Process parameter optimization method and device, electronic equipment and storage medium
By screening and preprocessing real-time process parameters and optimizing them using a quality prediction model, the problem of strong subjectivity in process parameter setting is solved, enabling rapid and automatic process parameter optimization and improving production efficiency and yield.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HONGFUJIN PRECISION ELECTRONICS ZHENGZHOU
- Filing Date
- 2025-12-23
- Publication Date
- 2026-05-08
AI Technical Summary
Existing process parameter setting methods are highly subjective, have poor reproducibility, and have long optimization cycles, resulting in poor production stability, large yield fluctuations, and an inability to adapt to the needs of rapid iterative production.
By acquiring real-time process parameters of the product in multiple processes, preprocessing them, and then screening key process parameters, the trained quality prediction model is used to predict multiple quality indicators. Under process constraints, the weighted total loss function is minimized to determine the target combination of process parameters, thereby achieving automatic optimization.
It can quickly and automatically find the globally optimal process parameters across multiple processes, significantly improving production efficiency and product yield, shortening the process improvement cycle, and enabling self-learning and continuous optimization.
Smart Images
Figure CN121998477A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of CNC machining technology, specifically to a method, apparatus, electronic device, and storage medium for optimizing process parameters. Background Technology
[0002] In complex multi-process product manufacturing processes, such as precision electronics manufacturing, the final quality of the product is determined by the process parameters during production.
[0003] However, the current process parameter setting mainly relies on engineers' personal experience for parameter initialization and adjustment, which is essentially a "trial and error" method. This method is not only highly subjective and difficult to replicate, but also requires waiting for the product to flow through the entire process and complete quality inspection after each parameter adjustment before the effect can be evaluated. This results in a long optimization cycle, delayed control, poor production stability, large yield fluctuations, and an inability to adapt to the needs of rapid iterative production. Summary of the Invention
[0004] In view of the above, it is necessary to propose a process parameter optimization method, device, electronic equipment and storage medium to solve the technical problems of existing process parameter setting methods, which are highly subjective, have poor reproducibility, long optimization cycle and lagging control, resulting in poor production stability, large yield fluctuations and inability to adapt to the needs of rapid iterative production.
[0005] In a first aspect, this application provides a process parameter optimization method applied to electronic devices. The method includes: acquiring real-time process parameters of a product in multiple manufacturing processes, and preprocessing the real-time process parameters to obtain key process parameters; based on a trained quality prediction model, obtaining multiple predicted quality indicators of the product according to the key process parameters; under preset process constraints, using the key process parameters as variables, minimizing the weighted total loss function value as the objective, solving for the minimization of the weighted total loss function to determine a target process parameter combination, wherein the weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators; and optimizing the real-time process parameters according to the target process parameter combination.
[0006] In the process parameter optimization method of this application embodiment, firstly, real-time process parameters of the product in multiple processes are obtained, and the real-time process parameters are preprocessed to obtain key process parameters; further, based on the trained quality prediction model, multiple predicted quality indicators of the product are obtained according to the key process parameters; further, under preset process constraints, the key process parameters are used as variables, and the weighted total loss function is minimized to determine the target process parameter combination, wherein the weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators; finally, the real-time process parameters are optimized according to the target process parameter combination. Based on this, this application constructs a quality prediction model that can output multiple quality indicators, inherently learns the complex relationships between quality indicators, and transforms the multi-objective optimization problem into minimizing a weighted total loss function based on the predicted values of quality indicators under process constraints. It uses optimization algorithms for intelligent solution, thereby finding the optimal combination of parameters with the best trade-off within the engineering feasible domain. It can quickly and automatically find the globally optimal process parameters across multiple processes, greatly shortening the cycle of existing process improvement and significantly improving production efficiency and product yield.
[0007] In some embodiments of this application, the step of preprocessing the real-time process parameters to obtain key process parameters includes: performing data cleaning and standardization on the real-time process parameters to obtain processed real-time process parameters; and filtering the processed real-time process parameters based on the correlation between the processed real-time process parameters and each quality indicator to obtain the key process parameters.
