Intelligent process optimization system and method for mold manufacturing process

Abstract

The invention discloses an intelligent process optimization system and method for a mold manufacturing process, and relates to the technical field of process optimization, and the system comprises a target change polling module, a target coverage rate analysis module, a parameter interval quality label updating instruction generation module, a parameter interval quality label updating module and an intelligent process optimization scheme generation module. According to the method, the self-evolution of the process parameter space and the dynamic adaptation of the optimization result can be realized, the delay caused by repeated simulation and manual judgment is reduced, the optimization process of the mold manufacturing process is converted from experience driving to data driving, the precision and response speed of the process decision in the simulation stage are improved, and the simulation efficiency is improved. And a high-credibility parameter combination basis is provided for subsequent actual manufacturing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of process optimization, in particular to an intelligent process optimization system and method for a mold manufacturing process. BACKGROUND

[0002] As an important basic link of modern manufacturing industry, the process design and manufacturing process of mold manufacturing directly affect the dimensional accuracy, surface quality and production efficiency of products. In order to reduce the trial and error cost and processing risk in the actual manufacturing stage, mold manufacturing enterprises generally carry out simulation analysis before actual production, and predict and verify the forming process, thermal stress distribution, cooling efficiency and material flow behavior by establishing virtual models of molds and processes. At present, the simulation process before mold manufacturing usually includes geometric modeling, meshing, boundary condition setting, material parameter definition, process parameter input and analysis of simulation calculation results. Common simulation software such as Moldflow, DEFORM, ProCAST, ANSYS and ABAQUS can analyze different types of mold process such as injection molding, forging, stamping, die casting and other multi-physical field coupling. Through simulation calculation, metal flow, temperature field distribution, stress and strain evolution and filling and cooling process can be observed in a virtual environment, thereby providing reference for process parameter setting and structure design.

[0003] For example, the Chinese invention patent with publication number CN120277984A discloses a mold stamping process optimization method and system. The method includes: obtaining a data set of a part, and extracting part features; performing a fitting simulation of the part and a cooperating part, and establishing a limiting space of the part; inputting the part features and product requirements of the part into a structure feature analysis model to perform structure analysis, and establishing a structure sensitivity identifier; performing mold optimization fitting, and establishing a mold rib scheme set; based on the scheme set, performing stamping key parameter optimization, and establishing a control parameter optimization set; performing mapping result evaluation analysis through a multi-dimensional fitness evaluation function, and generating a process optimization scheme.

[0004] For example, the Chinese invention patent with publication number CN118395774A discloses a super-high strength plate strip finishing process parameter multi-objective optimization method and system. A response surface model is established through a finishing process experiment, a relationship between process parameters and strip elongation is established, and the elongation is taken as a constraint condition in a multi-objective optimization model; an energy consumption model including strip deformation power, friction power and idling power is established; a bending roller load is simulated, and a relationship between the bending roller load and the process parameters is established according to a virtual experiment result; a multi-objective optimization model considering energy consumption and load balance is established, and a NSGA-II algorithm is used to solve the process parameter multi-objective optimization model established above, and two groups of Pareto solution sets are obtained.

[0005] The prior art at least has the following technical problems:

[0006] In the mold manufacturing process optimization, engineering practice often needs to consider multiple objectives at the same time, such as the dimensional accuracy of the molded part, the mold life, the defect rate, and the production efficiency, etc. The focus and weight of each target may be adjusted at any time in different production stages or batches. This dynamic demand of multi-objective and multi-weight directly leads to the definition of the superior / inferior interval in the process parameter space to have significant dynamics. Specifically, a part of the interval in the parameter space may change its effectiveness and value under different target or weight settings. For example, some parameter combinations (such as high temperature + fast cooling) may belong to the inferior interval (such as large deformation) when the dimensional accuracy is the main target, but when the optimization target is switched to the mold life, this interval may become the superior interval (low stress, long life). Therefore, the superiority and inferiority of the parameter interval are not static and unchangeable, but closely depend on the current optimization target and weight distribution. For this reason, the parameter interval originally marked as invalid or low value in the historical simulation data may need to be re-examined and even focused on mining with the change of the optimization target and weight. SUMMARY

[0007] In order to solve the above technical problems existing in the prior art, the embodiments of the present application provide an intelligent process optimization system and method for mold manufacturing process. The technical scheme is as follows:

[0008] On the one hand, an intelligent process optimization system for mold manufacturing process is provided, comprising:

[0009] A target change polling module is configured to perform intelligent process optimization simulation based on the mold manufacturing process, set a polling period, poll the configuration file target change at a regular time, and generate a target coverage judgment instruction when the target change is polled.

[0010] A target coverage analysis module is configured to, after receiving the target coverage judgment instruction, analyze a key sub-interval set of each parameter based on the current target information, and synchronously query an existing parameter set in a production database, and then analyze the target coverage rate, thereby determining a target information coverage completeness label.

[0011] A parameter interval superior / inferior label update instruction generation module is configured to, when the target information coverage completeness label is not completely covered, generate a parameter completion instruction, and when it is judged that the parameter completion is completed, generate a parameter interval superior / inferior label dynamic update instruction, and when the target information coverage completeness label is completely covered, directly generate a parameter interval superior / inferior label dynamic update instruction.

[0012] A parameter interval superior / inferior label update module is configured to, after receiving the parameter interval superior / inferior label dynamic update instruction, analyze the comprehensive score of the current target information in each historical parameter interval, thereby updating the parameter interval superior / inferior label, and writing it into the attribute field in the buffer pool.

[0013] The intelligent process optimization scheme generation module is used to filter parameter combinations with superior and inferior labels for parameter ranges as preferred parameter ranges, and generate intelligent process optimization schemes by combining current target information and through Pareto fronts.

[0014] On the other hand, a smart process optimization method for mold manufacturing is provided, including the following steps:

[0015] Perform intelligent process optimization simulation based on mold manufacturing process, set polling cycle, periodically poll for changes in configuration file target, and generate target coverage judgment instruction when target change is detected.

[0016] Upon receiving the target coverage assessment instruction, the system analyzes the key sub-interval set of each parameter based on the current target information, and simultaneously queries the existing parameter set in the production database to analyze the target coverage rate, thereby determining the target information coverage completeness label.

[0017] When the target information coverage label is not fully covered, a parameter completion instruction is generated. After the parameter completion is completed, a parameter range quality label dynamic update instruction is generated. When the target information coverage label is fully covered, a parameter range quality label dynamic update instruction is directly generated.

[0018] Upon receiving the instruction to dynamically update the superiority / inferiority labels for parameter intervals, the system analyzes the comprehensive score of the current target information across each historical parameter interval, updates the superiority / inferiority labels for the parameter intervals accordingly, and writes them into the attribute fields in the buffer pool.

