Foaming production method, control device, electronic equipment and storage medium
By acquiring production data from the foaming production line, using a preset model to predict process defects and automatically adjust, the problems of reliance on manual labor and slow response speed are solved, achieving efficient foaming production.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-11
- Publication Date
- 2026-04-03
AI Technical Summary
The production process of polyurethane foam insulation layers in existing refrigerators is highly dependent on manual labor and has a slow response time, which affects production efficiency.
By acquiring production data from the foaming production line, using a preset foaming process model to predict process defect information, and determining adjustment parameters based on the prediction results, the foaming production line is automatically adjusted to avoid the production of substandard products.
It enables timely adjustments during the production process, reduces the defect rate, eliminates reliance on manual management, and improves production efficiency.
Smart Images
Figure CN121785253A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of foam production technology, and particularly relates to a foam production method, control device, electronic device and storage medium. Background Technology
[0002] In the current production process of polyurethane foam insulation layers for refrigerators, the main reliance is on manual labor to monitor raw material fluctuations, equipment status changes, and environmental temperature and humidity effects in the production workshop. Adjustments to the production line can only be made based on experience after defective products appear. This high degree of reliance on manual labor and slow response time affects production efficiency. Summary of the Invention
[0003] This application provides a foaming production method, control device, electronic device, and storage medium to solve the problem that the polyurethane foam insulation layer of existing refrigerators is highly dependent on manual labor and has a slow response speed, which affects production efficiency.
[0004] This application provides a foaming production method, the method comprising: Obtain production data from the foaming production line; Based on the production data, process defect information is predicted using a preset foaming process model. Adjustment parameters are determined based on the process defect information; The foaming production line is adjusted based on the aforementioned adjustment parameters.
[0005] Optionally, the production data includes: raw material data, process data, and environmental data; wherein, The raw material data includes one or more of the following: raw material type, raw material batch, raw material temperature, raw material ratio, and raw material viscosity. The process data includes one or more of the following: mold flow rate, injection pressure, mold temperature, and mold casting volume; The environmental data includes one or more of the following: ambient temperature and ambient humidity in the production workshop.
[0006] Optionally, the foaming process model includes a mechanism-based physical sub-model and a machine learning-based data-driven sub-model; The prediction of process defect information based on the preset foaming process model includes: Predict the first defect risk based on the physical sub-model; Predict the risk of the second defect based on the data-driven sub-model; The process defect information is generated by integrating the first defect risk and the second defect risk, wherein the process defect information includes defect type, risk level, confidence level and cause analysis.
[0007] Optionally, predicting the first defect risk based on the physical sub-model includes: The raw material temperature and the ambient temperature are input into the physical sub-model to obtain the risk of uneven cell structure. The mold flow rate and the injection pressure are input into the physical sub-model to obtain the risk of insufficient filling or overfilling. The raw material ratio and the raw material temperature are input into the physical sub-model to obtain the risk of incomplete reaction or excessive cross-linking. The first defect risk is obtained by quantifying and normalizing the risks of uneven cell structure, insufficient filling, overfilling, inadequate reaction, and excessive crosslinking.
[0008] Optionally, predicting the second defect risk based on the data-driven sub-model includes: Feature extraction is performed on the production data to obtain a defect feature vector; The defect feature vector is input into the data-driven sub-model to obtain a probability vector; The second defect risk is obtained based on the probability vector and the preset threshold.
[0009] Optionally, the fusion of the first defect risk and the second defect risk includes: The first defect risk and the second defect risk are merged according to preset weights; The preset weight is determined based on the historical reliability of the first defect risk and the historical reliability of the first defect risk type.
[0010] Optionally, after acquiring the production data of the foaming production line, the method further includes: The production data is aggregated, cleaned, and time-series aligned to obtain processed production data; The step of predicting process defects based on the production data and a preset foaming process model includes: Based on the processed production data, process defects are predicted using a preset foaming process model.