[0008] In some embodiments of this application, the step of optimizing the real-time process parameters according to the target process parameter combination includes: applying the target process parameter combination to the actual production of the multiple processes and collecting actual quality inspection data of the newly produced products; comparing the actual quality inspection data with the predicted quality index to determine the quality yield of the newly produced products; and if the quality yield of the newly produced products meets the preset requirements, updating the real-time process parameters in the multiple processes to the target process parameter combination.
[0009] In some embodiments of this application, the method further includes: if the quality yield of the newly produced product does not meet the preset requirements, updating the parameters of the quality prediction model based on the target process parameter combination.
[0010] In some embodiments of this application, the training steps of the quality prediction model include: acquiring historical process parameters and corresponding quality inspection data of the product in multiple processes, and preprocessing the historical process parameters to obtain historical key process parameters; configuring the contribution weight of the corresponding process stage for the historical key parameters based on preset engineering experience values to obtain weighted historical key process parameters; using the weighted historical key process parameters as input features and the quality indicators in the quality inspection data as input labels to construct and train the quality prediction model.
[0011] In some embodiments of this application, the method further includes: constructing the weighted total loss function based on the quality prediction model obtained from the training.
[0012] In some embodiments of this application, the plurality of processes include at least a sandblasting process for the product, a preceding process for the sandblasting process, and a following process for the sandblasting process.
[0013] Secondly, this application also provides a process parameter optimization device applied to electronic devices. The device includes: an acquisition module for acquiring real-time process parameters of a product in multiple manufacturing processes and preprocessing the real-time process parameters to obtain key process parameters; a prediction module for obtaining multiple predicted quality indicators of the product based on a trained quality prediction model and the key process parameters; a solution module for minimizing the weighted total loss function under preset process constraints, using the key process parameters as variables, to determine a target combination of process parameters, wherein the weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators; and an optimization module for optimizing the real-time process parameters according to the target combination of process parameters.
[0014] Thirdly, this application also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the process parameter optimization method described in the above embodiments.
[0015] Fourthly, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the process parameter optimization method described in the above embodiments.
[0016] Understandably, the process parameter optimization apparatus of the second aspect, the electronic equipment of the third aspect, and the computer-readable storage medium of the fourth aspect provided above all correspond to the process parameter optimization method of the first aspect. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding process parameter optimization methods provided above, and will not be repeated here. Attached Figure Description
[0017] Figure 1 This is a schematic flowchart of a process parameter optimization method provided in an embodiment of this application.
[0018] Figure 2 This is a schematic diagram of the functional modules of a process parameter optimization device provided in an embodiment of this application.
[0019] Figure 3 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application.
[0020] Explanation of main component symbols Electronic devices 10 Memory 11 Processor 12 Process parameter optimization device 100 Get Module 110 Solver Module 120 Prediction Module 130 Optimization Module 140 The following detailed description, in conjunction with the accompanying drawings, will further illustrate this application. Detailed Implementation
[0021] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the following detailed description is provided in conjunction with embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0022] To provide a clearer understanding of the embodiments of the present invention, the invention will be described in detail below with reference to the accompanying drawings and specific examples. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.
[0023] Please see Figure 1 This is a flowchart illustrating a process parameter optimization method provided in an embodiment of this application.
[0024] The process parameter optimization method in this application embodiment can be applied to... Figure 3The one or more electronic devices 10 shown are devices that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Their hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.
[0025] Specifically, the process parameter optimization method includes the following steps. Depending on different needs, the order of some steps in the flowchart can be changed, and some steps can be omitted.
[0026] Step S10: Obtain the real-time process parameters of the product in multiple processes, and preprocess the real-time process parameters to obtain the key process parameters.
[0027] Among them, multiple processes include, but are not limited to, the sandblasting process of the product in the production line, the preceding processes of the sandblasting process (such as stamping, CNC machining) and the following processes (such as anodizing, assembly).