[0019] The parameter ranges with superior and inferior labels are selected as the optimal parameter ranges. Combined with the current target information, a smart process optimization scheme is generated through the Pareto front.

[0020] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0021] 1. This invention provides an intelligent process optimization system and method for mold manufacturing, which can achieve multi-objective dynamic process optimization and adaptive parameter range updates during the simulation stage before mold manufacturing, thereby improving the targeting and efficiency of process parameter optimization. By introducing an objective change polling mechanism and objective coverage self-checking logic into the simulation process, it can detect adjustments to optimization objectives or weights in real time and automatically trigger coverage analysis and completion of the parameter space, ensuring that parameter ranges under different optimization objectives have sufficient simulation support data. By identifying and determining the validity of key sub-ranges, it can relatively accurately identify representative parameter combination regions under the current objective, realizing dynamic correction and supplementary sampling of the parameter space. Furthermore, by introducing a comprehensive scoring mechanism based on quality evaluation parameters and a hierarchical update strategy for parameter range superiority and inferiority labels, it can quickly complete the intelligent division of parameter superiority and inferiority ranges based on historical simulation data, and use the preferred range for Pareto front analysis, thereby generating an optimal process scheme with multi-objective balance. In summary, this invention enables the self-evolution of process parameter space and dynamic adaptation of optimization results, reduces delays caused by repetitive simulations and manual judgments, transforms the optimization process of mold manufacturing from experience-driven to data-driven, improves the accuracy and response speed of process decisions in the simulation stage, and provides a highly reliable parameter combination basis for subsequent actual manufacturing.

[0022] 2. This invention analyzes target coverage to determine the completeness label of target information coverage, enabling precise quantification and dynamic judgment of parameter space coverage during the simulation phase. This avoids the problem of missing simulation data in parameter intervals due to target changes. When insufficient target coverage is detected, a parameter completion process is automatically triggered to ensure sufficient data support for each key sub-interval. When coverage is complete, the process directly enters the superior / inferior interval update stage, achieving optimal utilization of simulation resources. Therefore, the process optimization process is always carried out under conditions of sufficient data and consistent targets, making subsequent process parameter analysis and optimization scheme generation more complete and reliable.

[0023] 3. This invention, by updating the superiority / inferiority labels of parameter intervals, can dynamically reflect the overall performance of each process parameter interval under different optimization objectives or weight switching, thereby achieving adaptive evolution of parameter space evaluation results. By introducing a comprehensive scoring calculation and a hierarchical mechanism for optimal, critical, and inferior intervals, the superiority / inferiority status of each interval can be automatically corrected based on the latest simulation data and quality evaluation parameters, avoiding parameter failure or misjudgment problems caused by traditional fixed labels. This enables continuous optimization and refined management of process parameter intervals, making the generated optimization schemes more consistent with the performance requirements of the current simulation objectives, and improving the accuracy and intelligence level of process optimization. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of an intelligent process optimization system for mold manufacturing provided in an embodiment of the present invention.

[0026] Figure 2 This is a flowchart of an intelligent process optimization method for mold manufacturing provided by an embodiment of the present invention.

[0027] Figure 3 This is a flowchart of the target coverage judgment and parameter completion process involved in the embodiments of the present invention.

[0028] Figure 4 This is a flowchart of the parameter range quality label update process involved in this embodiment.

[0029] Figure 5 It is a cloud map of the melt temperature distribution during the filling process inside the mold cavity.

[0030] Figure 6 It is a temperature distribution cloud map of the cooling channels (circuit) inside the injection mold.

[0031] Figure 7 This is a trend graph showing the change in the lowest flow front temperature over injection time. Detailed Implementation

[0032] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0033] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0034] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0035] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0036] To make the technical problems, technical solutions and advantages of the present invention clearer, the following will describe them in detail with reference to the accompanying drawings, using injection mold process optimization simulation as a specific embodiment.

[0037] This invention provides an intelligent process optimization system and method for mold manufacturing, such as... Figure 1 The diagram shows a schematic of an intelligent process optimization system for mold manufacturing, which includes: a target change polling module, a target coverage analysis module, a parameter range advantage / disadvantage label update instruction generation module, a parameter range advantage / disadvantage label update module, and an intelligent process optimization scheme generation module.

[0038] The target change polling module is used to perform intelligent process optimization simulation based on the mold manufacturing process. It sets the polling cycle, periodically polls for target changes in the configuration file, and generates a target coverage judgment instruction when a target change is detected.

[0039] Taking the virtual simulation and parameter optimization process applied to injection mold manufacturing as an example, the system first sets a polling cycle, such as every 10 minutes, 30 minutes, or a polling cycle customized by the operator according to the actual process rhythm, and automatically checks the configuration file stored locally or in the cloud at regular intervals. This configuration file can contain multi-dimensional information such as the current production target of the injection molded product, quality evaluation parameters (such as dimensional accuracy, defect rate, molding cycle, mold life, etc.), quality evaluation weight factor settings for each quality evaluation parameter, batch process requirements, and recent quality feedback.

[0040] At the end of each polling cycle, the system automatically reads and parses the configuration file to determine if there have been any changes to the process optimization goals. These changes include proactive adjustments from the production management system, MES, or process engineers, as well as shifts in optimization goals triggered by real-time quality feedback, equipment status monitoring, and adaptive process adjustments. For example, if the initial batch focused on shortening the molding cycle, but due to a recent increase in defect rates, the system or engineers might adjust the optimization goal to prioritize reducing the defect rate and update this in the configuration file. Once a goal change is detected, the system automatically pauses the current simulation optimization task and generates a goal coverage judgment instruction in real time. This instruction serves as an internal logical trigger signal to initiate subsequent process parameter space analysis.

[0041] Through the above process, the system can continuously and efficiently monitor the dynamic changes of process optimization objectives during the injection mold manufacturing process, avoiding a decrease in the accuracy of process parameter optimization results due to changes in objectives.

[0042] like Figure 3 The diagram shows a flowchart of the target coverage judgment and parameter completion process according to an embodiment of the present invention. The process begins with the generation of a target coverage judgment instruction. Upon receiving this instruction, the system first analyzes the set of key sub-intervals for each parameter based on the current target information. Then, it performs target coverage analysis on the analyzed set of key sub-intervals. The system determines whether the current target coverage is greater than the target coverage threshold and whether it meets the key sub-interval coverage conditions. If the determination result is negative, the target information coverage completeness label is recorded as incomplete coverage, and a parameter completion instruction is generated. Next, parameter completion is executed, and the completion status is judged. If completion is incomplete, the parameter completion process continues until completion is complete. After completion, the system re-analyzes the target coverage and judges the coverage completeness, ultimately generating a dynamic update instruction for parameter interval quality labels. If the determination result is positive, the target information coverage completeness label is directly recorded as complete coverage, and a dynamic update instruction for parameter interval quality labels is generated. This process ensures that when the target coverage is incomplete, the system can gradually complete the coverage of key sub-intervals through parameter completion loops, thereby providing a reliable data foundation for the dynamic updating of the superior and inferior labels of subsequent parameter intervals.

