CN-121998477-A - Process parameter optimization method and device, electronic equipment and storage medium

CN121998477ACN 121998477 ACN121998477 ACN 121998477ACN-121998477-A

Abstract

The application provides a process parameter optimization method, a device, electronic equipment and a storage medium, wherein the process parameter optimization method comprises the steps of obtaining real-time process parameters of a product in a plurality of processes, and preprocessing the real-time process parameters to obtain key process parameters; the method comprises the steps of obtaining a plurality of predicted quality indexes of a product based on a trained quality prediction model according to key process parameters, taking the key process parameters as variables under a preset process constraint condition, taking a minimum weighted total loss function value as a target, carrying out minimum solution on the weighted total loss function, determining a target process parameter combination, wherein the weighted total loss function is a loss weighted sum corresponding to each quality index in the plurality of predicted quality indexes, and optimizing the real-time process parameters according to the target process parameter combination. Based on the method, the global optimal process parameters crossing a plurality of processes can be dynamically found, the period of improving the existing process is shortened, and the production efficiency and the product yield are improved.

Inventors

ZHOU JING
ZENG TAO
ZHOU BING
ZHANG NENGWEI
Liao Qianshen
YUE WEILING

Assignees

鸿富锦精密电子（成都）有限公司

Dates

Publication Date: 20260508
Application Date: 20251223

Claims (10)

1. A process parameter optimization method applied to an electronic device, the method comprising: acquiring real-time process parameters of a product in a plurality of processes, and preprocessing the real-time process parameters to obtain key process parameters; Based on the trained quality prediction model, obtaining a plurality of prediction quality indexes of the product according to the key process parameters; Under a preset process constraint condition, taking the key process parameters as variables, taking a minimization weighted total loss function value as a target, carrying out minimization solution on the weighted total loss function, and determining a target process parameter combination, wherein the weighted total loss function is a loss weighted sum corresponding to each quality index in the plurality of prediction quality indexes; and optimizing the real-time process parameters according to the target process parameter combination.
2. The process parameter optimization method of claim 1, wherein the step of preprocessing the real-time process parameters to obtain key process parameters comprises: performing data cleaning and standardization treatment on the real-time process parameters to obtain treated real-time process parameters; And screening the processed real-time process parameters based on the correlation between the processed real-time process parameters and each quality index to obtain the key process parameters.
3. The process parameter optimization method of claim 1, wherein said step of optimizing said real-time process parameters based on said target process parameter combination comprises: applying the target process parameter combination to actual production of the plurality of processes, and collecting actual quality detection data of newly produced products; Comparing the actual quality detection data with the predicted quality index to determine the quality yield of the newly produced product; And if the quality yield of the newly produced product meets the preset requirement, updating the real-time process parameters in the multiple processes into the target process parameter combination.
4. The process parameter optimization method of claim 3, further comprising updating parameters of said quality prediction model based on a target process parameter combination if a quality yield of said newly produced product does not meet said preset requirement.
5. The process parameter optimization method of claim 1, wherein the training step of the quality prediction model comprises: acquiring historical process parameters and corresponding quality detection data of the product in a plurality of processes, and preprocessing the historical process parameters to obtain historical key process parameters; Based on a preset engineering experience value, configuring contribution degree weights of corresponding process stages for the historical key parameters to obtain weighted historical key process parameters; and taking the weighted historical key process parameters as input characteristics, taking quality indexes in the quality detection data as input labels, and constructing and training the quality prediction model.
6. The process parameter optimization method of claim 5, further comprising constructing the weighted total loss function based on the training to obtain the quality prediction model.
7. The method of claim 1, wherein the plurality of processes includes at least a blasting process of the product, a pre-process and a post-process of the blasting process.
8. A process parameter optimization apparatus for use in an electronic device, the apparatus comprising: The acquisition module is used for acquiring real-time process parameters of the product in a plurality of processes and preprocessing the real-time process parameters to obtain key process parameters; The prediction module is used for obtaining a plurality of prediction quality indexes of the product according to the key process parameters based on the trained quality prediction model; The solving module is used for taking the key process parameters as variables under the preset process constraint condition, taking the minimized weighted total loss function value as a target, performing minimized solving on the weighted total loss function, and determining a target process parameter combination, wherein the weighted total loss function is a loss weighted sum corresponding to each quality index in the plurality of prediction quality indexes; and the optimization module is used for optimizing the real-time process parameters according to the target process parameter combination.
9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed by the processor implements the process parameter optimization method according to any one of claims 1 to 7.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the process parameter optimization method according to any one of claims 1 to 7.