[0011] This application embodiment also provides a foaming production control device, the device comprising: The production data acquisition module is configured to acquire production data from the foaming production line; The prediction module is configured to predict process defect information based on the production data and a preset foaming process model. The analysis module is configured to determine adjustment parameters based on the process defect information; The control module is configured to adjust the foaming production line based on the adjustment parameters.
[0012] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the foaming production method described above.
[0013] This application embodiment also provides a storage medium storing control instructions, which, when executed by a processor, implement the foaming production method described above.
[0014] The foaming production method provided in this application predicts potential process defects based on production data from the foaming production line and a preset foaming process model during the production process. It then determines corresponding adjustment parameters based on the prediction results and makes timely adjustments to the foaming production line. This allows for automatic data detection and timely adjustments during production to avoid producing substandard products. Compared to existing methods that rely on manual analysis of already produced substandard products to adjust the production data for the next batch, the foaming production method provided in this application is more responsive, reduces the defect rate, and eliminates reliance on manual management, thus achieving intelligent foaming technology. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings. In the following description, the same reference numerals denote the same parts.
[0017] Figure 1 This is a schematic flowchart of a foaming production method provided in an embodiment of this application.
[0018] Figure 2 This is a schematic flowchart of a foaming production method provided in another embodiment of this application.
[0019] Figure 3 This is a schematic diagram of a model structure of a foaming production method provided in an embodiment of this application.
[0020] Figure 4 This is a schematic diagram of the structure of a foaming production control device provided in an embodiment of this application.
[0021] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0023] In the description of the embodiments of this application, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, and memory, and may also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, a microprocessor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc.
[0024] This application provides a foaming production method, control device, electronic device, and storage medium to solve the problem that the polyurethane foam insulation layer of existing refrigerators is highly dependent on manual labor and has a slow response speed, which affects production efficiency. The following description is in conjunction with the accompanying drawings.
[0025] The foaming production method provided in this application embodiment is described in the following reference. Figure 1 The method includes the following steps: Step S101: Obtain production data from the foaming production line. The foaming production line is a continuous or intermittent automated production system that mixes liquid or molten polymer materials (such as polyurethane, polystyrene, rubber, etc.) with a foaming agent, and through chemical reactions or physical methods, creates a honeycomb-like porous structure inside, which is then solidified into a predetermined shape.
[0026] Production data includes raw material data, process data, and environmental data. Raw material data includes one or more of the following: raw material type, raw material batch, raw material temperature, raw material ratio, and raw material viscosity. Process data includes one or more of the following: mold flow rate, injection pressure, mold temperature, and mold casting volume. Environmental data includes one or more of the following: ambient temperature and ambient humidity in the production workshop.
[0027] The method of acquiring production data can be determined based on the specific type of production data. Production data can be obtained through channels such as online monitoring equipment, vision systems, MES (Manufacturing Execution System), and SCADA (Supervisory and Data Acquisition). For example, the viscosity of raw materials can be obtained through online rheometers or online viscometers; the temperature of raw materials can be obtained through near-infrared spectrometers or thermocouples; the proportion of raw materials can be obtained through mass flow meters, volumetric flow meters, or weighing sensors; the type and batch of raw materials can be obtained through MES / ERP systems; the flow rate of the mold can be obtained through flow meters built into the casting head or dedicated flow meters at the mold inlet; the injection pressure can be obtained through pressure transmitters or sensors; the mold temperature can be obtained through embedded resistance thermometers; the mold casting volume can be obtained through integral calculation using weighing sensors or high-precision metering pumps; and the ambient temperature and humidity of the production workshop can be obtained through digital temperature and humidity sensors / transmitters.
[0028] Step S102: Based on production data, predict process defect information using a pre-set foaming process model. This pre-set foaming process model is a pre-built model constructed based on the foaming production process and trained using massive amounts of historical data before being deployed to the foaming production line control system. Process defect information includes defect type, risk level, confidence level, and causal analysis. For example, based on production data, predict the foam flow front position, gelation time, final density distribution, and potential defects (such as incomplete filling or cavitation) within the next 30 to 60 seconds.