[0043] The target coverage analysis module is used to analyze the set of key sub-intervals of each parameter based on the current target information after receiving the target coverage judgment instruction, and simultaneously query the existing parameter set in the production database to analyze the target coverage and thus determine the target information coverage completeness label.

[0044] Furthermore, based on the current target information, the key sub-interval set of each parameter is analyzed. The specific analysis process is as follows:

[0045] Based on the current target information, obtain the parameter ranges for this optimization.

[0046] Determine the step size for each parameter, and then divide the interval of each parameter based on the step size to obtain the sub-intervals of each parameter.

[0047] In a specific embodiment, it is assumed that the current injection mold manufacturing process is for the mass production of a precision electronic housing component. The current optimization target information is to require the product dimensional deviation to be less than 0.03mm, the defect rate to be no higher than 0.8%, and the single-piece molding cycle to be no more than 35 seconds. Based on this optimization target information, the system obtains the process parameters and their ranges required for this optimization as follows: the holding pressure is set between 8MPa and 12MPa, with a step size of 0.5 MPa, and the parameter range is: [8.0, 8.5), [8.5, 9.0), [9.0, 9.5), [9.5, 10.0), [10.0, 10.5), [10.5, 11.0), [11.0, 11.5), [11.5, 12.0]MPa. The injection speed is set between 35 mm / s and 50 mm / s, with a step size of 5 mm / s, and the parameter range is [35, 40), [40, 45), [45, 50] mm / s. The cooling time is set between 18 seconds and 26 seconds, with a step size of 2 seconds, and the parameter range is [18, 20), [20, 22), [22, 24), [24, 26] seconds. The mold temperature is set between 65℃ and 80℃, with a step size of 5℃, and the parameter range is [65, 70), [70, 75), [75, 80]℃.

[0048] For a given parameter, obtain each sub-interval of that parameter, query the switching frequency of each sub-interval from the candidate pool, and record the sub-interval with a switching frequency greater than the preset switching frequency as the first key sub-interval.

[0049] Switching frequency is an indicator used to measure the activity of label status changes in each sub-interval of injection mold process parameters during process optimization. In each simulation optimization iteration, the system records the good and bad labels for all sub-intervals. When the label of a sub-interval changes between two analyses (e.g., from a critical parameter interval to a preferred parameter interval or from a preferred parameter interval to a disadvantageous parameter interval), it is considered a switch, and the system increments the switching count for that sub-interval by one.

[0050] Furthermore, a switch is recorded when a sub-interval is first identified as a preferred parameter interval and added to the candidate pool. By statistically analyzing the label changes of all sub-intervals, the system can calculate the switching frequency of each sub-interval throughout the entire optimization cycle. The switching frequency is the number of times the label of that sub-interval switches per unit time or per optimization round. The system defines sub-intervals with switching frequencies greater than a preset switching frequency threshold (the switching frequency threshold is set manually by experts in the field based on experience and can be specifically set based on different real-time scenarios) as the first critical sub-intervals, indicating that the process performance of these intervals is highly sensitive to the optimization objective.

[0051] Determine the basic key sub-intervals based on the current target information.

[0052] In this embodiment, the system is applied to injection mold process optimization simulation. By subdividing the process parameter space, it achieves automatic identification of basic critical sub-intervals. Specifically, after receiving the current optimization target information, the system first analyzes the process indicators (such as dimensional accuracy, defect rate, molding cycle, etc.) and their target setting ranges that need to be focused on in this round of process optimization. Subsequently, the system automatically retrieves historical simulation data and interval scoring results for each optimization parameter (such as holding pressure, injection speed, cooling time, mold temperature, etc.), and divides the value range of the optimization parameter into several sub-intervals evenly or empirically. For all sub-intervals of each parameter, the system analyzes the historical mean, variance, best and worst performance of each quality evaluation parameter in that interval, and, combined with the process requirements set for the current target, screens out those sub-intervals that historically performed closest to or were most easily met within the current target setting range. Specifically, if most of the historical simulation samples in a certain sub-interval meet the requirements of the current target for all quality evaluation indicators (such as dimensional deviations all being less than the target threshold, defect rates all being lower than the limit, etc.), then the system marks that sub-interval as a basic critical sub-interval. Furthermore, for sub-intervals within each optimization parameter interval that exhibit extreme values ​​among multi-objective indicators (such as minimum defect rate, shortest molding cycle, and optimal dimensional accuracy), the system also includes them in the set of basic critical sub-intervals. Through the above process, the system can automatically determine the basic critical sub-intervals in each process parameter space based on the current optimization objective information.

[0053] The first key sub-interval and the basic key sub-interval are combined to form the key sub-interval set for this parameter.

[0054] Iterate through each parameter to obtain the set of key sub-intervals for each parameter.

[0055] Furthermore, the target coverage is analyzed, and the specific analysis process is as follows:

[0056] Based on the current target information, query the existing historical parameter set in the production database.

[0057] After receiving the current optimization target information, the system first extracts the process parameters and related process indicator requirements that need to be focused on in this round of optimization. The system then accesses the database and automatically constructs a database query statement based on the parameter names, value ranges, and quality evaluation indicator conditions set for the current optimization target. The specific query process includes: retrieving all historical simulation data records stored in the database and filtering out all historical parameter combinations within the current optimization parameter space. The query conditions are not only limited to the names and ranges of the optimization parameters but also include the relevant constraints of the process quality indicators and the current target settings. For example, if this round of optimization requires a holding pressure of 8.0 to 12.0 MPa, an injection speed of 35 to 50 mm / s, and a dimensional accuracy deviation of less than 0.03 mm, the system will filter out all historical parameter combinations where the holding pressure and injection speed fall within the above ranges, and the corresponding dimensional accuracy indicators meet the requirements. The query results form an existing historical parameter set, which is the set of all existing historical parameters in the current production database that meet the parameter range and quality requirements of this round of optimization targets. This existing set of historical parameters will serve as the foundation for subsequent target coverage analysis, parameter range label determination, data reuse, and parameter completion, providing data support for intelligent process optimization.

[0058] Based on the existing set of historical parameters, if a certain historical parameter falls into the sub-interval of its corresponding parameter, then the sub-interval of that parameter is recorded as the initial valid sub-interval, and the number of parameters falling into it is recorded; otherwise, the sub-interval of that parameter is recorded as the invalid sub-interval.

