the second preset range, the source of the deviation is determined to be a change in the physical properties of the raw materials. When the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, the source of the offset is determined to be a fault in the injection molding mold. When the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first preset range, but the difference between the pressure maintenance rate and the reference pressure maintenance rate does not exceed the second preset range, or when the scaling ratio of the pressure difference relative to the reference pressure difference does not exceed the first preset range, but the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, it is determined that the source of the offset cannot be determined.

[0010] Optionally, when the current production cycle is within the applicable boundaries and the offset originates from changes in raw material properties or injection mold failure, the steps for generating the first risk result include: When the source of the offset is a change in the physical properties of raw materials, the first preset risk value assigned to the current production module is obtained as the product defect risk estimate. When the source of the offset is a failure of the injection molding mold, the second preset risk value assigned to the current production batch is obtained as the product defect risk estimate, wherein the second preset risk value is greater than the first preset risk value; The estimated product defect risk is output as the first risk result, which includes a preset high confidence level marker and an offset marker indicating the source of the offset.

[0011] Optionally, when the current production cycle is not within the applicable boundaries or the source of the offset cannot be determined, the step of generating a second risk result includes: Calculate the deviation of the pressure maintenance rate of the current production cycle from the baseline characteristic range; The risk compensation value is calculated based on the difference between the degree of deviation and the preset safety deviation threshold, where the risk compensation value is positively correlated with the difference. The product defect risk estimate is obtained by adding the first preset risk value to the risk compensation value. The estimated product defect risk is output as the second risk result, which is marked with a preset low confidence level.

[0012] Optionally, the steps for selecting control actions for the injection molding process based on the first risk outcome or the second risk outcome include: When the first risk result is marked with a high confidence level and the offset mark indicates that the offset source is a change in the physical properties of the raw materials, the preset conventional multi-objective optimization control action is executed to optimize the injection molding cycle and energy consumption. When the first risk result is marked with a high confidence level and the offset mark indicates that the offset source is a failure of the injection molding die, an equipment-level abnormal response action is executed, and a stop command is issued to the injection molding machine controller; When the second risk result is marked with a low confidence level, a conservative compensation control action is performed, extending the holding time and increasing the holding pressure setting value.

[0013] Optionally, during the execution of conservative compensation control actions, the following may also be included: Obtain the product weight detection value for the current production batch; The average residual pressure in the root region of the reinforcing rib during the pressure holding stage is obtained within a preset time window, as well as the instantaneous pressure decay time constant at the end of the pressure holding stage. When the product weight detection value exceeds the preset standard tolerance range, the conservative compensation control action is maintained, and the holding pressure setting value is increased in the next production cycle. When the product weight detection value is within the preset standard tolerance range, and the average residual pressure is lower than the preset effective compaction threshold or the pressure decay time constant exceeds the preset healthy range, the conservative compensation control action is maintained, and the pressure holding time is extended in the next production cycle. When the product weight detection value is within the preset standard tolerance range, and the average residual pressure is not lower than the preset effective compaction threshold, and the pressure decay time constant is within the preset healthy range, the actual product quality feedback data corresponding to the preset number of consecutive production cycles is obtained. If the actual product quality feedback data meets the preset safe molding conditions, the pressure characteristic values ​​corresponding to the preset number of consecutive production cycles are integrated into the historical process database.

[0014] Secondly, this application also discloses an electronic control box assembly manufacturing process management system for performing the method described above, including: The data acquisition unit is used to acquire the pressure timing data of the current production cycle in the injection molding process of the electrical control box assembly, extract the pressure feature value of the current production cycle based on the pressure timing data, and retrieve the pressure feature value of the corresponding historical production cycle from the historical process database. The judgment unit is used to perform applicable boundary judgment and offset source judgment on the current production module based on the pressure characteristic value of the current production module and the pressure characteristic value of the historical production module. The applicable boundary judgment is used to determine whether the pressure characteristic value of the current production module deviates from the reference characteristic range corresponding to the pressure characteristic value of the historical production module. The offset source judgment is used to determine the offset source of the characteristic change of the pressure characteristic value of the current production module relative to the pressure characteristic value of the historical production module. The risk result generation unit is used to generate a first risk result when the current production cycle is within the applicable boundary and the source of the deviation is a change in the physical properties of the raw material or a failure of the injection molding die; and to generate a second risk result when the current production cycle is not within the applicable boundary or the source of the deviation cannot be determined. The control unit is used to select control actions for the injection molding process based on either the first risk outcome or the second risk outcome.

[0015] This application introduces feature extraction and application boundary judgment of pressure time series data during injection molding, which can detect feature offset caused by changes in raw material properties or mold failure in real time. Through a dual judgment mechanism, it can distinguish the source of feature offset and perform risk compensation calibration for out-of-distribution data. At the same time, by introducing a closed-loop feedback mechanism of cross-validation of macro and micro information, it can achieve adaptive updating of benchmark features. Under the premise of ensuring production quality, it realizes dynamic optimization and closed-loop control of process parameters, which significantly improves the stability and reliability of the manufacturing process of electrical control box components. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for managing the manufacturing process of an electrical control box assembly, as provided in an embodiment of this application.

[0017] Figure 2 This is a schematic diagram of the structure of a manufacturing process management system for an electrical control box assembly provided in an embodiment of this application.

[0018] Labeling explanation: 210, data acquisition unit; 220, judgment unit; 230, risk result generation unit; 240, control unit. Detailed Implementation

[0019] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally 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 this 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.

[0020] It should be noted that similar reference numerals 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. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] In the continuous injection molding production of electrical control box components, to achieve real-time quality control, sensors are typically used to collect pressure distribution and melt flow state during the molding process. Feature vectors are extracted and input into a machine learning model trained offline based on a historical process database to output the probability of defects such as shrinkage marks. However, in actual production, injection molding plants frequently switch raw material batches due to supply adjustments or cost control. When the melt index of the newly changed batch is significantly higher than that of commonly used batches in historical production, resulting in a significant decrease in melt viscosity, the melt flowability at the same injection rate is enhanced, leading to an overall decrease in pressure within the mold cavity, especially in areas far from the gate, such as the roots of reinforcing ribs. The extracted pressure feature values ​​shift downwards, falling into the long-tailed, low-density region of the historical training sample distribution. Because the prediction function of existing machine learning models in this sparse region has significant uncertainty, fixed models tend to map low pressure features to low shrinkage mark probabilities, underestimating the risk of shrinkage marks. Upon receiving the underestimated shrinkage mark probability, the injection rate is no longer forcibly reduced; in fact, the rate may even be increased, and the triggering of extended holding pressure may be canceled. Because the injection rate was not effectively controlled and the holding pressure was insufficient, thick-walled areas such as the root of the reinforcing ribs did not receive enough melt feeding during the cooling stage, which ultimately led to the concentrated occurrence of shrinkage defects in critical areas.