[0029] Step S103: Determine adjustment parameters based on process defect information. These adjustment parameters are executable instructions that directly affect the foaming production line control system, aiming to correct predicted defects. These parameters cover comprehensive adjustments from formulation and process conditions to equipment operation. For example, formulation and proportion adjustment parameters include, but are not limited to, isocyanate index adjustment, foaming agent dosage adjustment, catalyst ratio adjustment, and other additive adjustments; process condition adjustment parameters include, but are not limited to, temperature adjustment parameters (raw material temperature, mold temperature, ambient temperature); pressure and flow adjustment parameters (injection pressure, circulation pressure, injection speed / flow rate); and equipment operation and timing adjustment parameters include, but are not limited to, timing parameters, trajectory and position parameters, and mixing head operation parameters. For instance, when insufficient filling is predicted in the corner of the mold box, the injection pressure or flow rate in that area is fine-tuned; when abnormal raw material viscosity is detected, the A / B material ratio is automatically calculated and compensated; when the weather temperature changes in real time, the catalyst dosage or mold temperature is determined, and the ambient temperature and humidity of the foaming environment are also determined.
[0030] Step S104: Adjust the foaming production line based on the adjustment parameters. For example, precisely control the servo motor of the high-pressure foaming machine, adjust the injection ratio and pressure, control the temperature control system of each area of the mold to achieve precise local temperature control, and adapt the optimal flow rate, pressure, ratio and temperature before foaming injection for products with different heights, widths, depths and internal structures, and precisely control the filling state and density distribution of the box.
[0031] The foaming production method provided in this application, during the production process, predicts potential process defects based on production data from the foaming production line and a preset foaming process model. Then, based on the prediction results, it determines corresponding adjustment parameters and makes timely adjustments to the foaming production line, forming a real-time closed loop of "perception-decision-execution-feedback." This allows for automatic data detection and timely adjustments during production to avoid producing substandard products. Compared to existing methods that rely on manual analysis of already produced substandard products to adjust the production data for the next batch, the foaming production method provided in this application offers a more timely response, reduces the defect rate, and eliminates reliance on manual management, achieving intelligent foaming technology.
[0032] Optionally, please refer to Figure 2 The foaming process model includes a mechanism-based physical sub-model and a machine learning-based data-driven sub-model. Based on the preset foaming process model, process defect information is predicted, including: predicting the first defect risk based on the physical sub-model; predicting the second defect risk based on the data-driven sub-model; and integrating the first and second defect risks to generate process defect information, which includes defect type, risk level, confidence level, and cause analysis.
[0033] The physical sub-model is a mathematical model built upon fundamental physicochemical laws such as conservation of mass, energy, momentum, and chemical reaction kinetics. Because each step of its reasoning has physical meaning, its predictions are highly reliable and interpretable. Furthermore, even with entirely new formulations or processes, it can still provide reasonable inferences as long as the mechanism is correct, demonstrating strong extrapolation capabilities. The data-driven sub-model is a statistical model that automatically learns the complex mapping relationship between "input features" and "output defects" from historical data. It does not concern itself with the underlying mechanisms but only with the patterns in the data. It should be understood that the data-driven sub-model can use the same production data as the physical sub-model as input, or the physical sub-model can provide preprocessed input to the data-driven sub-model. It should also be understood that the physical sub-model and the data-driven sub-model run simultaneously after receiving a unified process data stream.
[0034] For the first defect risk, the defect type is usually a known defect with a clear mechanism (such as shrinkage cavity, turbulence, thermal stress cracking). Its risk value is usually a theoretical index or Boolean value calculated by formula (such as violation of design rules). Its cause usually points directly to the violated physical law or process window (such as "cooling rate is too low"). Its advantages are usually strong interpretability, clear physical meaning, and high reliability within the known theoretical range.