[0059] If the number of parameters falling into the range is greater than the preset threshold for the number of parameters falling into the range, it is determined that the historical sample size of the sub-interval is sufficient and the statistical representativeness is good. The initial valid sub-interval can be directly recorded as a valid sub-interval. Otherwise, it indicates that the historical data of the initial valid sub-interval is limited and the representativeness is insufficient. The initial valid sub-interval will be recorded as an invalid sub-interval.

[0060] The total number of each sub-interval for each parameter is counted and recorded as the total number of sub-intervals. The number of valid sub-intervals is counted simultaneously.

[0061] The target coverage rate is obtained by dividing the number of valid sub-intervals by the total number of sub-intervals.

[0062] It should be noted that the target coverage value ranges from 0 to 1. The higher the value, the more comprehensive the distribution of historical simulation data in the current parameter space and the more complete the coverage.

[0063] Furthermore, the target information coverage completeness label is determined, and the specific determination process is as follows:

[0064] If the target coverage rate is greater than the preset target coverage rate threshold and meets the key sub-interval coverage condition, it means that most parameter sub-intervals have sufficient and representative historical data, the parameter space is fully covered, and the historical data can meet all the requirements of subsequent intelligent optimization. At this time, the system can directly enter the dynamic update process of parameter interval quality labels without supplementing the data. The target information coverage completeness label is recorded as complete coverage, and the dynamic update instruction for parameter interval quality labels is directly generated.

[0065] If the target coverage rate is less than or equal to the preset target coverage rate threshold, it means that a considerable proportion of sub-interval data in the parameter space is still insufficient or missing, which cannot guarantee the representativeness and reliability of the optimization results, or the key sub-interval coverage conditions are not met, indicating that there are data blind spots in key risk intervals in the parameter space, which may directly affect the robustness of the process optimization results. In this case, the target information coverage completeness label is marked as incomplete coverage, a parameter completion instruction is generated, and parameter completion is performed.

[0066] The key sub-interval coverage condition is that all key sub-intervals are valid sub-intervals.

[0067] The parameter range quality label update instruction generation module is used to generate parameter completion instructions when the target information coverage completeness label is not fully covered, and to generate parameter range quality label dynamic update instructions after the parameter completion is completed. When the target information coverage completeness label is fully covered, the parameter range quality label dynamic update instructions are directly generated.

[0068] Furthermore, parameter completion is performed, and the specific execution process is as follows:

[0069] All invalid sub-intervals are counted to form a list of invalid sub-intervals, and then the re-sampling task for each invalid sub-interval is generated.

[0070] The task of re-sampling each invalid sub-interval includes setting the parameters and configuring the simulation input for each invalid sub-interval.

[0071] In this embodiment, based on the determination results of target coverage and key sub-interval coverage conditions, the system statistically analyzes and categorizes all invalid sub-intervals that are not covered by historical valid data, ultimately forming an invalid sub-interval list. This invalid sub-interval list clearly records the parameter ranges that, during this round of process optimization, cannot meet the target coverage requirements due to missing historical data or insufficient sample size. Examples include the holding pressure range [9.5, 10.0) MPa, the injection speed range [45, 50) mm / s, and the cooling time range [22, 24) seconds. For each item in the invalid sub-interval list, the system automatically generates a corresponding supplementary sampling task. Each supplementary sampling task includes specific parameter settings for the sub-interval to be supplemented and a complete simulation input configuration. Taking the invalid holding pressure sub-interval [9.5, 10.0) MPa as an example, the supplementary sampling task will explicitly specify that the holding pressure in the simulation input parameters must be set to the representative value of this interval (e.g., 9.75 MPa). At the same time, according to the needs of the current round of optimization, the values ​​of other process parameters such as injection speed, cooling time, and mold temperature will be set (historical best values, empirically recommended values, or a full parameter combination strategy can be adopted). The system will automatically generate the input files required by the simulation software to ensure that all invalid sub-intervals have independent and reasonable parameter settings. Subsequently, all supplementary sampling tasks are submitted in batches to the task queue of the simulation platform (such as Moldflow (selected in this embodiment) or other injection molding simulation software), and the completion simulation will be automatically started.

[0072] By using the simulation software's API interface, all invalid sub-interval re-sampling tasks are submitted in batches to the simulation task queue, thereby executing all invalid sub-interval re-sampling tasks and outputting the simulation result data for each invalid sub-interval.

[0073] Collect simulation results data for each invalid sub-interval, perform data checks, record data that passes the data check as qualified data, and mark the corresponding invalid sub-interval as completed.

[0074] Data inspection includes data integrity inspection and data consistency inspection.

[0075] In this embodiment, data integrity checking refers to the system verifying each simulation result data item by item to confirm that all necessary process quality indicators have been generated completely. Taking injection mold process simulation as an example, the system will check whether the simulation results contain all key indicators required by the current optimization target, such as dimensional deviation, defect rate, molding cycle, mold temperature distribution, pressure peak, etc. If there are missing items, data not output, or file corruption in the simulation results, the data is judged as unqualified and needs to be re-simulated and re-collected.

[0076] Data consistency checks involve verifying the parameter labels in the simulation results against the input settings to ensure that the process parameter settings for each simulation data point strictly correspond to the invalid sub-intervals specified in the supplementary sampling task. For example, in a supplementary sampling simulation with a holding pressure range of [9.5, 10.0) MPa, the holding pressure in the output file must fall within this range, and all other process parameters (such as injection rate, cooling time, etc.) must be consistent with the simulation task configuration. If parameter misalignment, input-output discrepancies, or abnormal data fluctuations are found (such as quality indicators deviating from historical distribution ranges), it is also considered data inconsistency.

[0077] In a specific embodiment, continuing the above specific embodiment, after system query, it was found that the number of historical data in the two sub-intervals of holding pressure parameters [9.5, 10.0) MPa and [10.0, 10.5) MPa was 1, far below the system's set qualified threshold of 3. Therefore, these two intervals were included in the invalid sub-interval list. Subsequently, the system automatically generated a supplementary sampling task for each invalid sub-interval. Taking the [9.5, 10.0) MPa interval as an example, the simulation input parameters were set to holding pressure 9.75 MPa, injection speed 40 mm / s, cooling time 20 seconds, and mold temperature 70℃. For the [10.0, 10.5) MPa interval, the simulation parameters were a holding pressure of 10.25 MPa, with other parameters remaining the same. By calling the Moldflow simulation software API interface, the system submitted the above two supplementary sampling tasks in batches and automatically executed the simulation. After the simulation was completed, the system collected the output results of the supplementary sampling tasks. For the [9.5, 10.0) MPa range, the simulation results are: dimensional deviation 0.018 mm, defect rate 0.40%, and molding cycle 28.5 seconds. For the [10.0, 10.5) MPa range, the simulation results are: dimensional deviation 0.021 mm, defect rate 0.32%, and molding cycle 29.0 seconds. The system first performs a completeness check on all output data to ensure that all indicators such as dimensional deviation, defect rate, and molding cycle are output and their values ​​are valid. Subsequently, the system verifies that the simulation output parameters are completely consistent with the input of the supplementary data acquisition task, and no misalignment or configuration abnormalities are found, thus determining that the data consistency is qualified. Finally, both sets of supplementary data are marked as qualified data, and the corresponding invalid sub-range status is updated to complete.