[0028] Real-time process parameters include, but are not limited to, process parameters in the production processes of stamping, CNC machining, sandblasting, anodizing, and assembly, such as travel speed, oscillation speed, pressure of each spray gun, temperature, flow, and other 200+ dimension time series data.
[0029] In some embodiments of this application, the electronic device 10 can capture real-time process parameters daily / batch from the production execution system, equipment programmable logic controller, sensors, and detection instruments through an application programming interface / extraction conversion loading tool, and then enter the data processing layer for preprocessing, such as format conversion, anomaly detection, and parameter filtering.
[0030] In some embodiments of this application, the step of preprocessing real-time process parameters to obtain key process parameters includes: performing data cleaning and standardization on the real-time process parameters to obtain processed real-time process parameters; and filtering the processed real-time process parameters based on their correlation with each quality indicator to obtain key process parameters.
[0031] Specifically, for each parameter, a moving average and standard deviation are calculated for each model (the time window is configurable). The three sigma criterion is applied, and data points exceeding the mean ± 3 × standard deviation are marked as outliers. A cumulative sum algorithm or a sliding window mean is used to detect drift. Based on the magnitude and duration of the deviation, it is divided into three levels: "warning", "critical" and "fatal". For missing values, forward and backward filling or filling with the mean of the same model is used. Outliers are smoothed or removed, and differences in units and numerical ranges between different parameters are eliminated so that they can be compared and calculated under the same standard.
[0032] Furthermore, from over 200 real-time process parameters, the 30 most influential key parameters for each predicted quality indicator (e.g., gloss, color difference, etc.) are selected. Specifically, the dependency between each real-time process parameter and the target variable (i.e., quality indicator, such as gloss, color difference, etc.) is calculated to quantify the nonlinear relationship. Then, the parameters are sorted in descending order of dependency scores, and the top N parameters (e.g., the top 100 parameters) are selected. Next, the minimum absolute shrinkage and selection operator regression or elastic network with L1 regularization is used to input the top 100 parameters and quality indicators into the regression model. The regularization strength is adjusted through cross-validation so that the regression model retains exactly 30 non-zero coefficient features, thus obtaining a 30-dimensional list of key parameters for each quality indicator, which is then saved as a key process parameter file.
[0033] In the above embodiments, by cleaning massive amounts of process parameters, noise and outliers are removed from the data, data errors are corrected, and data naming rules and formats are standardized, ensuring parameter consistency and standardization. This provides a clean and reliable data foundation for all subsequent analyses. From the massive amount of parameters, based on factors such as the relevance and importance of the information and its impact on product quality, a set of key process parameters that significantly affect quality is selected. This achieves data dimensionality reduction, improves data processing efficiency and analytical accuracy, and reduces the complexity and overfitting risk of the quality prediction model.
[0034] Step S20: Based on the trained quality prediction model, obtain multiple predicted quality indicators for the product according to key process parameters.
[0035] In some embodiments of this application, the training steps of the quality prediction model include: acquiring historical process parameters and corresponding quality inspection data of the product in multiple processes, and preprocessing the historical process parameters to obtain historical key process parameters; configuring the contribution weight of the corresponding process stage for the historical key parameters based on preset engineering experience values (for example, the parameters of the sandblasting process have a greater impact on surface roughness, so the weight is higher), to obtain weighted historical key process parameters; using the weighted historical key process parameters as input features and the quality indicators in the quality inspection data as input labels to construct and train the quality prediction model.
[0036] The quality inspection data includes, but is not limited to, measurements of gloss, color difference ΔE, warpage, and protrusion after each process.
[0037] It should be noted that the historical process parameters and real-time process parameters have the same composition and processing steps. To avoid repetition, this application will not repeat them here.
[0038] In some embodiments of this application, to balance prediction accuracy and interpretability, the quality prediction model employs a dual-model strategy: Model 1: An end-to-end deep learning model (black box, high precision), employing a multilayer perceptron or a one-dimensional convolutional neural network. Input features include 30 selected key process parameters and quality inspection data. Output data consists of predicted quality indicators (gloss, color difference, etc.) used for high-precision prediction of the final yield.