[0035] For the second defect risk, the defect type can usually be a known defect or an unknown but categorizable abnormal pattern (such as "unknown defect pattern A"). Its risk value is usually the probability value output by the model (such as 0.87), representing the similarity with historical defect patterns. Its cause is usually the key influencing features pointed out by interpretable AI (such as SHAP) (such as "feature X: holding pressure variance, contribution +0.35"). It should be noted that this is a statistical association, not physical causality. Its advantages usually include the ability to discover complex and implicit data patterns and strong predictive ability, and it is especially suitable for multivariate nonlinear problems.
[0036] By leveraging the complementary strengths and deep integration of physical and data-driven sub-models, "mechanism-guided data intelligence" is formed. By integrating the prediction results of the two models, the credibility of the prediction results is enhanced, avoiding the inability to promptly check or verify when the prediction results deviate due to the use of a single prediction model, thus ensuring the reliability of process defect information prediction.
[0037] Optionally, the prediction of the first defect risk based on the physical sub-model includes: inputting the raw material temperature and ambient temperature into the physical sub-model to obtain the risk of uneven cell structure; inputting the mold flow rate and injection pressure into the physical sub-model to obtain the risk of insufficient filling or overfilling; inputting the raw material ratio and raw material temperature into the physical sub-model to obtain the risk of insufficient reaction or excessive cross-linking; and quantifying and normalizing the risks of uneven cell structure, insufficient filling or overfilling, and insufficient reaction or excessive cross-linking to obtain the first defect risk.
[0038] Please refer to Figure 3 The physical sub-model includes a thermodynamics module, a fluid dynamics module, and a chemical reaction dynamics module.
[0039] For example, the raw material temperature and ambient temperature are input into the physical sub-model to obtain the risk of non-uniform cell structure. Specifically, the raw material temperature and ambient temperature are input into the thermodynamic module to obtain the risk of non-uniform cell structure. Among them, the risk of non-uniform cell structure mainly corresponds to the risk corresponding to the process of bubble nucleation, growth, merging and stabilization. The relevant key process parameters include foaming agent concentration, dispersion, temperature field uniformity, and pressure curve.
[0040] For example, mold flow rate and injection pressure are input into the physical sub-model to obtain the risk of underfilling or overfilling. Specifically, mold flow rate and injection pressure are input into the fluid dynamics module to obtain the risk of underfilling or overfilling. The risk of underfilling or overfilling mainly corresponds to the risk associated with the flow and filling behavior of the polymer / foam within the mold cavity. The relevant key process parameters include injection speed, injection pressure, mold temperature, and material viscosity.
[0041] For example, the raw material ratio and temperature are input into the physical sub-model to obtain the risk of incomplete reaction or excessive crosslinking. Specifically, the raw material ratio and temperature are input into the chemical reaction kinetics module to obtain the risk of incomplete reaction or excessive crosslinking. The risk of incomplete reaction or excessive crosslinking is mainly the risk corresponding to the kinetic process of polymerization / crosslinking reaction, and its related key process parameters include raw material temperature, mold temperature, catalyst concentration, and reaction time.
[0042] It should be noted that the risks of uneven cell structure, insufficient filling or overfilling, insufficient reaction or excessive crosslinking are quantified and normalized. This includes separately quantifying and modeling the risks of uneven cell structure, insufficient filling or overfilling, and insufficient reaction or excessive crosslinking, obtaining the quantified values for each risk, and then mapping multiple quantified values to a unified range of [0, 1] and assigning them clear risk levels. This makes the quantified values of risks with different dimensions and ranges comparable, which is convenient for integration with the second defect risk (data-driven, which is itself a probability value).
[0043] For example, the risk of non-uniform cell structure can be quantified, with core indicators being the uniformity of cell size distribution and the anisotropy of cell shape. Example of a quantification model: Rcell=1 U* Nactual / Nideal Where Rcell is the quantification value of the risk of non-uniformity in cell structure, U is the uniformity index (ranging from 0 to 1), Nactual is the actual nucleation density, and Nideal is the theoretical nucleation density. When the actual nucleation density is low or unevenly distributed, the quantification value of the risk of non-uniformity in cell structure approaches 1.