[0078] Further, the system checks if the parameter completion is complete. The specific check process is as follows:

[0079] Traverse each invalid sub-interval. If the marking results of all invalid sub-intervals are complete, generate a coverage detection instruction and re-analyze the target coverage.

[0080] If the target coverage rate of the reanalysis is greater than the target coverage rate threshold and meets the key sub-interval coverage condition, it means that the proportion of the number of valid sub-intervals among all the divided parameter sub-intervals has exceeded the target coverage rate threshold, the historical and newly added simulation data can fully represent the overall parameter space, and all the previously identified key sub-intervals have been covered by valid historical or newly supplemented data. There are no key weak points in the system data, so the parameter completion is judged to be complete.

[0081] If the target coverage rate after reanalysis is less than or equal to the target coverage rate threshold, or does not meet the key sub-interval coverage conditions, it indicates that there are still weak data links or coverage blind spots in the parameter space. The system must generate a continue completion instruction, automatically initiate a new round of parameter completion process, and reanalyze the target coverage rate after each execution of the parameter completion process until the target coverage rate of the current analysis is greater than the target coverage rate threshold and meets the key sub-interval coverage conditions. Then, the parameter completion is considered complete.

[0082] It should be explained that in this embodiment, the system performs batch supplementary sampling simulations for all invalid sub-intervals and conducts strict consistency and integrity checks on each supplementary data entry to ensure that all supplementary data is qualified and added to the historical parameter set. Nevertheless, when re-analyzing the target coverage and key sub-interval coverage conditions subsequently, situations may still arise where the target coverage is less than or equal to the target coverage threshold, or the key sub-interval coverage conditions are not met. This situation is usually not due to the supplementary data itself being invalid, but rather because, with a finely divided parameter space or a high sample threshold, a single round of supplementary sampling may only add one or a small number of new data entries to each invalid sub-interval, resulting in the historical sample count (i.e., the number of parameters falling into the range) of some sub-intervals still failing to exceed the set threshold for the number of parameters falling into the range, and thus failing to be recognized as valid sub-intervals by the system.

[0083] like Figure 4The diagram shows the flowchart for updating the superiority / inferiority labels of parameter intervals in this embodiment. The process begins with receiving a dynamic update instruction for the superiority / inferiority labels of parameter intervals. The system selects each quality evaluation parameter based on the current optimization target information and calculates a comprehensive score for each historical parameter interval based on these parameters. The system compares the comprehensive score of each historical parameter interval with preset preferred thresholds and critical thresholds: if the comprehensive score is greater than or equal to the preferred threshold, the historical parameter interval is marked as a preferred parameter interval; if the comprehensive score is greater than the critical threshold but less than the preferred threshold, the historical parameter interval is marked as a critical parameter interval; if the comprehensive score is less than or equal to the critical threshold, the historical parameter interval is marked as a inferior parameter interval. After determining the superiority / inferiority labels for one historical parameter interval, the system iterates through the next historical parameter interval, repeating the above comparison and marking process until the superiority / inferiority labels for all historical parameter intervals are updated, thereby providing reliable superiority / inferiority information for subsequent intelligent process optimization scheme generation.

[0084] The parameter interval quality label update module is used to analyze the comprehensive score of the current target information in each historical parameter interval after receiving the parameter interval quality label dynamic update instruction, thereby updating the parameter interval quality label and writing it into the attribute field in the buffer pool.

[0085] Furthermore, the comprehensive score of the current target information across various historical parameter intervals is analyzed. The specific analysis process is as follows:

[0086] Based on the current optimization target information, select each quality evaluation parameter.

[0087] Continuing with the specific implementation example above, let's assume the optimization goals for this batch of products are set as follows: dimensional accuracy weight 0.5, defect rate weight 0.3, molding cycle weight 0.2, requiring the final dimensional deviation of the product to not exceed 0.03mm, the defect rate to not exceed 0.8%, and the molding cycle per part to not exceed 35 seconds. Based on these optimization goals, the system automatically selects the quality evaluation parameters to focus on in this round of optimization. These parameters include: dimensional accuracy, which is the deviation between the molded product and the design dimensions, used to evaluate the precision of the product; defect rate, which is the proportion of defects such as porosity, short shots, and warpage in all injection molded parts, used to reflect the stability of the production process and the product yield; and molding cycle, which is the total time taken for a single product from injection, holding pressure, cooling to demolding, used to measure production efficiency.

[0088] Obtain the quality evaluation parameters of existing historical samples in each historical parameter interval.

[0089] For a given historical parameter range, after normalizing each quality evaluation parameter, a quality evaluation weight factor is introduced to perform weighted coupling processing on the normalization results of each quality evaluation parameter, thereby obtaining the comprehensive score for that historical parameter range.

[0090] In a specific embodiment, the comprehensive score for a historical parameter interval is defined as: Where A is the comprehensive score for the historical parameter range, and X... i Let X be the i-th quality evaluation parameter in the historical parameter interval. i,max X is the preset maximum value of the i-th quality evaluation parameter in this historical parameter interval. i,min w is the preset minimum value of the i-th quality evaluation parameter in this historical parameter interval. i Let i be the quality evaluation weight factor of the i-th quality evaluation parameter in the historical parameter interval, where i is the number of each quality evaluation parameter, i=1,2,...,n, and n is the number of quality evaluation parameters.

[0091] In this embodiment, the preset minimum and maximum values ​​of each quality evaluation parameter are used to normalize the numerical range of different process quality indicators in the historical parameter range, so as to facilitate weighting and synthesizing a comprehensive score. The system sets minimum and maximum values ​​for each quality evaluation parameter based on actual process requirements, product design standards, and historical data distribution. Specifically, taking dimensional accuracy, defect rate, and molding cycle as examples, the system determines the preset value ranges for each parameter based on the current batch production target and industry standards: For the dimensional accuracy parameter, based on product design tolerances and past high-quality production data, the system presets a minimum value of 0.010mm, representing the optimal dimensional deviation achieved in actual production, and a maximum value of 0.030mm, corresponding to the maximum allowable dimensional deviation for this batch; for the defect rate parameter, considering past stable production batches and product quality requirements, the system presets a minimum value of 0.10%, representing the historical best yield rate, and a maximum value of 0.80%, corresponding to the upper limit of the defect rate required by the customer; for the molding cycle parameter, referencing equipment performance and the lower limit of process optimization, the system presets a minimum value of 24 seconds and a maximum value of 35 seconds, representing the fastest molding and the longest acceptable molding time, respectively. The preset minimum and maximum values ​​for each quality evaluation parameter can be obtained from statistical data of historical excellent production batches or determined based on industry standards, product drawing requirements, etc.