[0022] Regarding this, firstly, see... Figure 1 This application provides a method for managing the manufacturing process of an electronic control box assembly, including: S1. Obtain the pressure timing data of the current production cycle in the injection molding process of the electrical control box assembly, extract the pressure characteristic value of the current production cycle based on the pressure timing data, and retrieve the pressure characteristic value of the corresponding historical production cycle from the historical process database. S2. Based on the pressure characteristic value of the current production module and the pressure characteristic value of the historical production modules, perform application boundary judgment and offset source judgment on the current production module. The application boundary judgment is used to determine whether the pressure characteristic value of the current production module deviates from the benchmark characteristic range corresponding to the pressure characteristic value of the historical production module. The offset source judgment is used to determine the offset source of the characteristic change of the pressure characteristic value of the current production module relative to the pressure characteristic value of the historical production module. S3. When the current production cycle is within the applicable boundary and the source of the offset is a change in the material properties or a failure of the injection molding die, a first risk result is generated; when the current production cycle is not within the applicable boundary or the source of the offset cannot be determined, a second risk result is generated. S4. Select control actions for the injection molding process based on the first risk outcome or the second risk outcome.

[0023] Pressure time-series data refers to the numerical sequence of pressure changes inside the cavity that is continuously recorded over time during the injection molding cycle. It is usually acquired continuously according to the injection molding machine control cycle or the sampling cycle of the edge acquisition module. For example, in a production cycle, pressure sensors placed near the gate and near the root of the reinforcing rib will continuously output pressure readings that change over time. These readings arranged in chronological order constitute the pressure time-series data.

[0024] Pressure characteristic values ​​are key numerical indicators extracted from pressure time-series data that represent the melt flow and stress state of the current molding die. Pressure characteristic values ​​can include the pressure retention rate, which directly reflects the compensating ability. In addition to the pressure retention rate, in some auxiliary judgment scenarios, supplementary characteristics can be used such as the pressure difference between the gate area and the reinforcing rib root area, and the average residual pressure at the end of the holding pressure period. For example, if the peak filling values ​​of two dies are similar, but one die experiences a faster pressure decay in the later holding pressure stage, its pressure retention rate is lower, meaning that the compensating ability of that die is weaker.

[0025] The applicable boundary refers to the normal fluctuation range covered by the pressure characteristic values ​​corresponding to stable batches of raw materials in the historical process database. Exceeding this range means that the current operating conditions deviate from historical experience. Stable batches of raw materials here usually refer to the data set formed in historical production stages with stable continuous production results after quality verification, under given molds, raw material grades, and equipment conditions. The applicable boundary is not an arbitrarily given fixed constant, but rather a benchmark characteristic range obtained from the statistics of historical stable samples.

[0026] The source of deviation refers to the fundamental physical cause that causes the current pressure characteristic value to deviate from the historical benchmark. For example, changes in the physical properties of the raw material itself or hardware abnormalities of the molding equipment. Changes in the physical properties of the raw material can be manifested as changes in melt flowability after batch switching of raw materials, changes in water content, or changes in overall flow behavior caused by fine-tuning of the formula. Injection molding mold failures can be manifested as local pressure transmission distortion caused by premature condensation of the gate, poor venting, local blockage, abnormal wear of the runner, etc.

[0027] The first risk result refers to a high-confidence defect occurrence risk assessment output when the operating conditions are within the known applicable range, or when deviations occur but the causes are clear and controllable. It includes not only the risk estimate, but also preferably a confidence level label and a deviation label, so that the subsequent control unit can distinguish whether to continue with routine optimization or directly execute the device-level response.

[0028] The second risk outcome refers to the risk assessment marked with uncertainty when the operating conditions deviate from the known applicable range, or when the cause of the deviation is unknown and causes the original prediction model to fail.

[0029] In a preferred embodiment, on an injection molding production line for an electrical control box assembly, multiple distributed pressure sensors are deployed within the cavity of the injection mold, such as in the gate area and the root area of ​​the reinforcing ribs. During the holding pressure phase of each production cycle, these sensors collect pressure timing data in real time. Based on this pressure timing data, the pressure maintenance rate during the holding pressure phase (the ratio of the final pressure value to the initial pressure value) is calculated as the pressure characteristic value of the current production cycle.

[0030] Simultaneously, the system acquires the current mold number (e.g., M001) and the currently used raw material grade (e.g., PP-Grade A). Based on this information, it retrieves the pressure maintenance rate data from the historical process database for historical production cycles matching the M001 mold and PP-Grade A raw material, using this data as a benchmark. The numerical range of the historical production cycle pressure maintenance rate is then used as a benchmark characteristic range. If the pressure maintenance rate of the current production cycle falls within this benchmark characteristic range, the current cycle is determined to be within the applicable boundary.

[0031] Simultaneously, the pressure difference between the gate region and the reinforcing rib root region in the current batch is acquired, and this pressure difference and the pressure maintenance rate of the current batch are compared with the baseline pressure difference and baseline pressure maintenance rate of historical production batches, respectively. If the scaling ratio of the pressure difference relative to the baseline pressure difference does not exceed a preset first range (e.g., ±5%), and the difference between the pressure maintenance rate and the baseline pressure maintenance rate does not exceed a preset second range (e.g., ±2%), then the source of the deviation is determined to be a change in the physical properties of the raw materials.

[0032] If the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first range, and the difference between the pressure holding rate and the reference pressure holding rate exceeds the second range, then the source of the offset is determined to be a failure of the injection molding mold.

[0033] If other combinations of conditions occur, the source of the offset cannot be determined.

[0034] When the current module is within the applicable boundary and the source of the offset is the change in raw material properties, a first preset risk value (e.g., 0.3) is assigned as the product defect risk estimate, and a first risk result with a high confidence level label and a raw material property change offset label is generated.

[0035] When the current mold is within the applicable boundary and the source of the offset is a failure of the injection molding mold, a second preset risk value (e.g., 0.7, greater than 0.3) is assigned as the product defect risk estimate, and a first risk result with a high confidence level mark and a mold failure offset mark is generated.

[0036] When the current module is not within the applicable boundaries or the source of the deviation cannot be determined, the deviation of the pressure maintenance rate of the current module from the benchmark characteristic range is calculated. The risk compensation value is calculated based on the difference between the deviation degree and the safety deviation threshold, and then added to the first preset risk value to obtain the product defect risk estimate, generating a second risk result marked with a low confidence level.

[0037] If the first risk result indicates a change in raw material properties, perform routine multi-objective optimization control actions, such as fine-tuning the injection rate and holding time, to optimize the injection molding cycle and energy consumption.

[0038] If the first risk result indicates a mold failure, execute the equipment-level abnormal response action and issue a stop command to the injection molding machine controller for mold inspection and repair.

[0039] If the second risk result is marked with a low confidence level, implement conservative compensation control actions, such as extending the holding time by 5% and increasing the holding pressure setting by 10%, to reduce the risk of shrinkage marks.

[0040] During the implementation of conservative compensation controls, continuous monitoring is conducted: Product weight detection value: If it exceeds the preset standard tolerance range, maintain conservative compensation control and continue to increase the holding pressure setting value in the next cycle.

[0041] Mean residual pressure and instantaneous pressure decay time constant in the root region of the reinforcing rib: If the product weight is within the tolerance range, but the mean residual pressure is lower than the effective compaction threshold or the pressure decay time constant exceeds the healthy range, maintain conservative compensation control and continue to extend the holding time in the next cycle.

[0042] If the product weight is within tolerance and the average residual pressure and pressure decay time constant are normal, obtain actual product quality feedback data for five consecutive molding cycles. If these data meet the safe molding conditions, then integrate the pressure characteristic values ​​of these five cycles into the historical process database and update the baseline data.