[0044] For example, the risk of underfilling or overfilling can be quantified using key indicators such as flow front velocity, cavity end pressure, and filler volume fraction. An example of a quantification model is provided below. Risk of incomplete mold filling:
[0045] Where Rshort is the quantification value of the risk of insufficient filling. For the arrival time of the flow front, The gel time of the material.
[0046] Overfilling risk:
[0047] in, R "over" represents the overfilled risk quantification value. This represents the actual injection volume. For the theoretical volume of the cavity, For end pressure, To ensure the mold can withstand pressure, λ This is the proportionality coefficient.
[0048] Overall liquidity risk: Rflow = max(Rshort, Rover) Rflow is the quantification value of the risk of underfilling or overfilling.
[0049] For example, the risks of incomplete reaction or excessive cross-linking can be quantified using core indicators such as reaction conversion rate, cross-linking density, and peak exothermic reaction. Example of a quantification model: Risk of inadequate response: Runder=max(0,1 Xpred / Xtarget) Wherein, Runder is the quantification value of insufficient response risk, Xpred is the predicted conversion rate, and Xtarget is the target conversion rate.
[0050] Risk of excessive cross-linking: RoverX=min(1, Δρ / ρcritical ) Where RoverX is the quantification value of excessive crosslinking risk, Δρ is the additional crosslinking density increment after the target conversion rate is reached at the process temperature, and ρcritical is the crosslinking density.
[0051] Overall reaction risk: Rreact = max(Runder, RoverX) Rreact is the quantitative value of the risk of insufficient reaction or excessive crosslinking.
[0052] Optionally, predicting the second defect risk based on the data-driven sub-model includes: extracting features from production data to obtain a defect feature vector; inputting the defect feature vector into the data-driven sub-model to obtain a probability vector; and obtaining the second defect risk based on the probability vector and a preset threshold.
[0053] By extracting features from production data, the original, high-dimensional, and time-series production data is transformed into low-dimensional feature vectors that can effectively characterize the process status and point to specific defect patterns. Then, using a pre-trained data-driven sub-model, the defect feature vectors are mapped to the quantified probability of defect risk, i.e., probability vectors. Finally, the obtained probability vectors are evaluated and filtered according to a preset threshold to obtain the second defect risk. This realizes the transformation of continuous model output, i.e., probability vectors, into discrete and operable "second defect risk" judgments.
[0054] For example, in the production process of polyurethane rigid foam insulation board, the following features are first extracted from the unified production data: raw material features, process features (injection stage and foaming stage), and morphological / temporal features. Based on the above features, an array is constructed (i.e., a defect feature vector is obtained). The system loads the pre-trained XGBoost multi-classification model, inputs the defect feature vector into the model, and the model outputs the raw score. Softmax converts the raw score into a probability vector. If the preset threshold is 0.7 (above is high risk, below is low risk), the probability vector is judged according to the preset threshold to obtain whether the defect risk type corresponds to high risk or low risk.
[0055] Optionally, the first defect risk and the second defect risk are integrated, including: integrating the first defect risk and the second defect risk according to a preset weight; wherein the preset weight is determined based on the historical reliability of the first defect risk and the historical reliability of the first defect risk type.
[0056] It is understandable that the weights of the first and second defect risks are allocated according to their historical reliability, that is, the higher the historical reliability, the higher the weight is allocated, in order to improve the reliability of the prediction results.
[0057] As an alternative implementation, the preset weights can also be determined based on the current operating condition type. For example, if the current operating condition is a normal operating condition, the data-driven sub-model is given a higher weight; if the current operating condition is an edge condition or a new operating condition, the physical sub-model is given a higher weight.
[0058] Optionally, when there is a serious conflict between the type and probability results of the first defect risk and the second defect risk (e.g., the physical sub-model predicts "normal" and the data-driven sub-model predicts "high risk"), a confidence arbitration mechanism can be initiated. Specifically, by checking factors such as the quality of the input data and whether the current operating condition is within the training set distribution, a decision can be made on which result to adopt or to trigger manual intervention.