[0092] By iterating through each historical parameter interval, a comprehensive score for each historical parameter interval is obtained.

[0093] Furthermore, the superiority / inferiority labels for the parameter ranges are updated. The specific update process is as follows:

[0094] Extract the preset preferred threshold and critical threshold from the database.

[0095] The parameter intervals whose comprehensive scores are greater than or equal to the preferred threshold are tagged as the preferred parameter intervals.

[0096] The parameter intervals whose overall score is less than the preferred threshold but greater than the critical threshold are denoted as the critical parameter intervals.

[0097] The parameter intervals with a comprehensive score less than or equal to the critical threshold in the historical parameter intervals are labeled as inferior parameter intervals.

[0098] In this embodiment, the system uses an optimal threshold and a critical threshold to classify the comprehensive score of each historical parameter interval into hierarchical management, thereby achieving a scientific classification of the superiority and inferiority labels of parameter intervals. The optimal threshold indicates that when the comprehensive score reaches or exceeds this threshold, the parameter interval performs excellently and can be directly used as a recommended option for optimization. The critical threshold indicates that when the comprehensive score reaches or exceeds this threshold but does not reach the optimal threshold, the parameter interval is in a suboptimal state, performing close to optimal but with room for improvement. Intervals below or equal to the critical threshold are considered inferior, indicating that the parameter combination is difficult to meet the process optimization objectives or has significant shortcomings. The specific presets for these two thresholds are typically determined by combining actual process requirements, historical data distribution, optimization target weights, and enterprise internal control standards. Taking a comprehensive score range of 0-1 as an example, the system can be set based on historical production statistics: the optimal threshold is preset to 0.85, meaning that parameter ranges with a comprehensive score of 0.85 or higher are considered optimal, representing excellent performance in all process quality evaluation parameters; the critical threshold is preset to 0.70, meaning that parameter ranges with a comprehensive score between 0.70 (excluding) and 0.85 are considered critical, representing performance close to the target requirements in some process quality evaluation parameters, but not yet perfect; parameter ranges with a comprehensive score equal to or lower than 0.70 are classified as inferior parameter ranges. The optimal and critical thresholds can be set by enterprise process experts based on past experience, data analysis, and optimization results, or dynamically recommended through statistical analysis of historical simulation score distribution ranges, reference target quality achievement rates, or even the introduction of machine learning methods. All threshold information is centrally stored in the database and can be flexibly accessed and adjusted to adapt to the actual needs of different batches, products, or optimization goals.

[0099] Iterate through each historical parameter interval to obtain the superior and inferior labels for each historical parameter interval, and then update the superior and inferior labels for each historical parameter interval.

[0100] The intelligent process optimization scheme generation module is used to filter parameter combinations with superior and inferior labels for parameter ranges as preferred parameter ranges, and generate intelligent process optimization schemes by combining current target information and through Pareto fronts.

[0101] Furthermore, a smart process optimization scheme is generated, the specific process of which is as follows:

[0102] Each optimal parameter interval is selected from the historical parameter intervals, and the comprehensive score of each optimal parameter interval is extracted based on the comprehensive score of each historical parameter interval.

[0103] Based on the current optimization objective information, obtain the set of optimization parameters for this round.

[0104] It should be noted that the determination of the optimized parameter set is based on the target requirements, process characteristics and engineer settings of this round of production process, and selects all process parameters that actually participate in the parameter combination and optimization in this optimization (such as holding pressure, injection speed, cooling time and mold temperature in this embodiment).

[0105] Based on the optimized parameter set, the optimal parameter intervals are selected from each preferred parameter interval and denoted as the preferred parameter intervals for each optimized parameter. That is, for each optimized parameter, the parameter type is selected from all preferred intervals. For example, the preferred intervals for parameters such as holding pressure and injection speed are extracted from all preferred intervals and listed separately.

[0106] Quality evaluation parameters are extracted from each optimized parameter range based on each quality evaluation parameter.

[0107] Based on the set of optimized parameters and the quality evaluation parameters of each optimal parameter range, Pareto front analysis is used to comprehensively judge the performance of all optimized parameters in each objective dimension, and the Pareto front solution set is obtained.

[0108] It should be explained that Pareto front analysis refers to the selection of all parameter combinations in multi-objective optimization where improving one objective does not lead to a deterioration in other objectives; that is, the optimal compromise solution that cannot be surpassed by other combinations in all objective dimensions.

[0109] All optimization parameters located at the Pareto front solution set, along with their corresponding quality evaluation parameters and comprehensive scores, are output as intelligent process optimization schemes.

[0110] Continuing with the specific implementation example for mass production of precision electronic housing components, the optimization objectives for this round are set as follows: dimensional accuracy weight 0.5, defect rate weight 0.3, and molding cycle weight 0.2, simultaneously achieving high precision, low defects, and high efficiency. The core optimization parameter set includes holding pressure, injection speed, cooling time, and mold temperature. First, the system traverses all parameter ranges in the historical database and extracts the comprehensive scores for these ranges (calculated and stored according to the aforementioned formula). For example, the holding pressure ranges are [8.0,8.5), [8.5,9.0), [9.0,9.5), [9.5,10.0), [10.0,10.5), [10.5,11.0), [11.0,11.5), [11.5,12.0] MPa, with historical simulation samples and corresponding scores for each range. The system sets the optimization threshold to 0.85, and filters all intervals with scores ≥ 0.85 as the optimal parameter intervals. Assume the filtering results are: [9.0, 9.5), [9.5, 10.0), [10.0, 10.5), [10.5, 11.0) MPa. Based on the objectives of this round, the system clarifies the set of parameters to be optimized (holding pressure, injection speed, cooling time, mold temperature), and filters all intervals with scores ≥ 0.85 from their respective optimal interval lists, resulting in parameters such as injection speed [40, 45), [45, 50) mm / s, cooling time [20, 22), [22, 24) seconds, and mold temperature [70, 75) ℃. For each optimal interval of each optimization parameter, all historical simulation quality evaluation parameters within that interval are extracted from the database. For example, under [9.5, 10.0) MPa, the dimensional accuracy of the samples is 0.019 mm and 0.018 mm, the defect rate is 0.41% and 0.39%, and the molding cycle is 28.6 and 28.5 seconds, respectively. The optimal intervals for holding pressure, injection speed, cooling time, and mold temperature are arranged and combined to generate all candidate solutions. For each combination, the system collects all its quality evaluation parameters and uses the Pareto front analysis method to screen out the optimal compromise solution that cannot be comprehensively surpassed by other solutions in the three target dimensions of dimensional accuracy, defect rate, and molding cycle. All parameter combinations located at the Pareto front, along with their corresponding quality evaluation parameters and comprehensive scores, are output as the intelligent process optimization solution for this round. For example, Output Option 1: Holding pressure 9.75MPa, injection speed 45mm / s, cooling time 22 seconds, mold temperature 75℃, dimensional accuracy 0.017mm, defect rate 0.36%, molding cycle 28.7 seconds, overall score 0.91. Option 2: Holding pressure 10.25MPa, injection speed 40mm / s, cooling time 20 seconds, mold temperature 70℃, dimensional accuracy 0.018mm, defect rate 0.38%, molding cycle 29.1 seconds, overall score 0.90.