[0039] Model 2: Two-stage interpretable model (white box, analyzable). First, an autoencoder or principal component analysis is used to compress the high-dimensional input into a low-dimensional feature vector (e.g., 10-dimensional). The low-dimensional feature vector represents the core potential factors affecting quality. Then, linear regression or a shallow neural network is used to obtain the predicted quality index based on the low-dimensional feature vector, which is used for contribution backtracking and root cause analysis, because each dimension has a clear direction of influence and weight.
[0040] In some embodiments of this application, the core architecture of the quality prediction model is input space (X) → weighting and fusion → nonlinear transformation → output space (Y), specifically: The first phase: The input space gathers 30 key process parameters after the product has undergone six critical processes (CNC sampling, CNC7, post-cleaning, post-sandblasting, post-anodizing, and assembly), which have been screened and comprise a total of 6 x 30 = 180 core features. This constitutes the input vector X of the quality prediction model. For example, X_Sandblasting might include: walking speed, swaying, and pressure of each spray gun, while X_anodizing might include: bath temperature, voltage density, current density, and electrolyte concentration. It is no longer data from a single process, but rather a "digital fingerprint" of the entire product lifecycle, forming the basis for cross-process correlation analysis.
[0041] The second stage involves dynamically allocating the contribution of features. Different weights (contribution percentages) are assigned to the six major processes mentioned above. For example, CNC machining accounts for 40%, post-anodizing for 25%, and CNC incoming material parameters for 15%. These percentages are learned from historical data, quantifying "which process's parameter set has the greatest overall impact on the final quality." The weights may vary depending on the machine type and quality indicator (Y). For example, for "color difference," the contribution of the post-anodizing process may be as high as 60%; while for "warpage," CNC machining may dominate.
[0042] The third stage: the mapping function f(X->Y) is the prediction engine of the quality prediction model. The input X is processed by a nonlinear transformation function h(x), and finally outputs the predicted quality index Y. This is an example of a linear model (such as linear regression). It is the output quality index coefficient. This represents the weighting coefficient for each specific parameter. The value of (i.e., the magnitude of the coefficient) is determined by both the percentage contribution of its respective process and engineering experience. For example, The contribution of CNC7 machining parameters: 40%. The contribution of CNC incoming material parameters is 15%, and this application does not impose any restrictions on this.
[0043] Phase 4: Output Space (Y) The final output of the quality prediction model is the key quality indicators that need to be predicted, such as Y=[Gloss, Color_Delta_E, BowingKingking, Bluge]]. For the same set of inputs X, the model can output the predicted values of multiple quality indicators in parallel, realizing one-stop quality assessment.
[0044] In the above embodiments, the intrinsic relationship between the quality indicators of each process and the process parameters is explored in depth, the influence mechanism of different process parameters on the quality of the final product is clarified, and the results of "data screening" and "correlation analysis" are solidified into a calculable and interpretable mathematical model to achieve the prediction of quality indicators.
[0045] Step S30: Under the preset process constraints, the key process parameters are used as variables. The goal is to minimize the weighted total loss function value. The weighted total loss function is minimized to determine the target combination of process parameters.
[0046] In some embodiments of this application, a weighted total loss function is constructed based on the quality prediction model obtained through training. The weighted total loss function is the weighted sum of the losses corresponding to each of the multiple predicted quality indicators.
[0047] In some embodiments of this application, the weighted total loss function is defined as: Where Yi_pred is the i-th predicted quality index output by the quality prediction model, Yi_target is the expected target value of the predicted quality index, Lossn is the loss function corresponding to the predicted quality index (e.g., mean square error, absolute error), and wi is a pre-set weight coefficient that reflects the importance of the predicted quality index in the optimization of process parameters.
[0048] In some embodiments of this application, process constraints include: the physical feasible range of each key process parameter (e.g., the sandblasting pressure cannot exceed the equipment limit), and process rules between each key process parameter (e.g., the temperature of process A must be higher than the temperature of process B), etc.