[0043] Through the above technical solution, this application addresses the problem in the continuous injection molding production of electrical control box components where, due to a significant decrease in melt viscosity caused by raw material batch switching, the pressure feature values ​​extracted by edge computing shift downwards, falling into the long-tailed low-density interval of the historical training sample distribution. This results in significant uncertainty in the prediction function of the fixed model in this sparse interval, tending to map low pressure features as low shrinkage probability, underestimating the risk of shrinkage, and consequently causing the failure of process parameter adjustments. By acquiring pressure time-series data in real time and extracting pressure feature values, combined with historical process databases for applicable boundary judgment and offset source judgment, abnormal situations in the production process can be accurately identified, and raw material property changes and mold failures can be distinguished. Therefore, targeted control actions can be selected based on different risk outcomes, avoiding the underestimation of shrinkage risk and the failure of process parameter adjustments due to model prediction uncertainty. By dynamically adjusting the holding time, holding pressure setpoint, or taking measures such as shutdown, the challenges brought by raw material batch switching are effectively addressed, ensuring the production quality and efficiency of electrical control box components.

[0044] Furthermore, the steps of acquiring pressure timing data during the injection molding process of the electrical control box assembly in the current production cycle, and extracting pressure feature values ​​for the current production cycle based on the pressure timing data, include: Pressure timing data during the holding phase is acquired by a distributed pressure sensor array deployed within the injection molding mold cavity. The pressure maintenance rate during the pressure holding phase is calculated based on pressure time series data and used as the pressure characteristic value of the current production cycle. The pressure maintenance rate is the ratio of the pressure value at the end of the pressure holding phase to the pressure value at the beginning of the pressure holding phase.

[0045] Specifically, multiple pressure sensors are installed inside the cavity of the injection mold. These sensors are distributed in an array to monitor and collect pressure change data of the melt in real time during the holding pressure stage. Piezoelectric or piezoresistive sensors can be used, which convert pressure signals into electrical signals and record them through a data acquisition system to form pressure time-series data. For example, sensors can be placed at key locations such as the gate area, the end of the runner, and thick-walled areas of the cavity to obtain comprehensive pressure distribution information.

[0046] The pressure maintenance rate during the pressure holding phase is calculated using the collected pressure time-series data and used as the pressure characteristic value for the current production cycle. Specifically, this can be achieved as follows: First, the initial and final pressure values ​​of the pressure holding phase are determined from the pressure time-series data. The initial pressure value typically refers to the instantaneous pressure at the beginning of the pressure holding phase or the average pressure at a preset time point during the initial phase; the final pressure value refers to the instantaneous pressure at the end of the pressure holding phase or the average pressure at a preset time point during the final phase. Then, the final pressure value is divided by the initial pressure value to obtain the pressure maintenance rate. For example, if the initial pressure during pressure holding is 100 MPa and the final pressure during pressure holding is 80 MPa, the pressure maintenance rate is 0.8. It should be noted that when calculating the pressure maintenance rate during the pressure holding phase, you can first calculate the pressure maintenance rate of each pressure sensor based on the pressure time series data of each pressure sensor, and then use the mean, weighted average, or maximum value of these pressure maintenance rates as the effective pressure maintenance rate; or, you can first calculate the average pressure time series data, weighted average pressure time series data, or maximum pressure time series data of each pressure sensor as the effective pressure time series data, and then calculate the pressure maintenance rate based on the effective pressure time series data.

[0047] Through the above technical solution, this application solves the problems of how to specifically acquire pressure time-series data and extract pressure characteristic values ​​based on it, as well as how to define and calculate these pressure characteristic values. By deploying a distributed pressure sensor array within the injection molding mold cavity, the pressure changes of the melt during the holding pressure stage can be monitored in real time and at multiple points, thereby acquiring high-quality pressure time-series data. By defining the ratio of the final pressure value to the initial pressure value during the holding pressure stage as the pressure maintenance rate and using it as the pressure characteristic value, the degree of pressure decay during the holding pressure stage can be intuitively and accurately reflected. This provides a reliable basis for subsequent judgment of applicable boundaries and offset sources, effectively avoiding judgment errors caused by inaccurate characteristic value extraction, thereby improving the accuracy and reliability of injection molding process management.

[0048] Furthermore, the steps for retrieving the pressure characteristic values ​​of the corresponding historical production modules from the historical process database include: Obtain the injection mold number and the raw material grade information for the current production batch; Based on the mold number information and raw material grade information, load the historical production cycles and pressure maintenance rates of the historical production cycles that match the mold number information and raw material grade information from the historical process database.

[0049] Firstly, by obtaining the injection mold number and the raw material grade information for the current production batch, the key production conditions for that batch were identified. Mold number and raw material grade information are crucial factors affecting the injection molding process and product quality; different molds and raw material grades can lead to different pressure characteristics.

[0050] Secondly, based on the obtained mold number and raw material grade information, historical production cycles and their pressure maintenance rates that match this information are loaded from the historical process database. This precise matching based on mold number and raw material grade ensures a high degree of consistency between the retrieved historical data and the current production cycle in terms of production conditions. This provides a more reliable benchmark for subsequent determination of applicable boundaries and sources of deviation, effectively avoiding judgment biases caused by mismatched historical data and improving the accuracy and effectiveness of the entire management method.

[0051] Obtaining the injection mold number information refers to the unique identifier that identifies the injection mold currently in use. Specifically, this can be achieved by reading the mold ID from the injection molding machine controller or production management system, or by scanning the QR code or RFID tag on the mold.

[0052] Obtaining the raw material grade information for the current production module refers to identifying the model or brand of the raw materials used in the current production batch. This can be achieved by querying the raw material information of the current work order from the material management system or MES system, or by having the operator manually enter or scan the barcode on the raw material packaging.

[0053] The historical process database refers to a database that stores a large amount of key process parameters and product quality data from past production cycles. Specifically, it can be implemented using a relational database (such as MySQL or PostgreSQL) or a non-relational database (such as MongoDB). The database should include fields such as mold number, raw material grade, and pressure holding rate.

[0054] Therefore, by obtaining mold number information and raw material grade information, historical data that is highly consistent with the production conditions of the current production cycle can be accurately located. Through the above technical solution, this application solves the problem that when retrieving pressure characteristic values ​​from the historical process database, simply retrieving the pressure characteristic values ​​of the corresponding historical production cycle without considering the compatibility of the mold and raw materials may lead to a mismatch between the retrieved historical data and the current production cycle, thus affecting the accuracy of the judgment and causing deviations in subsequent risk assessment and control actions.

[0055] Further, the applicable boundary determination is performed through the following steps: Obtaining a reference feature range corresponding to pressure characteristic values of historical production molds, wherein the reference feature range is a numerical interval of pressure maintenance rates of historical production molds; If the pressure maintenance rate of the current production mold is within the reference feature range, it is determined that the current production mold is within the applicable boundary; otherwise, it is determined that the current production mold is not within the applicable boundary.

[0056] Wherein, obtaining the reference feature range corresponding to the pressure characteristic values of historical production molds refers to determining a numerical interval, which represents the fluctuation range of the pressure maintenance rate of historical production molds under normal production conditions. This can be specifically implemented by statistical methods. For example, a large amount of pressure maintenance rate data of historical production molds can be collected, then the average value and standard deviation of these data are calculated, and the average value plus or minus a certain multiple of the standard deviation is taken as the reference feature range. For example, the three-standard-deviation principle can be adopted. A percentile method can also be used, for example, the 5th percentile and 95th percentile of historical data are taken as the reference feature range. In addition, machine learning models, such as clustering algorithms, can also be used to analyze historical data, identify the pressure maintenance rate distribution of normal production molds, and determine the reference feature range based on this.