[0111] Again, taking injection molds as an example, in another specific embodiment, for a shell-type injection molded structure, see [reference needed]. Figure 5 , Figure 6 and Figure 7 The figures show the melt temperature distribution cloud map during the mold filling process, the temperature distribution cloud map of the cooling channels (circuit) within the injection mold, and the trend diagram of the lowest flow front temperature changing with injection time. The shell-type injection molded structure shown in the figures has a thin wall thickness, a long strip-shaped main cavity, local openings and reinforcing ribs, a side gate, a long runner, and a symmetrical cavity structure. Since melt temperature distribution plays a decisive role in product quality, excessively low temperatures may result in incomplete mold filling, while excessively high temperatures can cause defects such as decomposition and shrinkage. Further reference... Figure 5 As shown, the left side shows the flow front temperature, and the right side shows the three-dimensional mold cavity and flow channel structure. The temperature of each part of the cavity is between 244.7°C and 249.1°C, and the distribution of hot and cold spots is clear.

[0112] Because the temperature distribution in the cooling channels directly affects the thermal balance of different areas of the mold, it influences the cooling rate, shrinkage, warpage, and cycle time of the product. A uniform temperature distribution contributes to dimensional consistency, while excessive temperature differences can lead to inconsistent cooling rates throughout the product, causing potential quality issues. See further details. Figure 6 As shown, the left-hand menu displays the loop cooling temperature, and the right-hand menu shows the three-dimensional cooling channel temperature distribution. The color bars indicate that the temperature range within the channel is between 25.01°C and 25.74°C, and the temperature gradually increases from the inlet to the outlet as the cooling medium flows through the cooling loop.

[0113] The minimum flow front temperature reflects the temperature change of the region the melt first reaches during mold filling. If the temperature drops too low, the melt may partially solidify before completely filling the mold cavity, leading to poor filling or reduced strength. See further... Figure 7 As shown, the background represents the mold cavity structure outline of this shell-type injection molding structure. The curves are depicted with triangular dot markers, with the vertical axis representing temperature (°C) and the horizontal axis representing injection time (s). Under conditions of a mold temperature of 55.00°C and a melt temperature of 263.5°C, the evolution trajectory of the minimum melt front temperature over time during the injection molding process was automatically collected. As can be seen from the figure, the minimum flow front temperature continuously decreases from approximately 280°C initially to approximately 185°C, reflecting the melt heat loss and flow cooling behavior during injection molding.

[0114] like Figure 2 The flowchart shown illustrates an intelligent process optimization method for mold manufacturing. The method includes:

[0115] Perform intelligent process optimization simulation based on mold manufacturing process, set polling cycle, periodically poll for changes in configuration file target, and generate target coverage judgment instruction when target change is detected.

[0116] Upon receiving the target coverage assessment instruction, the system analyzes the key sub-interval set of each parameter based on the current target information, and simultaneously queries the existing parameter set in the production database to analyze the target coverage rate, thereby determining the target information coverage completeness label.

[0117] When the target information coverage label is not fully covered, a parameter completion instruction is generated. After the parameter completion is completed, a parameter range quality label dynamic update instruction is generated. When the target information coverage label is fully covered, a parameter range quality label dynamic update instruction is directly generated.

[0118] Upon receiving the instruction to dynamically update the superiority / inferiority labels for parameter intervals, the system analyzes the comprehensive score of the current target information across each historical parameter interval, updates the superiority / inferiority labels for the parameter intervals accordingly, and writes them into the attribute fields in the buffer pool.

[0119] The parameter ranges with superior and inferior labels are selected as the optimal parameter ranges. Combined with the current target information, a smart process optimization scheme is generated through the Pareto front.

[0120] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0121] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0122] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0123] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0124] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0125] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0126] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0127] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0128] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An intelligent process optimization system for mold manufacturing, characterized in that, The system includes: The target change polling module is used to perform intelligent process optimization simulation based on the mold manufacturing process. It sets the polling cycle, periodically polls for target changes in the configuration file, and generates a target coverage judgment instruction when a target change is detected. The target coverage analysis module is used to analyze the set of key sub-intervals of each parameter based on the current target information after receiving the target coverage judgment instruction, and simultaneously query the existing parameter set in the production database to analyze the target coverage and thus determine the target information coverage completeness label. The parameter range quality label update instruction generation module is used to generate parameter completion instructions when the target information coverage completeness label is not fully covered, and to generate parameter range quality label dynamic update instructions after the parameter completion is completed. When the target information coverage completeness label is fully covered, the parameter range quality label dynamic update instructions are directly generated. The parameter interval merit label update module is used to analyze the comprehensive score of the current target information in each historical parameter interval after receiving the parameter interval merit label dynamic update instruction, thereby updating the parameter interval merit label and writing it into the attribute field in the buffer pool. The intelligent process optimization scheme generation module is used to filter parameter combinations with superior and inferior labels for parameter ranges as preferred parameter ranges, and generate intelligent process optimization schemes by combining current target information and through Pareto fronts.

2. The intelligent process optimization system for mold manufacturing according to claim 1, characterized in that, The key sub-interval set for each parameter is analyzed based on the current target information. The specific analysis process is as follows: Based on the current target information, obtain the parameter ranges for this optimization; Determine the step size for each parameter, and divide the interval of each parameter into intervals based on the step size of each parameter to obtain the sub-intervals of each parameter; For a parameter, obtain each sub-interval of the parameter, query the switching frequency of each sub-interval from the candidate pool, and record the sub-interval with a switching frequency greater than the preset switching frequency as the first key sub-interval. Determine the basic key sub-intervals based on the current target information; The first key sub-interval and the basic key sub-interval are combined to form the set of key sub-intervals for this parameter; Iterate through each parameter to obtain the set of key sub-intervals for each parameter.