[0049] Specifically, the current key process parameters are used as the initial point. Optimization algorithms (such as gradient descent, genetic algorithms, and particle swarm optimization) are used to search under process constraints, continuously adjusting these key process parameters. New predicted quality indicators are obtained through a quality prediction model, and the weighted total loss function value is calculated. Finally, the combination of process parameters that minimizes the weighted total loss function value is found; this is the "target process parameter combination."
[0050] Step S40: Optimize the real-time process parameters based on the target combination of process parameters.
[0051] In some embodiments of this application, the step of optimizing real-time process parameters according to a target process parameter combination includes: applying the target process parameter combination to the actual production of multiple processes and collecting actual quality inspection data of newly produced products; comparing the actual quality inspection data with predicted quality indicators to determine the quality yield of newly produced products; if the quality yield of newly produced products meets preset requirements (e.g., actual yield ≥ 99%), then the optimization is proven to be effective, and the real-time process parameters in multiple processes are updated to the target process parameter combination.
[0052] In some embodiments of this application, the method further includes: if the quality yield of the newly produced product does not meet the preset requirements, triggering online learning or retraining of the quality prediction model, updating the parameters of the quality prediction model based on the target process parameter combination, thereby forming a continuously improving closed-loop system.
[0053] In some embodiments of this application, when the current product's yield meets preset requirements, it directly flows into the next process for continued production. When the current product's yield does not meet preset requirements, an alarm is triggered, the product is isolated and reported to prevent it from flowing into subsequent stages. The system analyzes which core features or original parameters contributed most to the quality problem, pinpointing the root cause. Based on the analysis results, optimization suggestions are automatically generated, such as adjusting CNC parameters (adjusting the machining parameters of the CNC machine tool) and adjusting anodizing parameters (adjusting the process parameters of surface treatment, such as anodizing). Furthermore, these optimization suggestions are converted into specific, executable instructions and sent to the production execution system to directly adjust the production line equipment, achieving adaptive process optimization and preventing subsequent products from experiencing the same problem again.
[0054] The above embodiments realize a quality control model transformation from "post-event detection" to "pre-event prediction and in-process control": by utilizing historical and real-time data, product quality can be predicted in advance, turning passive into proactive. This not only identifies problems but also pinpoints specific process parameters, enabling precise root cause analysis. The analysis results are directly fed back to the production equipment, automatically adjusting parameters to form a self-learning, continuously optimizing intelligent production closed loop, thereby improving yield, reducing waste, and lowering costs.
[0055] In some embodiments of this application, the electronic device 10 can also provide decision support for production and management personnel by visually displaying parameter optimization suggestions, thereby realizing intelligent management and control of the sandblasting process.
[0056] In the process parameter optimization method of this application embodiment, the real-time process parameters of the product in multiple processes are first obtained, and the real-time process parameters are preprocessed to obtain key process parameters. Further, based on the trained quality prediction model, multiple predicted quality indicators of the product are obtained according to the key process parameters. Further, under preset process constraints, the key process parameters are used as variables, and the goal is to minimize the weighted total loss function value. The weighted total loss function is then minimized to determine the target process parameter combination. The weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators. Finally, the real-time process parameters are optimized according to the target process parameter combination. Based on this, this application constructs a quality prediction model capable of outputting multiple quality indicators, inherently learning the complex relationships between quality indicators, and transforming the multi-objective optimization problem into minimizing a weighted total loss function based on the predicted value of quality indicators under process constraints. An optimization algorithm is used for intelligent solution, thereby finding the optimal parameter combination with the best trade-off within the engineering feasible domain. This allows for rapid and automatic finding of globally optimal process parameters across multiple processes, greatly shortening the cycle of existing process improvement and significantly improving production efficiency and product yield.
[0057] Please see Figure 2 This is a schematic diagram of the functional modules of a process parameter optimization device 100 provided in an embodiment of this application.