[0057] That the reference feature range is the numerical interval of the pressure maintenance rate of historical production molds refers to defining the specific form of the reference feature range, that is, an interval composed of a minimum value and a maximum value. It can be specifically defined in the following manner: for example, it can be set as [P_min,P_max], wherein P_min is the minimum value of the pressure maintenance rate of historical production molds, and P_max is the maximum value of the pressure maintenance rate of historical production molds. Alternatively, it can be set as [μ-kσ,μ+kσ], wherein μ is the average value of the pressure maintenance rate of historical production molds, σ is the standard deviation, and k is a preset coefficient, for example, k=2 or k=3.

[0058] That if the pressure maintenance rate of the current production mold is within the reference feature range, it is determined that the current production mold is within the applicable boundary refers to comparing the pressure maintenance rate of the current production mold with the predetermined reference feature range. This can be specifically implemented by means of numerical comparison. For example, if the pressure maintenance rate P_current of the current production mold satisfies P_min≤P_current≤P_max, it is determined that the current production mold is within the applicable boundary. That otherwise it is determined that the current production mold is not within the applicable boundary refers to that when the pressure maintenance rate of the current production mold does not satisfy the condition of being within the reference feature range, that is, P_current<P_min or P_current>P_max, the current production mold is considered to deviate from the normal production state.

[0059] As a preferred embodiment, on an injection molding production line for an electrical control box assembly, assuming that by analyzing data from the past 1000 normal production cycles, the average pressure maintenance rate during the holding phase of these cycles is calculated to be 0.85, with a standard deviation of 0.02. To set a reasonable baseline characteristic range, the principle of mean ± 2 times standard deviation can be adopted, that is, the baseline characteristic range is [0.85-2*0.02, 0.85+2*0.02], or [0.81, 0.89].

[0060] During injection molding in the current production cycle, a distributed pressure sensor array deployed within the mold cavity collects real-time pressure timing data during the holding pressure phase. Based on this data, the pressure maintenance rate for the current production cycle is calculated. For example, if the final pressure value during the holding pressure phase of the current production cycle is 50 MPa and the initial pressure value is 60 MPa, then the pressure maintenance rate is approximately 50 / 60 ≈ 0.833.

[0061] Next, the calculated pressure maintenance rate of the current production cycle, 0.833, is compared with the preset baseline characteristic range [0.81, 0.89]. Since 0.833 is between 0.81 and 0.89, i.e., 0.81≤0.833≤0.89, the current production cycle is determined to be within the applicable boundary.

[0062] If the pressure maintenance rate of another production cycle is calculated to be 0.79, then since 0.79 < 0.81, the pressure maintenance rate of this cycle is not within the baseline characteristic range, and therefore it is determined that this production cycle is not within the applicable boundary.

[0063] The above technical solution clarifies that the benchmark feature range is the numerical range of the pressure maintenance rate of historical production modules, and directly compares whether the pressure maintenance rate of the current production module falls within this range, thus achieving a rapid and objective judgment of the production status.

[0064] Further, the offset source is determined through the following steps: Obtain the pressure difference between the gate area and the root area of ​​the reinforcing rib in the injection mold cavity during the current production cycle; The pressure difference and pressure maintenance rate are compared with the baseline pressure difference and baseline pressure maintenance rate of historical production cycles, respectively. When the scaling ratio of the pressure difference relative to the reference pressure difference does not exceed the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate does not exceed the second preset range, the source of the deviation is determined to be a change in the physical properties of the raw materials. When the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, the source of the offset is determined to be a fault in the injection molding mold. When the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first preset range, but the difference between the pressure maintenance rate and the reference pressure maintenance rate does not exceed the second preset range, or when the scaling ratio of the pressure difference relative to the reference pressure difference does not exceed the first preset range, but the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, it is determined that the source of the offset cannot be determined.

[0065] The acquisition of the pressure difference between the gate area and the reinforcing rib root area within the injection mold cavity for the current production cycle refers to real-time monitoring of the pressure in these two specific areas using sensors inside the mold, and calculating the instantaneous pressure difference. This can be achieved by deploying high-precision pressure sensors in both the gate area and the reinforcing rib root area. For example, piezoelectric or strain gauge pressure sensors can be used. These sensors convert pressure signals into electrical signals, which are then synchronously acquired and the difference calculated by a data acquisition system. The gate area is the initial point where the melt enters the mold cavity, and its pressure directly reflects the injection pressure and melt flow resistance. The reinforcing rib root area is typically a thick-walled portion of the product structure, sensitive to pressure holding effects and shrinkage defects; its pressure reflects the filling and pressure holding state of the melt deep within the mold cavity.

[0066] Comparing the pressure difference and pressure maintenance rate with the baseline pressure difference and baseline pressure maintenance rate of historical production cycles means comparing the pressure difference and pressure maintenance rate measured in the current production cycle with baseline values ​​representing stable production states obtained from the historical process database. Specifically, the pressure difference can be compared with the historical baseline pressure difference by scaling, while the pressure maintenance rate is compared with the historical baseline pressure maintenance rate by difference. The historical process database stores a large amount of pressure data accumulated under normal production conditions. By statistically analyzing this data, the range or average value of the baseline pressure difference and baseline pressure maintenance rate can be established.

[0067] The first and second preset ranges are thresholds set based on historical data and process experience to distinguish characteristics from different sources of offset. Specifically, the first preset range can be set as, for example, a scaling factor of ±5%, and the second preset range can be set as, for example, a difference of ±2%. This judgment logic is based on the following principle: changes in raw material properties, such as slight changes in melt index or viscosity, usually lead to a uniform shift in the overall pressure level within the mold cavity, but do not significantly alter the melt flow pattern or local pressure distribution differences within the mold cavity. Therefore, the relative change in pressure difference is small, while the pressure maintenance rate may show a slight but uniform shift.

[0068] When the scaling factor of the pressure difference relative to the reference pressure difference exceeds a first preset range, and the difference between the pressure holding rate and the reference pressure holding rate exceeds a second preset range, the source of the deviation is determined to be a mold malfunction in the injection molding process. Specifically, when the scaling factor of the pressure difference exceeds, for example, ±10%, and the difference in the pressure holding rate exceeds, for example, ±5%, it can be determined to be a mold malfunction. This judgment logic is based on the following principle: mold malfunctions, such as cavity wear, gate blockage, runner deformation, or sensor damage, directly affect the flow path and pressure transmission of the melt within the mold cavity, leading to significant changes in local pressure distribution, thereby causing significant changes in the pressure difference. At the same time, mold malfunctions may also affect the overall pressure holding effect, resulting in a significant abnormality in the pressure holding rate.

[0069] Specifically, when the pressure differential changes significantly while the pressure maintenance rate does not, or vice versa, this indicates that the current deviation pattern does not fully conform to the typical characteristics of changes in raw material properties or mold failure. For example, there may be a combination of factors, or a new anomaly not identified by the model. Under such uncertainty, instead of forcibly attributing the cause, it is marked as uncertain to avoid misjudgment and provide a more cautious basis for subsequent risk assessment and control.