[0059] Optionally, after obtaining the production data of the foaming production line, the method further includes: aggregating, cleaning, and aligning the production data according to time sequence to obtain processed production data; predicting process defects based on the production data and a preset foaming process model, including: predicting process defects based on the processed production data and a preset foaming process model.
[0060] This process involves three main steps: First, production data aggregation. This involves extracting raw data from various isolated systems (such as DCS, PLC, SCADA, and MES) and centralizing it on a data platform (such as a data lake). Second, production data cleaning. This involves identifying and repairing dirty data based on process knowledge (such as flow rate not being zero and temperature having a reasonable range) and logical rules (such as chronological order). Third, time-series alignment. This involves using the "birth of a product" as the core narrative line, integrating data generated at different times and on different devices through time offset calculations and key event correlations, and finally extracting statistically significant features (mean, variance, duration, cumulative amount, etc.).
[0061] By aggregating, cleaning, and aligning the data in sequence, the processed production data can accurately reflect the complete causal chain of "what kind of raw materials, under what process conditions, after what kind of processing, ultimately produce what kind of quality product," laying a solid foundation for subsequent foaming process model prediction of process defects.
[0062] It is understood that the foaming production method provided in this application embodiment can also be applied to the automotive industry (foaming of car seats and interior parts), the building materials industry (continuous foaming production of insulation boards), the furniture industry (foaming of upholstered furniture sponges), etc.
[0063] By implementing the foaming production method provided in this application, "zero defects" in product quality can be achieved. Through real-time prediction and compensation, the foaming defect rate can be reduced from the traditional 2% to below 0.2%, significantly improving the first-pass yield. Precise injection and proportioning control reduce overfilling, achieving 3%-5% raw material savings. In 2024, the refrigerator market sold approximately 40.19 million units. Assuming a single unit uses 8kg of foaming material, this translates to a reduction of 0.24-0.4kg of raw material per unit, saving 0.3-0.5 yuan per unit, resulting in annual savings of 12.05-20.09 million yuan. Fully automated closed-loop control reduces production line downtime and debugging time, improving overall equipment efficiency by over 10%, truly maximizing equipment capacity. It reduces reliance on highly skilled workers, lowering labor and training costs. Engineers' experience is transformed into replicable and optimizable algorithmic models, forming core enterprise data assets. The system continuously evolves with the accumulation of production data, becoming increasingly intelligent with use. By overcoming fluctuations in raw materials and the environment, the quality of the insulation layer of products manufactured throughout the year remains highly consistent across different seasons.
[0064] This application also provides a foaming production control device; please refer to [link / reference]. Figure 4 The device includes a production data acquisition module 201, a prediction module 202, an analysis module 203, and a control module 204. The production data acquisition module 201 is configured to acquire production data from the foaming production line; the prediction module 202 is configured to predict process defect information based on the production data and a preset foaming process model; the analysis module 203 is configured to determine adjustment parameters based on the process defect information; and the control module 204 is configured to adjust the foaming production line based on the adjustment parameters.
[0065] This application also provides an electronic device 300, please refer to... Figure 5 The system includes a memory 301, a processor 302, and a computer program 3011 stored in the memory 301 and executable on the processor 302. When the processor 302 executes the computer program 3011, it implements the foaming production method described above. The method includes the following steps: Step S101: Acquire production data from the foaming production line. Step S102: Based on the production data, predict process defect information based on a preset foaming process model. Step S103: Determine adjustment parameters based on the process defect information. Step S104: Adjust the foaming production line based on the adjustment parameters.
[0066] This application embodiment also provides a storage medium storing control instructions. When the control instructions are executed by a processor, they implement the foaming production method described above. The method includes the following steps: Step S101: Acquire production data from the foaming production line. Step S102: Based on the production data, predict process defect information based on a preset foaming process model. Step S103: Determine adjustment parameters based on the process defect information. Step S104: Adjust the foaming production line based on the adjustment parameters.