[0058] In this embodiment, based on the above... Figure 1 Using the same concept as the process parameter optimization method in the illustrated embodiments, this application also provides a process parameter optimization apparatus 100, which can be used to execute the above-described process parameter optimization method. For ease of explanation, the schematic diagram of the process parameter optimization apparatus 100 embodiment only shows the parts related to the embodiments of this application. Those skilled in the art will understand that the illustrated structure does not constitute a limitation on the process parameter optimization apparatus 100, and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.
[0059] Specifically, the process parameter optimization device 100 provided in this application embodiment includes an acquisition module 110, a solution module 120, an optimization module 130, and an optimization module 140.
[0060] The acquisition module 110 is used to acquire real-time process parameters of the product in multiple processes and preprocess the real-time process parameters to obtain key process parameters; the solution module 120 is used to obtain multiple predicted quality indicators of the product based on the trained quality prediction model and the key process parameters; the prediction module 130 is used to minimize the weighted total loss function value as a target, taking the key process parameters as variables under preset process constraints, and determine the target process parameter combination, wherein the weighted total loss function is the weighted sum of the losses corresponding to each of the multiple predicted quality indicators; the optimization module 140 is used to optimize the real-time process parameters according to the target process parameter combination.
[0061] In the process parameter optimization device 100 of this application embodiment, the real-time process parameters of the product in multiple processes are first acquired, and the real-time process parameters are preprocessed to obtain key process parameters. Further, based on the trained quality prediction model, multiple predicted quality indicators of the product are obtained according to the key process parameters. Further, under preset process constraints, the key process parameters are used as variables, and the goal is to minimize the weighted total loss function value. The weighted total loss function is then minimized to determine the target process parameter combination. The weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators. Finally, the real-time process parameters are optimized according to the target process parameter combination. Based on this, this application constructs a quality prediction model capable of outputting multiple quality indicators, inherently learns the complex relationships between quality indicators, and transforms the multi-objective optimization problem into minimizing a weighted total loss function based on the predicted value of the quality indicators under process constraints. An optimization algorithm is used for intelligent solution, thereby finding the optimal parameter combination with the best trade-off within the engineering feasible domain. This allows for rapid and automatic finding of the globally optimal process parameters across multiple processes, greatly shortening the cycle of existing process improvement and significantly improving production efficiency and product yield.
[0062] Please see Figure 3 This is a schematic diagram of the hardware structure of an electronic device 10 provided in an embodiment of this application.
[0063] The electronic device 10 provided in this application includes, but is not limited to, a memory 11, a processor 12, and a computer program stored in the memory 11 and executable on the processor 12, such as a process parameter optimization program. When the computer program is executed by the processor 12, it implements the process parameter optimization method as described in the above embodiments.
[0064] Figure 3 Only the electronic device 10 with memory 11 and processor 12 is shown. It will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 10, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.
[0065] In some embodiments of this application, the electronic device 10 can be communicatively connected to devices such as desktop computers, laptops, handheld computers, and cloud servers.
[0066] In some embodiments of this application, the electronic device 10 can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control device.
[0067] In some embodiments of this application, the electronic device 10 may further include network devices and / or client devices. These network devices include, but are not limited to, a single network server, a server group consisting of multiple network servers, and a cloud server based on cloud computing, consisting of a large number of hosts or network servers.
[0068] In some embodiments of this application, the network where the electronic device 10 is located includes, but is not limited to, the Internet, wide area network, metropolitan area network, local area network, virtual private network (VPN), etc.
[0069] In some embodiments of this application, memory 11 stores multiple computer-readable instructions to implement a process parameter optimization method, and processor 12 can execute multiple instructions to achieve: acquiring real-time process parameters of the product in multiple processes, and preprocessing the real-time process parameters to obtain key process parameters; based on the trained quality prediction model, obtaining multiple predicted quality indicators of the product according to the key process parameters; under preset process constraints, taking the key process parameters as variables, minimizing the weighted total loss function value as the objective, solving the weighted total loss function to determine the target process parameter combination, where the weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators.