[0070] As a preferred embodiment, on an injection molding production line for an electrical control box assembly, to accurately identify the source of deviation in the pressure characteristic value of each production cycle, piezoelectric pressure sensors are first installed in the gate area and the reinforcing rib root area within the injection mold cavity. These sensors collect pressure data in real time during the current production cycle. For example, during the holding pressure stage, the pressure measured in the gate area is 80 MPa, and the pressure measured in the reinforcing rib root area is 60 MPa, from which a pressure difference of 20 MPa is calculated. Simultaneously, the pressure maintenance rate for the current cycle is calculated, for example, to be 0.75.

[0071] Subsequently, baseline data for historical production cycles matching the current mold number and raw material grade are retrieved from the historical process database. The historical baseline pressure difference is assumed to be 22 MPa, and the baseline pressure maintenance rate is 0.78.

[0072] Next, the offset source determination will be performed: First, calculate the scaling factor of the current pressure difference relative to the reference pressure difference. The current pressure difference is 20 MPa, and the reference pressure difference is 22 MPa. The scaling factor is (20-22) / 22 = -9.09%.

[0073] Then, calculate the difference between the current pressure maintenance rate and the reference pressure maintenance rate. The current pressure maintenance rate is 0.75, and the reference pressure maintenance rate is 0.78. The difference is 0.75 - 0.78 = -0.03.

[0074] Set the first preset range to ±5% and the second preset range to ±0.02.

[0075] In this case, the scaling factor of the pressure difference -9.09% exceeds the first preset range of ±5%. The difference in pressure maintenance rate -0.03 also exceeds the second preset range of ±0.02.

[0076] According to the judgment logic, when the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, the source of the offset is determined to be a fault in the injection molding mold.

[0077] Through the above technical solution, this application can more specifically and precisely distinguish between two different sources of deviation: changes in raw material properties and injection molding mold failure. This solves the problem that existing methods cannot clearly determine the source of deviation when performing deviation source judgment, thereby avoiding affecting the generation of subsequent risk results and the selection of control actions.

[0078] Furthermore, when the current production cycle is within the applicable boundaries and the source of the offset is a change in raw material properties or a failure of the injection molding die, the steps for generating the first risk result include: When the source of the offset is a change in the physical properties of raw materials, the first preset risk value assigned to the current production module is obtained as the product defect risk estimate. When the source of the offset is a failure of the injection molding mold, the second preset risk value assigned to the current production batch is obtained as the product defect risk estimate, wherein the second preset risk value is greater than the first preset risk value; The estimated product defect risk is output as the first risk result, which includes a preset high confidence level marker and an offset marker indicating the source of the offset.

[0079] Specifically, when the source of the deviation is determined to be a change in raw material properties, a first preset risk value assigned to the current production cycle is used as the product defect risk estimate. Based on preset experience or models, a relatively low risk assessment is given, because changes in raw material properties can usually be compensated for by adjusting process parameters. The first preset risk value can be an empirical value; for example, based on historical data statistics, when raw material properties change, the probability or severity of product defects is usually within a low range, so a risk value of 0.3 can be set.

[0080] When the source of the offset is determined to be a failure in the injection molding mold, a second preset risk value assigned to the current production batch is used as the estimated product defect risk, and this second preset risk value is greater than the first preset risk value. This reflects that mold failures typically pose a more severe product defect risk than changes in raw material properties, thus requiring a higher risk assessment to prompt more urgent and thorough interventions. For example, when a mold fails, the probability or severity of product defects is usually within a high range, so a second preset risk value of 0.7 can be set.

[0081] Finally, the calculated product defect risk estimate is output as the first risk result, which includes a pre-set high confidence level marker and an offset marker indicating the source of the deviation. The high confidence level marker indicates that the system has a high degree of confidence in the current risk assessment because the source of the deviation has been clearly identified. The offset marker directly indicates whether the root cause of the problem is a change in raw material properties or a mold failure, which is crucial for selecting targeted control actions. For example, if it is a raw material problem, it may be necessary to adjust the injection parameters; if it is a mold failure, it may be necessary to shut down the machine for maintenance.

[0082] Through the above technical solution, this application can accurately assess product defect risks and provide highly reliable risk information and clear indications of deviation sources to guide subsequent control actions.

[0083] Furthermore, when the current production cycle is not within the applicable boundaries or the source of the offset cannot be determined, the steps for generating a second risk result include: Calculate the deviation of the pressure maintenance rate of the current production cycle from the baseline characteristic range; The risk compensation value is calculated based on the difference between the degree of deviation and the preset safety deviation threshold, where the risk compensation value is positively correlated with the difference. The product defect risk estimate is obtained by adding the first preset risk value to the risk compensation value. The estimated product defect risk is output as the second risk result, which is marked with a preset low confidence level.

[0084] Calculating the deviation of the current production cycle's pressure maintenance rate from the baseline characteristic range refers to quantifying the difference between the current production cycle and the historical baseline. Specifically, the deviation can be calculated by comparing the current production cycle's pressure maintenance rate with the baseline characteristic range. For example, the distance between the current pressure maintenance rate and the boundary of the baseline characteristic range can be calculated, or its relative position outside the baseline characteristic range can be calculated.

[0085] The risk compensation value is calculated based on the difference between the degree of deviation and a preset safe deviation threshold. The risk compensation value is positively correlated with the difference, meaning that when a production cycle deviates from the baseline characteristic range, there is higher uncertainty and potential risk. This additional risk is reflected by calculating the risk compensation value. Specifically, this can be achieved by setting a preset safe deviation threshold, which represents an acceptable range of deviation. When the calculated degree of deviation exceeds this safe deviation threshold, the difference between the two is calculated. The risk compensation value can be designed as a linear or non-linear function of this difference, as long as it ensures a positive correlation between the risk compensation value and the difference. For example, a compensation coefficient can be set, and the risk compensation value is obtained by multiplying the difference by this coefficient.

[0086] Adding the initial risk value to the risk compensation value to obtain the product defect risk estimate means combining basic risk with additional risk to form a comprehensive risk assessment. The initial risk value represents the basic risk under normal or controllable conditions; for example, the initial risk value allocated when the source of deviation is a change in raw material properties. Adding this basic risk value directly to the previously calculated risk compensation value yields the product defect risk estimate.

[0087] The estimated product defect risk is output as a second risk outcome, marked with a pre-defined low confidence level. This clearly informs the user of the uncertainty in the current risk assessment. Specifically, this can be achieved by attaching a low confidence level marker along with the estimated product defect risk. This marker can be a text label (e.g., low confidence), color coding (e.g., red or yellow), or a specific numerical identifier. When a production cycle is outside the applicable boundaries or the source of the deviation cannot be determined, the model's predictive ability for the current state decreases. Therefore, marking the risk outcome as low confidence reminds the user to be more cautious when taking control actions or to require further verification and analysis. This improves the transparency and reliability of the entire management approach, avoiding misjudgments caused by blindly trusting risk outcomes.

[0088] As a preferred embodiment, on an injection molding production line for an electrical control box assembly, it is assumed that the historical process database records a baseline characteristic range of pressure maintenance rate for historical production cycles of 0.85 to 0.95 under specific mold numbers and raw material grades. The pressure maintenance rate calculated from the pressure timing data collected by a distributed pressure sensor array during the holding pressure stage of the current production cycle is 0.80.

[0089] First, calculate the deviation of the current pressure maintenance rate from the baseline characteristic range. The current pressure maintenance rate of 0.80 is lower than the lower limit of the baseline characteristic range of 0.85, and the deviation is 0.85-0.80=0.05.