[0067] For example, a computer program can be divided into one or more modules / units, which are stored in memory and executed by a processor to perform the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in an electronic device.
[0068] Electronic devices can be desktop computers, laptops, handheld computers, and cloud servers, among other electronic devices. Electronic devices may include, but are not limited to, processors and memory. For example, electronic devices may also include input / output devices, network access devices, buses, etc.
[0069] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.
[0070] In the embodiments provided by this invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.
[0071] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, the functional units in the various embodiments of this 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. The integrated units described above can be implemented in hardware or as software functional units.
[0072] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.
[0073] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.
[0074] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more features.
[0075] The foaming production method, control device, electronic device, and storage medium provided in the embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A foaming production method, characterized in that, The method includes: Obtain production data from the foaming production line; Based on the production data, process defect information is predicted using a preset foaming process model. Adjustment parameters are determined based on the process defect information; The foaming production line is adjusted based on the aforementioned adjustment parameters.
2. The foaming production method according to claim 1, characterized in that, The production data includes: raw material data, process data, and environmental data; among which, The raw material data includes one or more of the following: raw material type, raw material batch, raw material temperature, raw material ratio, and raw material viscosity. The process data includes one or more of the following: mold flow rate, injection pressure, mold temperature, and mold casting volume; The environmental data includes one or more of the following: ambient temperature and ambient humidity in the production workshop.
3. The foaming production method according to claim 2, characterized in that, The foaming process model includes a mechanism-based physical sub-model and a machine learning-based data-driven sub-model. The prediction of process defect information based on the preset foaming process model includes: Predict the first defect risk based on the physical sub-model; Predict the risk of the second defect based on the data-driven sub-model; The process defect information is generated by integrating the first defect risk and the second defect risk, wherein the process defect information includes defect type, risk level, confidence level and cause analysis.
4. The foaming production method according to claim 3, characterized in that, The prediction of the first defect risk based on the physical sub-model includes: The raw material temperature and the ambient temperature are input into the physical sub-model to obtain the risk of uneven cell structure. The mold flow rate and the injection pressure are input into the physical sub-model to obtain the risk of insufficient filling or overfilling. The raw material ratio and the raw material temperature are input into the physical sub-model to obtain the risk of incomplete reaction or excessive cross-linking. The first defect risk is obtained by quantifying and normalizing the risks of uneven cell structure, insufficient filling, overfilling, inadequate reaction, and excessive crosslinking.
5. The foaming production method according to claim 3, characterized in that, The prediction of the second defect risk based on the data-driven sub-model includes: Feature extraction is performed on the production data to obtain a defect feature vector; The defect feature vector is input into the data-driven sub-model to obtain a probability vector; The second defect risk is obtained based on the probability vector and the preset threshold.
6. The foaming production method according to claim 3, characterized in that, The fusion of the first defect risk and the second defect risk includes: The first defect risk and the second defect risk are merged according to preset weights; The preset weight is determined based on the historical reliability of the first defect risk and the historical reliability of the first defect risk type.
7. The foaming production method according to claim 1, characterized in that, After acquiring the production data of the foaming production line, the method further includes: The production data is aggregated, cleaned, and time-series aligned to obtain processed production data; The step of predicting process defects based on the production data and a preset foaming process model includes: Based on the processed production data, process defects are predicted using a preset foaming process model.
8. A foaming production control device, characterized in that, The device includes: The production data acquisition module is configured to acquire production data from the foaming production line; The prediction module is configured to predict process defect information based on the production data and a preset foaming process model. The analysis module is configured to determine adjustment parameters based on the process defect information; The control module is configured to adjust the foaming production line based on the adjustment parameters.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the foaming production method as described in any one of claims 1-7.
10. A storage medium, characterized in that, The storage medium stores control instructions, which, when executed by a processor, implement the foaming production method as described in any one of claims 1-7.