[0070] Specifically, the processor 12's implementation method for the above instructions can be found in [reference needed]. Figure 1 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.
[0071] Those skilled in the art will understand that the schematic diagram is merely an example of the electronic device 10 and does not constitute a limitation on the electronic device 10. The electronic device 10 can be a bus topology or a star topology. The electronic device 10 may also include more or fewer other hardware or software than shown in the diagram, or different component arrangements. For example, the electronic device 10 may also include input / output devices, network access devices, etc.
[0072] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, in... Figure 3 The symbol is represented by only one arrow, but this does not mean that there is only one bus or one type of bus. The bus is configured to implement communication between memory 11 and at least one processor 12, etc.
[0073] It should be noted that electronic device 10 is only an example. Other existing or future electronic products that are suitable for this application should also be included within the scope of protection of this application and are incorporated herein by reference.
[0074] In some embodiments of this application, the processor 12 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 12 is the control unit of the electronic device 10, connecting various components of the electronic device 10 via various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., executing a process parameter optimization program) and calls data stored in the memory 11 to perform various functions and process data of the electronic device 10.
[0075] Processor 12 executes the operating system of electronic device 10 and various installed applications. Processor 12 executes these applications to implement the steps in each of the above-described embodiments of the process parameter optimization method, for example... Figure 1 The steps are shown.
[0076] For example, a computer program may be divided into one or more modules / units, one or more of which are stored in memory 11 and executed by processor 12 to complete this application. One or more modules / units may be a series of computer-readable instruction segments capable of performing a specific function, which describe the execution process of the computer program in electronic device 10. For example, the computer program may be divided into an acquisition module 110, a solution module 120, a prediction module 130, and an optimization module 140.
[0077] The integrated unit implemented as a software functional module described above can be stored in a computer-readable storage medium. This software functional module, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, computer equipment, or network device, etc.) or processor to execute a portion of a process parameter optimization method according to various embodiments of this application.
[0078] If the modules / units integrated in the electronic device 10 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 can also be implemented by a computer program instructing related hardware devices. 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.
[0079] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can 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, and other types of memory.
[0080] This application also provides a computer-readable storage medium (not shown), which stores computer-readable instructions that are executed by a processor in an electronic device 10 to implement a process parameter optimization method of any of the above embodiments.
[0081] Specifically, computer-readable storage media can be non-volatile or volatile. Computer-readable storage media include flash memory, portable hard drives, multimedia cards, card-type memories (e.g., SD memory, DX memory, etc.), magnetic storage, magnetic disks, optical disks, etc. In some embodiments, memory 11 can be an internal storage unit of electronic device 10, such as a portable hard drive of electronic device 10. In other embodiments, memory 11 can also be an external storage device of electronic device 10, such as a plug-in portable hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on electronic device 10. Memory 11 can be used not only to store application software and various types of data installed on electronic device 10, such as the code of a process parameter optimization program, but also to temporarily store data that has been output or will be output.
[0082] Furthermore, the computer-readable storage medium may primarily include a stored program area and a stored data area, wherein the stored program area may store the operating system, an application program required for at least one function, etc.; and the stored data area may store data created based on the use of blockchain nodes, etc.
[0083] In the embodiments of this application, it should be noted that, unless otherwise explicitly specified and limited, "multiple" means two or more.
[0084] In the embodiments of this application, it should be noted that, unless otherwise expressly specified and limited, the word "for example" is used to indicate an example, illustration, or description. Any embodiment or design scheme described as "for example" in the embodiments of this application should not be construed as being better or more advantageous than other embodiments or design schemes. Specifically, the use of the word "for example" is intended to present the relevant concepts in a specific manner.
[0085] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a connection that allows for communication; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.
[0086] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, 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. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features.
[0087] In the description of this application, it should be noted that, unless otherwise explicitly specified and limited, "and / or" 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. Furthermore, the character " / " in this application generally indicates that the preceding and following related objects have an "or" relationship.