[0090] Secondly, the preset safety deviation threshold is 0.02. The risk compensation value is calculated based on the difference between the deviation level of 0.05 and the safety deviation threshold of 0.02. The difference is 0.05 - 0.02 = 0.03. Assuming the risk compensation value is positively correlated with the difference, for example, risk compensation value = difference * 10, then the risk compensation value is 0.03 * 10 = 0.3.

[0091] Next, assume the first preset risk value is 0.2 (representing the basic defect risk under normal circumstances). Add the first preset risk value of 0.2 to the risk compensation value of 0.3 to obtain the estimated product defect risk value of 0.2 + 0.3 = 0.5.

[0092] Finally, the estimated product defect risk of 0.5 is output as the second risk result, along with a preset low confidence level label. For example, the output result is: Estimated product defect risk: 0.5, Confidence level: Low confidence.

[0093] Through the above technical solution, this application calculates the degree of deviation between the pressure maintenance rate and the benchmark characteristic range, and calculates the risk compensation value based on the difference between the degree of deviation and the preset safety deviation threshold. The first preset risk value and the risk compensation value are added together to obtain the product defect risk estimate, and a second risk result with a low confidence level is output, thereby improving the accuracy and reliability of risk assessment and providing a more accurate basis for subsequent control actions.

[0094] Furthermore, the steps for selecting control actions for the injection molding process based on the first or second risk outcome include: When the first risk result is marked with a high confidence level and the offset mark indicates that the offset source is a change in the physical properties of the raw materials, the preset conventional multi-objective optimization control action is executed to optimize the injection molding cycle and energy consumption. When the first risk result is marked with a high confidence level and the offset mark indicates that the offset source is a failure of the injection molding die, an equipment-level abnormal response action is executed, and a stop command is issued to the injection molding machine controller; When the second risk result is marked with a low confidence level, a conservative compensation control action is performed, extending the holding time and increasing the holding pressure setting value.

[0095] Among them, the control actions for the injection molding process are selected based on the first risk result or the second risk result, aiming to achieve refined and adaptive production management.

[0096] Conventional multi-objective optimization control refers to adjusting injection molding parameters, such as injection speed, holding pressure, holding time, and cooling time, to optimize the injection molding cycle and energy consumption while ensuring product quality. This can be achieved using intelligent optimization algorithms based on genetic algorithms, particle swarm optimization, or neural networks. These algorithms can search for the optimal combination of parameters. For example, when changes in raw material properties lead to a decrease in melt viscosity, optimization algorithms can adjust the injection speed and holding pressure to adapt to the change in raw material, while avoiding unnecessary downtime or overcompensation. This improves production efficiency and reduces energy consumption while maintaining product quality.

[0097] Equipment-level anomaly response refers to immediate emergency measures taken when the system is highly certain that the risk stems from a more serious problem, such as mold failure, which could lead to widespread defects. This can be achieved by sending a specific stop command to the injection molding machine controller. This command immediately interrupts the current injection molding process, preventing further losses and product scrap. For example, when a mold experiences cracks, blockages, or deformation, a stop command is immediately issued to ensure production safety and product quality, avoiding a large number of defective products due to mold failure.

[0098] Conservative compensation control measures refer to measures taken when the reliability of risk outcomes is low or the source of deviation cannot be clearly identified. Specifically, this can be achieved by extending the holding time and increasing the holding pressure setting. For example, the holding time can be extended by 10%-20%, and the holding pressure setting increased by 5%-10%. By extending the holding time and increasing the holding pressure setting, the filling and compaction of the melt into the cavity can be enhanced. This compensates for potential defect risks as much as possible under conditions of high uncertainty, improving the stability of product quality. For instance, when it is unclear whether the abnormal pressure characteristic value is caused by changes in raw material properties or slight mold wear, conservative compensation measures are taken by extending the holding time and increasing the holding pressure to ensure product quality and avoid misjudgments and losses due to insufficient information.

[0099] The above approach allows for detailed differentiation based on the specific circumstances of risk outcomes, ensuring the precision and effectiveness of control actions, thereby achieving more refined process management and effectively avoiding potential product defects.

[0100] As a preferred embodiment, assuming the current production cycle detects a pressure maintenance rate of 0.85, and the corresponding baseline pressure maintenance rate range in the historical process database is 0.88-0.92. First, an application boundary judgment is performed, determining that the current pressure maintenance rate of 0.85 is not within the baseline characteristic range, therefore the current production cycle is not within the application boundary.

[0101] Simultaneously, the pressure difference between the gate region and the reinforcing rib root region in the current production cycle is acquired and compared with the baseline pressure difference in historical production cycles. It is assumed that the scaling ratio of the pressure difference relative to the baseline pressure difference does not exceed a first preset range, but the difference between the pressure maintenance rate and the baseline pressure maintenance rate exceeds a second preset range. In this case, it is determined that the source of the offset cannot be determined.

[0102] Since the current production cycle is not within the applicable boundaries and the source of the offset cannot be determined, a second risk result is generated. This second risk result is marked with a preset low confidence level.

[0103] Based on the second risk result marked with a low confidence level, a conservative compensation control action is selected. Specifically, an instruction is issued to the injection molding machine controller to extend the holding time of the current production cycle from the preset 5 seconds to 6 seconds, and to increase the holding pressure setting from the preset 80MPa to 85MPa.

[0104] During the implementation of conservative compensation control, the product weight is continuously monitored. If the product weight exceeds the preset standard tolerance range, the conservative compensation control is maintained, and the holding pressure setting is increased in the next production cycle. If the product weight is within the preset standard tolerance range, the average residual pressure and instantaneous pressure decay time constant at the root of the reinforcing rib are further monitored during the holding phase to determine whether to extend the holding time or integrate the current process parameters into the historical process database.

[0105] Furthermore, the implementation of conservative compensation control actions also includes: Obtain the product weight detection value for the current production batch; The average residual pressure in the root region of the reinforcing rib during the pressure holding stage is obtained within a preset time window, as well as the instantaneous pressure decay time constant at the end of the pressure holding stage. When the product weight detection value exceeds the preset standard tolerance range, the conservative compensation control action is maintained, and the holding pressure setting value is increased in the next production cycle. When the product weight detection value is within the preset standard tolerance range, and the average residual pressure is lower than the preset effective compaction threshold or the pressure decay time constant exceeds the preset healthy range, the conservative compensation control action is maintained, and the pressure holding time is extended in the next production cycle. When the product weight detection value is within the preset standard tolerance range, and the average residual pressure is not lower than the preset effective compaction threshold, and the pressure decay time constant is within the preset healthy range, the actual product quality feedback data corresponding to the preset number of consecutive production cycles is obtained. If the actual product quality feedback data meets the preset safe molding conditions, the pressure characteristic values ​​corresponding to the preset number of consecutive production cycles are integrated into the historical process database.

[0106] Obtaining the product weight detection value for the current production cycle refers to weighing the injection-molded product using a weighing device to obtain its quality data. This can be achieved using a high-precision electronic scale. For example, an automatic weighing device can be installed next to the injection molding machine. After each cycle, the product is removed by a robotic arm and placed directly on this device for weighing, with the data transmitted to the control system in real time.

