Secondary chamfering process parameter determination method and device, electronic equipment and storage medium

Abstract

The disclosure provides a secondary chamfer process parameter determination method and device, electronic equipment and a storage medium. The method comprises: obtaining historical process data of a target product, the historical process data comprising secondary chamfer process parameters and edge polishing process parameters corresponding to the secondary chamfer process parameters; determining a parameter correspondence relationship between the secondary chamfer process parameters and the edge polishing process parameters based on the historical process data; and determining a parameter range of the secondary chamfer process parameters that causes the edge polishing process parameters to fall within an edge polishing process parameter range according to the edge polishing process parameter range to be met by the target product in the edge polishing process and the parameter correspondence relationship.

Description

Technical Field

[0001] This disclosure relates to the field of semiconductor manufacturing technology, and in particular to a method, apparatus, electronic device, and storage medium for determining parameters of a secondary chamfering process. Background Technology

[0002] In semiconductor manufacturing processes, the process parameters of the secondary chamfering step affect the processing results of the target product in the subsequent edge polishing step. For target products with strict edge polishing specifications, it is necessary to determine the process parameters in the secondary chamfering step that match those of the subsequent edge polishing steps.

[0003] Currently, the adjustment of secondary chamfering process parameters typically relies on manual experience. The adjustment personnel need to set multiple sets of secondary chamfering process parameters, and then transfer the products processed by the secondary chamfering process to the edge polishing process for verification. Based on the verification results, the secondary chamfering process parameters are then selected. This method makes it difficult to directly determine the range of secondary chamfering process parameters that will ensure the edge polishing process parameters meet the target range. Summary of the Invention

[0004] This disclosure provides a method for determining parameters of a secondary chamfering process. The method acquires historical process data of the target product, including secondary chamfering process parameters and corresponding edge polishing process parameters; determines the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process data; and determines the parameter range of the secondary chamfering process parameters that will cause the edge polishing process parameters to fall within the specified range, based on the range of edge polishing process parameters to be satisfied by the target product in the edge polishing process and the parameter correspondence.

[0005] This disclosure also provides a device for determining parameters of a secondary chamfering process, including a data acquisition module, a relationship determination module, and a parameter determination module. The data acquisition module acquires historical process data of the target product, including secondary chamfering process parameters and corresponding edge polishing process parameters. The relationship determination module determines the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process data. The parameter determination module determines the parameter range of the secondary chamfering process parameters that makes the edge polishing process parameters fall within the range of edge polishing process parameters, based on the range of edge polishing process parameters to be satisfied by the target product in the edge polishing process and the parameter correspondence.

[0006] This disclosure also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the processor executes the computer program to implement the above-described method for determining the parameters of the secondary chamfering process.

[0007] This disclosure also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for determining parameters of the secondary chamfering process. Attached Figure Description

[0008] Figure 1 This is a schematic diagram illustrating an application scenario of a device for determining parameters in a secondary chamfering process, as provided in an embodiment of this disclosure. Figure 2 A flowchart illustrating a method for determining parameters in a secondary chamfering process, provided in an embodiment of this disclosure; Figure 3 This is a schematic diagram of a process for removing outliers from historical process data, provided in an embodiment of this disclosure. Figure 4 This is a schematic diagram illustrating the process for determining the correspondence between historical process data and parameters, provided in an embodiment of this disclosure. Figure 5 This is a schematic diagram illustrating the process of establishing a data correlation model and determining parameter correspondences according to an embodiment of the present disclosure; Figure 6 A flowchart illustrating how to determine the parameter range for a secondary chamfering process based on the correspondence between parameters, as provided in this embodiment of the disclosure; Figure 7 A flowchart illustrating a method for determining the parameter combination of a secondary chamfering process based on the parameter quality correspondence, provided in an embodiment of this disclosure; Figure 8 A schematic diagram of the equipment debugging process for a combination of parameters in a secondary chamfering process provided in this embodiment of the present disclosure; Figure 9 This is a schematic diagram of a device for determining parameters of a secondary chamfering process provided in an embodiment of the present disclosure; Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0009] The technical solutions in this disclosure will now be clearly and completely described with reference to the accompanying drawings.

[0010] During the parameter debugging process of the secondary chamfering process, the existing method usually requires process engineers to set multiple sets of secondary chamfering process parameters based on experience, and then continue to the edge polishing process for verification after the product completes the secondary chamfering process. Because there is a process-to-process relationship between the secondary chamfering process and the edge polishing process, when the specifications of the edge polishing process change or the target product imposes more stringent requirements on the parameter range of the edge polishing process, the existing method is difficult to directly determine the parameter range of the secondary chamfering process that can meet the parameter range of the edge polishing process.

[0011] Furthermore, manual debugging typically requires repeated parameter setting, product processing, verification, and statistical analysis of test results, resulting in a lengthy debugging cycle. If the parameters for the secondary chamfering process do not match those for the subsequent edge polishing process, it can lead to an increase in the number of defective products, affecting process debugging efficiency and production stability. In addition, during parameter debugging, historical production data, quality inspection data, and equipment debugging data are often not fully utilized to determine the correspondence between parameters of preceding and subsequent processes, making it difficult for process engineers to quickly obtain a suitable range of secondary chamfering process parameters for the target product based on existing data.