[0088] Unless otherwise specified, all steps in this application may be performed sequentially or randomly. For example, if a method includes steps A and B, it means that the method may include steps A and B performed sequentially, or it may include steps B and A performed sequentially. For example, if a method may also include step C, it means that step C may be added to the method in any order. For example, the method may include steps A, B, and C, or it may include steps A, C, and B, or it may include steps C, A, and B, etc.
[0089] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.
[0090] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.
[0091] In the various embodiments of this application, the functional modules 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 unit can be implemented in hardware or in the form of hardware plus software functional modules.
[0092] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices described in the specification may also be implemented by a single unit or device through software or hardware.
[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this application without departing from the spirit and scope of the technical solutions of this application.
Claims
1. A method for optimizing process parameters, applied to electronic equipment, characterized in that, The method includes: The real-time process parameters of the product in multiple processes are obtained, and the real-time process parameters are preprocessed to obtain key process parameters. Based on the trained quality prediction model, multiple predicted quality indicators of the product are obtained according to the key process parameters. Under preset process constraints, the key process parameters are used as variables. The goal is to minimize the weighted total loss function value. The weighted total loss function is minimized to determine the target combination of process parameters. The weighted total loss function is the weighted sum of losses corresponding to each quality indicator among the multiple predicted quality indicators. The real-time process parameters are optimized based on the target combination of process parameters.
2. The process parameter optimization method as described in claim 1, characterized in that, The step of preprocessing the real-time process parameters to obtain the key process parameters includes: The real-time process parameters are cleaned and standardized to obtain the processed real-time process parameters; Based on the correlation between the processed real-time process parameters and each quality indicator, the processed real-time process parameters are filtered to obtain the key process parameters.
3. The process parameter optimization method as described in claim 1, characterized in that, The step of optimizing the real-time process parameters based on the target combination of process parameters includes: The target process parameter combination is applied to the actual production of the multiple processes, and the actual quality inspection data of the newly produced products are collected. The actual quality inspection data is compared with the predicted quality indicators to determine the quality yield of the newly produced products. If the quality yield of the newly produced product meets the preset requirements, the real-time process parameters in the multiple processes are updated to the target process parameter combination.
4. The process parameter optimization method as described in claim 3, characterized in that, The method further includes: if the quality yield of the newly produced product does not meet the preset requirements, updating the parameters of the quality prediction model based on the target process parameter combination.
5. The process parameter optimization method as described in claim 1, characterized in that, The training steps for the quality prediction model include: The historical process parameters and corresponding quality inspection data of the product in multiple processes are obtained, and the historical process parameters are preprocessed to obtain historical key process parameters. Based on preset engineering experience values, the contribution weight of the corresponding process stage is configured for the historical key parameters to obtain the weighted historical key process parameters. The weighted historical key process parameters are used as input features, and the quality indicators in the quality inspection data are used as input labels to construct and train the quality prediction model.
6. The process parameter optimization method as described in claim 5, characterized in that, The method further includes: obtaining the quality prediction model based on the training, and constructing the weighted total loss function.
7. The process parameter optimization method as described in claim 1, characterized in that, The plurality of processes include at least the sandblasting process of the product, the preceding process of the sandblasting process, and the following process.
8. A process parameter optimization device, characterized in that, Applied to electronic devices, the device includes: The acquisition module is used to acquire real-time process parameters of the product in multiple processes, and to preprocess the real-time process parameters to obtain key process parameters. The prediction module is used to obtain multiple predicted quality indicators of the product based on the trained quality prediction model and the key process parameters. The solution module is used to minimize the weighted total loss function value under preset process constraints, taking the key process parameters as variables, and determining the target combination of process parameters by minimizing the weighted total loss function value. The weighted total loss function is the weighted sum of losses corresponding to each of the multiple predicted quality indicators. The optimization module is used to optimize the real-time process parameters based on the target combination of process parameters.
9. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the process parameter optimization method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the process parameter optimization method as described in any one of claims 1 to 7.