[0107] The acquisition of the average residual pressure in the reinforcing rib root region within a preset time window during the pressure holding stage, and the instantaneous pressure decay time constant at the end of the pressure holding stage, refers to real-time monitoring of pressure changes in the reinforcing rib root region during the pressure holding stage using pressure sensors deployed within the mold cavity. The average residual pressure can be obtained by integrating and averaging the pressure data within the preset time window, while the instantaneous pressure decay time constant can be calculated by analyzing the slope of the pressure drop curve at the end of the pressure holding stage or by fitting an exponential decay model. Specifically, piezoelectric or resistive pressure sensors can be used. These sensors are typically installed at specific locations within the mold cavity, such as the reinforcing rib root, to obtain accurate local pressure data.

[0108] Specifically, when the product weight exceeds the preset standard tolerance range, the conservative compensation control action is maintained, and the holding pressure setting is further increased in the next production cycle. This means that when the product weight does not meet the quality requirements, the current conservative compensation strategy continues, and the holding pressure is further increased in the next production cycle. The preset standard tolerance range can be set according to product design requirements and process experience. For example, for an electrical control box assembly weighing 100 grams, the tolerance range may be set to ±0.5 grams.

[0109] Specifically, when the product weight measurement is within the preset standard tolerance range, and the average residual pressure is lower than the preset effective compaction threshold or the pressure decay time constant exceeds the preset healthy range, conservative compensation control is maintained, and the holding time is extended in the next production cycle. This means that even if the product weight is qualified, but there are still problems with the internal compaction effect or mold condition, conservative compensation continues, and the holding time is increased in the next production cycle. The preset effective compaction threshold and healthy range are determined based on historical data and expert experience. For example, the average residual pressure may be set to be no less than 50 MPa, and the pressure decay time constant may be set between 0.5 seconds and 1.5 seconds.

[0110] Specifically, when the product weight is within the preset standard tolerance range, the average residual pressure is not lower than the preset effective compaction threshold, and the pressure decay time constant is within the preset healthy range, the actual product quality feedback data corresponding to a preset number of consecutive production cycles is acquired. If the actual product quality feedback data meets the preset safe molding conditions, the pressure characteristic values ​​corresponding to the preset number of consecutive production cycles are integrated into the historical process database. This means that when the product weight, internal compaction, and mold condition are all good, the actual product quality data of multiple consecutive cycles is further collected for final confirmation. If these data meet the safe molding conditions, the pressure characteristic values ​​of these cycles are updated to the historical process database. The preset number of production cycles can be set to 5 to 10, and the safe molding conditions may include no shrinkage marks, no bubbles, and qualified dimensional accuracy.

[0111] In a preferred embodiment, on an injection molding production line for an electrical control box assembly, when a conservative compensation control action is performed, the holding pressure time is extended to 3 seconds and the holding pressure setpoint is increased to 80 MPa. During this conservative compensation control action, the product weight detection value of the current production batch is acquired in real time. For example, an automatic weighing system located at the end of the production line detects that the product weight of the current batch is 98.2 grams, while the preset standard tolerance range is 100 grams ± 1 gram. Since 98.2 grams exceeds the standard tolerance range (99 grams - 101 grams), the conservative compensation control action is maintained, and the holding pressure setpoint is further increased in the next production batch, for example, increasing the holding pressure setpoint to 82 MPa.

[0112] In subsequent production cycles, assuming the product weight measurement returns to the preset standard tolerance range, for example, a measurement of 99.8 grams, the average residual pressure in the reinforcing rib root region during the pressure holding phase (e.g., 0.5 to 1.5 seconds) and the instantaneous pressure decay time constant at the end of the pressure holding phase are further obtained. Assuming the detected average residual pressure is 45 MPa, while the preset effective compaction threshold is 50 MPa; and the pressure decay time constant is 1.8 seconds, while the preset healthy range is 0.5 to 1.5 seconds, since the average residual pressure is below the effective compaction threshold and the pressure decay time constant exceeds the healthy range, conservative compensation control is maintained, and the pressure holding time is further extended in the next production cycle, for example, to 3.2 seconds.

[0113] After several rounds of adjustments, assuming the product weight stabilizes at 99.9 grams, the average residual pressure reaches 52 MPa (not lower than the effective compaction threshold of 50 MPa), and the pressure decay time constant is 1.2 seconds (within the healthy range of 0.5 to 1.5 seconds), actual product quality feedback data for five consecutive production rounds is obtained. For example, the product surface is checked for shrinkage marks and bubbles through manual visual inspection or machine vision system, and the dimensional accuracy is within tolerance, meeting the preset safe molding conditions. Once these conditions are met, the pressure characteristic values ​​(e.g., pressure maintenance rate during the holding phase) for these five consecutive production rounds are integrated into the historical process database as new stable production parameters for subsequent production reference and optimization.

[0114] Through the above technical solution, this application solves the problem that, during the execution of conservative compensation control actions, simply extending the holding time and increasing the holding pressure setpoint may not be sufficient to accurately determine whether the product quality of the current production cycle has truly improved, nor can it effectively identify when conservative compensation can be stopped or when the current process parameters can be integrated into the historical process database. By introducing multi-dimensional real-time feedback data such as product weight detection values, the average residual pressure in the reinforcing rib root region, and the pressure decay time constant, a refined evaluation and adaptive adjustment of the conservative compensation control action effect is achieved.

[0115] Secondly, see Figure 2 This application also provides a manufacturing process management system for an electronic control box assembly, including: The data acquisition unit 210 is used to acquire the pressure timing data of the current production cycle in the injection molding process of the electrical control box assembly, extract the pressure characteristic value of the current production cycle based on the pressure timing data, and retrieve the pressure characteristic value of the corresponding historical production cycle from the historical process database. The judgment unit 220 is used to perform applicable boundary judgment and offset source judgment on the current production module based on the pressure characteristic value of the current production module and the pressure characteristic value of the historical production module. The applicable boundary judgment is used to determine whether the pressure characteristic value of the current production module deviates from the reference characteristic range corresponding to the pressure characteristic value of the historical production module. The offset source judgment is used to determine the offset source of the characteristic change of the pressure characteristic value of the current production module relative to the pressure characteristic value of the historical production module. The risk result generation unit 230 is used to generate a first risk result when the current production cycle is within the applicable boundary and the source of the offset is a change in the physical properties of the raw material or a failure of the injection molding die; and to generate a second risk result when the current production cycle is not within the applicable boundary or the source of the offset cannot be determined. Control unit 240 is used to select control actions for the injection molding process based on a first risk result or a second risk result.

[0116] Through the collaborative work of the above-mentioned units, accurate perception, reliable diagnosis, and safe control of the injection molding process of the electrical control box assembly were achieved.

[0117] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for managing the manufacturing process of an electrical control box assembly, characterized in that, include: Obtain the pressure timing data of the current production cycle in the injection molding process of the electrical control box assembly, extract the pressure feature value of the current production cycle based on the pressure timing data, and retrieve the pressure feature value of the corresponding historical production cycle from the historical process database; Based on the pressure characteristic value of the current production module and the pressure characteristic value of the historical production module, an application boundary judgment and an offset source judgment are performed on the current production module. The application boundary judgment is used to determine whether the pressure characteristic value of the current production module deviates from the reference characteristic range corresponding to the pressure characteristic value of the historical production module. The offset source judgment is used to determine the offset source of the characteristic change of the pressure characteristic value of the current production module relative to the pressure characteristic value of the historical production module. A first risk result is generated when the current production cycle is within the applicable boundary and the source of the offset is a change in raw material properties or a failure of the injection molding die; a second risk result is generated when the current production cycle is not within the applicable boundary or the source of the offset cannot be determined. Select control actions for the injection molding process based on the first risk result or the second risk result.