3. The intelligent process optimization system for mold manufacturing process according to claim 2, characterized in that, The analysis of the target coverage rate is carried out in the following specific process: Based on the current target information, query the existing historical parameter set in the production database; Based on the existing set of historical parameters, if there exists an existing historical parameter that falls into the sub-interval of its corresponding parameter, then the sub-interval of that parameter is recorded as the initial valid sub-interval, and the number of parameters that fall into it is recorded; otherwise, the sub-interval of that parameter is recorded as the invalid sub-interval. If the number of parameters falling into the range exceeds the preset threshold for the number of parameters falling into the range, the initial valid sub-interval is recorded as a valid sub-interval; otherwise, the initial valid sub-interval is recorded as an invalid sub-interval. Count the total number of each sub-interval for each parameter, and record it as the total number of sub-intervals. Simultaneously count the number of valid sub-intervals. The target coverage rate is determined based on the number of valid sub-intervals and the total number of sub-intervals.

4. The intelligent process optimization system for mold manufacturing process according to claim 1, characterized in that, The process for determining the target information coverage completeness label is as follows: If the target coverage rate is greater than the preset target coverage rate threshold and meets the key sub-interval coverage condition, the target information coverage completeness label is recorded as complete coverage, and a parameter interval quality label dynamic update instruction is directly generated. If the target coverage rate is less than or equal to the preset target coverage rate threshold, or if the key sub-interval coverage condition is not met, the target information coverage completeness label is marked as incomplete coverage, a parameter completion instruction is generated, and parameter completion is executed. The key sub-interval coverage condition is that all key sub-intervals are valid sub-intervals.

5. The intelligent process optimization system for mold manufacturing process according to claim 4, characterized in that, The execution parameter completion process is as follows: All invalid sub-intervals are counted to form a list of invalid sub-intervals, and then the re-sampling task for each invalid sub-interval is generated; The task of re-sampling each invalid sub-interval includes setting parameters and configuring simulation input for each invalid sub-interval; By using the API interface of the simulation software, all invalid sub-interval re-sampling tasks are submitted in batches to the simulation task queue, thereby executing all invalid sub-interval re-sampling tasks and outputting the simulation result data of each invalid sub-interval. Collect simulation results data for each invalid sub-interval, perform data checks, record data that passes the data check as qualified data, and mark the corresponding invalid sub-interval as completed; The data inspection includes data integrity inspection and data consistency inspection.

6. The intelligent process optimization system for mold manufacturing process according to claim 1, characterized in that, The judgment parameters have been completed, and the specific judgment process is as follows: Traverse each invalid sub-interval. If the marking results of all invalid sub-intervals are complete, generate a coverage detection instruction and re-analyze the target coverage. If the target coverage rate after reanalysis is greater than the target coverage rate threshold and meets the key sub-interval coverage condition, then the parameter completion is considered complete. If the target coverage rate after reanalysis is less than or equal to the target coverage rate threshold, or if the key sub-interval coverage condition is not met, a continue completion instruction is generated, the parameter completion process is re-executed, and the target coverage rate is re-analyzed after each execution of the parameter completion process until the target coverage rate of the current analysis is greater than the target coverage rate threshold and the key sub-interval coverage condition is met, then the parameter completion is considered complete.

7. The intelligent process optimization system for mold manufacturing process according to claim 1, characterized in that, The analysis of the current target information's comprehensive score across various historical parameter intervals is as follows: Based on the current optimization target information, select the quality evaluation parameters; Obtain the quality evaluation parameters of existing historical samples in each historical parameter interval; For a given historical parameter interval, after normalizing each quality evaluation parameter, a quality evaluation weight factor is introduced to perform weighted coupling processing on the normalization results of each quality evaluation parameter, thereby obtaining the comprehensive score for that historical parameter interval. By iterating through each historical parameter interval, a comprehensive score for each historical parameter interval is obtained.

8. The intelligent process optimization system for mold manufacturing process according to claim 7, characterized in that, The specific update process for the superiority / inferiority labels within the updated parameter range is as follows: Extract the preset preferred threshold and critical threshold from the database; The parameter intervals whose comprehensive scores are greater than or equal to the preferred threshold are recorded as the preferred parameter intervals. The parameter interval whose overall score is less than the preferred threshold but greater than the critical threshold is denoted as the critical parameter interval. The parameter intervals whose overall score is less than or equal to the critical threshold in the historical parameter intervals are labeled as the inferior parameter intervals. Iterate through each historical parameter interval to obtain the superior and inferior labels for each historical parameter interval, and then update the superior and inferior labels for each historical parameter interval.

9. The intelligent process optimization system for mold manufacturing process according to claim 1, characterized in that, The specific process for generating the intelligent process optimization scheme is as follows: Each optimal parameter interval is selected from the historical parameter intervals, and the comprehensive score of each optimal parameter interval is extracted based on the comprehensive score of each historical parameter interval. Based on the current optimization objective information, obtain the set of optimization parameters for this round; Based on the set of optimized parameters, select each preferred parameter interval from each preferred parameter interval, and denot it as each preferred parameter interval of each optimized parameter. Quality evaluation parameters are extracted based on each quality evaluation parameter, and each optimal parameter range is extracted for each optimization parameter. Based on the set of optimized parameters and the quality evaluation parameters of each optimal parameter range, Pareto front analysis is used to comprehensively judge the performance of all optimized parameters in each objective dimension, and the Pareto front solution set is obtained. All optimization parameters located at the Pareto front solution set, along with their corresponding quality evaluation parameters and comprehensive scores, are output as intelligent process optimization schemes.

10. A method for intelligent process optimization of mold manufacturing process, applied to the intelligent process optimization system for mold manufacturing process as described in any one of claims 1-9, characterized in that, The method includes the following steps: Perform intelligent process optimization simulation based on mold manufacturing process, set polling cycle, periodically poll for changes in configuration file target, and generate target coverage judgment instruction when target change is detected. Upon receiving the target coverage judgment instruction, the system analyzes the key sub-interval set of each parameter based on the current target information, and simultaneously queries the existing parameter set in the production database to analyze the target coverage rate, thereby determining the target information coverage completeness label. When the target information coverage label is not fully covered, a parameter completion instruction is generated. After the parameter completion is completed, a parameter interval quality label dynamic update instruction is generated. When the target information coverage label is fully covered, a parameter interval quality label dynamic update instruction is directly generated. Upon receiving the instruction to dynamically update the superiority / inferiority labels for parameter intervals, the system analyzes the comprehensive score of the current target information in each historical parameter interval, thereby updating the superiority / inferiority labels for the parameter intervals and writing them into the attribute fields in the buffer pool. The parameter ranges with superior and inferior labels are selected as the optimal parameter ranges. Combined with the current target information, a smart process optimization scheme is generated through the Pareto front.

Images