Description

Process parameter optimization method and device, electronic equipment and storage medium Technical Field The application relates to the technical field of numerical control machining, in particular to a process parameter optimization method, a device, electronic equipment and a storage medium. Background In complex multi-process product manufacturing processes, such as precision electronics manufacturing, the final quality of the product is determined by the process parameters in the manufacturing process. However, the current process parameter setting mainly relies on the personal experience of the engineer to initialize and adjust the parameters, which essentially belongs to a "trial and error method". The method is high in subjectivity and poor in replicability, and the effect can be evaluated only by waiting for the product to flow through all the processes and finishing quality detection after each parameter adjustment, the optimization period is long, the regulation and control are delayed, the production stability is poor, the yield fluctuation is large, and the method cannot adapt to the production requirement of rapid iteration. Disclosure of Invention In view of the above, it is necessary to provide a process parameter optimization method, device, electronic equipment and storage medium, so as to solve the technical problems that the existing process parameter setting method has strong subjectivity, poor replicability, long optimization period, lag in regulation, poor production stability, large yield fluctuation and inability to adapt to the production requirement of rapid iteration. The application provides a process parameter optimization method which is applied to electronic equipment, and the method comprises the steps of obtaining real-time process parameters of a product in a plurality of processes, preprocessing the real-time process parameters to obtain key process parameters, obtaining a plurality of predicted quality indexes of the product according to the key process parameters based on a trained quality prediction model, taking the key process parameters as variables under a preset process constraint condition, taking a minimum weighted total loss function value as a target, carrying out minimum solution on the weighted total loss function, determining a target process parameter combination, wherein the weighted total loss function is a loss weighted sum corresponding to each quality index in the plurality of predicted quality indexes, and optimizing the real-time process parameters according to the target process parameter combination. In the process parameter optimization method, real-time process parameters of a product in a plurality of processes are firstly obtained, the real-time process parameters are preprocessed to obtain key process parameters, a plurality of predicted quality indexes of the product are further obtained according to the key process parameters based on a trained quality prediction model, the key process parameters are further used as variables under a preset process constraint condition, a minimum weighted total loss function value is used as a target, the minimum weighted total loss function is solved, a target process parameter combination is determined, the weighted total loss function is a loss weighted sum corresponding to each quality index in the plurality of predicted quality indexes, and finally the real-time process parameters are optimized according to the target process parameter combination. Based on the method, the quality prediction model capable of outputting a plurality of quality indexes is constructed, the complex relation among the quality indexes is internally learned, the multi-objective optimization problem is converted into a weighted total loss function based on the quality index prediction value under the process constraint, and the optimization algorithm is utilized to carry out intelligent solution, so that the optimal parameter combination of optimal balance is found in the engineering feasibility domain, the global optimal process parameters crossing a plurality of processes can be quickly and automatically found, the period of improving the existing process is greatly shortened, and the production efficiency and the product yield are remarkably improved. In some embodiments of the present application, the step of preprocessing the real-time process parameter to obtain a key process parameter includes performing data cleaning and normalization processing on the real-time process parameter to obtain a processed real-time process parameter, and screening the processed real-time process parameter based on correlation between the processed real-time process parameter and each quality index to obtain the key process parameter. In some embodiments of the present application, the step of optimizing the real-time process parameters according to the target process parameter combination includes applying the target process parameter combi