2. The method for managing the manufacturing process of an electrical control box assembly according to claim 1, characterized in that, The step of obtaining pressure timing data during the injection molding process of the electrical control box assembly in the current production cycle, and extracting pressure feature values ​​of the current production cycle based on the pressure timing data, includes: Pressure timing data during the holding pressure stage is acquired by a distributed pressure sensor array deployed in the cavity of the injection molding mold. The pressure maintenance rate of the pressure holding stage is calculated based on the pressure time series data as the pressure characteristic value of the current production cycle, wherein the pressure maintenance rate is the ratio of the pressure value at the end of the pressure holding stage to the pressure value at the beginning of the pressure holding stage.

3. The method for managing the manufacturing process of an electrical control box assembly according to claim 2, characterized in that, The step of retrieving the pressure characteristic value of the corresponding historical production module from the historical process database includes: Obtain the injection molding die number and the raw material grade information of the current production batch; Based on the mold number information and the raw material grade information, the historical production cycle and the pressure maintenance rate of the historical production cycle that match the mold number information and the raw material grade information are loaded from the historical process database.

4. The method for managing the manufacturing process of an electrical control box assembly according to claim 3, characterized in that, The applicable boundary determination is performed through the following steps: Obtain the baseline feature range corresponding to the pressure feature value of the historical production cycle, wherein the baseline feature range is the numerical range of the pressure maintenance rate of the historical production cycle; If the pressure maintenance rate of the current production cycle is within the reference characteristic range, then the current production cycle is determined to be within the applicable boundary; otherwise, the current production cycle is determined not to be within the applicable boundary.

5. The method for managing the manufacturing process of an electrical control box assembly according to claim 4, characterized in that, The offset source determination is performed through the following steps: Obtain the pressure difference between the gate area and the root area of ​​the reinforcing rib in the cavity of the injection mold for the current production cycle; The pressure difference and the pressure maintenance rate are compared with the baseline pressure difference and the baseline pressure maintenance rate of the historical production cycle, respectively. When the scaling ratio of the pressure difference relative to the reference pressure difference does not exceed the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate does not exceed the second preset range, it is determined that the source of the offset is the change in the physical properties of the raw material. When the scaling ratio of the pressure difference value relative to the reference pressure difference value exceeds the first preset range, and the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, it is determined that the source of the offset is a fault in the injection molding mold. When the scaling ratio of the pressure difference relative to the reference pressure difference exceeds the first preset range, but the difference between the pressure maintenance rate and the reference pressure maintenance rate does not exceed the second preset range, or when the scaling ratio of the pressure difference relative to the reference pressure difference does not exceed the first preset range, but the difference between the pressure maintenance rate and the reference pressure maintenance rate exceeds the second preset range, it is determined that the source of the offset cannot be determined.

6. The method for managing the manufacturing process of an electrical control box assembly according to claim 5, characterized in that, The step of generating a first risk result when the current production cycle is within the applicable boundary and the offset source is a change in raw material properties or a failure of the injection molding die includes: When the source of the offset is the change in the physical properties of the raw materials, a first preset risk value is obtained for the current production cycle and used as a product defect risk estimate. When the source of the offset is a failure of the injection molding mold, a second preset risk value is obtained for the current production batch and used as the product defect risk estimate, wherein the second preset risk value is greater than the first preset risk value; The estimated product defect risk is output as the first risk result, which includes a preset high confidence level marker and an offset marker indicating the source of the offset.

7. The method for managing the manufacturing process of an electrical control box assembly according to claim 6, characterized in that, The step of generating a second risk result when the current production cycle is not within the applicable boundary or the source of the offset cannot be determined includes: Calculate the degree of deviation between the pressure maintenance rate of the current production cycle and the baseline characteristic range; A risk compensation value is calculated based on the difference between the degree of deviation and a preset safety deviation threshold, wherein the risk compensation value is positively correlated with the difference. The product defect risk estimate is obtained by adding the first preset risk value to the risk compensation value. The estimated product defect risk is output as the second risk result, which is marked with a preset low confidence level.

8. The method for managing the manufacturing process of an electrical control box assembly according to claim 7, characterized in that, The step of selecting control actions for the injection molding process based on the first risk result or the second risk result includes: When the first risk result carries the high confidence level marker and the offset marker indicates that the offset source is a change in raw material properties, a preset conventional multi-objective optimization control action is executed to optimize the injection molding cycle and energy consumption. When the first risk result carries the high confidence level mark and the offset mark indicates that the offset source is a failure of the injection molding die, an equipment-level abnormal response action is executed, and a stop command is issued to the injection molding machine controller; When the second risk result carries the low confidence level flag, a conservative compensation control action is performed, extending the holding time and increasing the holding pressure setting value.

9. The method for managing the manufacturing process of an electrical control box assembly according to claim 8, characterized in that, During the execution of the conservative compensation control action, the following is also included: Obtain the product weight detection value for the current production cycle; The average residual pressure in the root region of the reinforcing rib during the pressure holding stage is obtained, as well as the instantaneous pressure decay time constant at the end of the pressure holding stage. When the product weight detection value exceeds the preset standard tolerance range, the conservative compensation control action is maintained, and the holding pressure setting value is increased in the next production cycle. When the product weight detection value is within the preset standard tolerance range, and the average residual pressure is lower than the preset effective compaction threshold or the pressure decay time constant exceeds the preset healthy range, the conservative compensation control action is maintained, and the pressure holding time is extended in the next production cycle. When the product weight detection value is within the preset standard tolerance range, and the average residual pressure is not lower than the preset effective compaction threshold, and the pressure decay time constant is within the preset healthy range, the actual product quality feedback data corresponding to a preset number of consecutive production cycles is obtained. If the actual product quality feedback data meets the preset safe molding conditions, the pressure characteristic values ​​corresponding to the preset number of consecutive production cycles are integrated into the historical process database.

10. A manufacturing process management system for an electronic control box assembly, used to execute the manufacturing process management method for an electronic control box assembly as described in any one of claims 1 to 9, characterized in that, The system includes: The data acquisition unit is used to acquire the pressure timing data of the current production cycle in the injection molding process of the electrical control box assembly, extract the pressure feature value of the current production cycle based on the pressure timing data, and retrieve the pressure feature value of the corresponding historical production cycle from the historical process database. The judgment unit is used to perform an application boundary judgment and an offset source judgment on the current production module based on the pressure characteristic value of the current production module and the pressure characteristic value of the historical production module. The application boundary judgment is used to determine whether the pressure characteristic value of the current production module deviates from the reference characteristic range corresponding to the pressure characteristic value of the historical production module. The offset source judgment is used to determine the offset source of the characteristic change of the pressure characteristic value of the current production module relative to the pressure characteristic value of the historical production module. The risk result generation unit is used to generate a first risk result when the current production cycle is within the applicable boundary and the source of the offset is a change in the physical properties of the raw material or a failure of the injection molding die; and to generate a second risk result when the current production cycle is not within the applicable boundary or the source of the offset cannot be determined. The control unit is used to select control actions for the injection molding process based on the first risk result or the second risk result.

Images