[0012] Based on this, embodiments of this disclosure provide a method, apparatus, electronic device, and storage medium for determining secondary chamfering process parameters. The method acquires historical process data corresponding to the target product, determines the parameter correspondence between secondary chamfering process parameters and edge polishing process parameters based on the historical process data, and determines the parameter range of secondary chamfering process parameters that will cause the edge polishing process parameters to fall within that range, based on the range of edge polishing process parameters to be satisfied by the target product in the edge polishing process and the corresponding parameter relationship. Thus, historical process data can be used to provide a data basis for determining secondary chamfering process parameters, and the determined results of the secondary chamfering process parameters correspond to the parameter requirements of subsequent edge polishing processes.

[0013] like Figure 1 As shown, Figure 1 An application scenario for a secondary chamfering process parameter determination device 100 is illustrated. The secondary chamfering process parameter determination device 100 can be applied between the secondary chamfering process and the edge polishing process in a semiconductor manufacturing process, used to determine the range of secondary chamfering process parameters for a target product 200 based on historical data from preceding and following processes.

[0014] In this embodiment, the target product 200 is the product whose secondary chamfering process parameters need to be determined. The sample product 210 is a historical product that has already completed the secondary chamfering and edge polishing processes. The historical process dataset 300 is used to store historical process data corresponding to multiple sample products 210. The historical process data may include the secondary chamfering process parameters of the sample product 210 during the secondary chamfering process, and the edge polishing process parameters of the sample product 210 after the edge polishing process. The secondary chamfering process parameters may be set parameters, measurement parameters, detection parameters, or recording parameters in the secondary chamfering process; the edge polishing process parameters may be process parameters formed, detected, or recorded after the product undergoes the edge polishing process. The same sample product 210, the same batch, the same process flow number, or the same production record number can be used to establish a data correspondence between the secondary chamfering process parameters and the edge polishing process parameters.

[0015] The historical process data of the target product refers to the historical process data corresponding to the target product 200, covering the historical process data of sample products 210 that are of the same type, specification, process version, or meet the same edge polishing specifications as the target product 200; the target product 200 itself does not need to have completed the secondary chamfering process or edge polishing process. This limitation is used to describe the compatibility between the historical data and the target product 200.

[0016] The historical process dataset 300 originates from the secondary chamfering processing equipment 500, the edge polishing process inspection equipment, the quality inspection system, the product specification system, the manufacturing execution system, the process database, or the manual data entry system. The historical process dataset 300 can be stored in the secondary chamfering process parameter determination device 100 or in an external database and can be accessed by the secondary chamfering process parameter determination device 100.

[0017] The secondary chamfering process parameter determination device 100 determines the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process dataset 300. The parameter correspondence can be represented as a mapping table, data table, statistical relationship, correlation analysis results, machine learning model output results, or parameter selection rules. The secondary chamfering process parameter determination device 100 can also obtain the range of edge polishing process parameters that the target product 200 needs to satisfy in the edge polishing process, and based on this range and the parameter correspondence, determine the parameter range of the secondary chamfering process parameters that makes the edge polishing process parameters fall within the range.

[0018] In some embodiments, the secondary chamfering processing equipment 500 is used to perform a secondary chamfering process. The secondary chamfering process parameter determining device 100 outputs the determined range of secondary chamfering process parameters to the process engineer, and may further determine the combination of secondary chamfering process parameters, and convert the combination of secondary chamfering process parameters into control commands that the secondary chamfering processing equipment 500 can recognize. After executing the control command, the secondary chamfering processing equipment 500 can provide feedback on the actual process parameters. The secondary chamfering process parameter determining device 100 can update the control commands according to the deviation between the actual process parameters and the target process parameters, so as to form a closed-loop debugging process for the secondary chamfering process parameters.

[0019] like Figure 2 As shown, Figure 2 A flowchart illustrating a method for determining parameters in a secondary chamfering process is provided. This method can be executed by the control system of a secondary chamfering process parameter determining device 100 or a secondary chamfering processing equipment 500. The method may include steps S210 to S230.

[0020] In step S210, historical process data of the target product is obtained.

[0021] Specifically, the historical process data refers to the sample historical process data corresponding to the target product, which may include secondary chamfering process parameters and the corresponding edge polishing process parameters. The correspondence between the secondary chamfering process parameters and the edge polishing process parameters is established based on the same sample product 210. For example, if a sample product 210 uses or forms a set of secondary chamfering process parameters in the secondary chamfering process and forms a set of edge polishing process parameters after the subsequent edge polishing process, then the secondary chamfering process parameters and the edge polishing parameters can form a set of corresponding data.

[0022] In some embodiments, the data acquisition process may include real-time acquisition of measurement data for secondary chamfering process parameters, acquisition of measurement data for edge polishing process parameters after the corresponding product has been transferred to the edge polishing process, and statistics on the number of qualified products, the number of unqualified products, and the parameter types. Secondary chamfering process parameters may include at least one of diameter parameters, angle parameters, chamfer width parameters, edge shape parameters, and thickness-related parameters. Edge polishing process parameters may also include at least one of diameter parameters, angle parameters, chamfer width parameters, edge shape parameters, and thickness-related parameters.

[0023] In some embodiments, the target product 200 can be product X. The secondary chamfering process parameters corresponding to product X may include at least one of the following: diameter parameter (DIA-1), first edge dimension parameter (A-1), angle parameter (ANG-1), bevel width parameter (BC-1), and thickness-related parameter (TR-1). The edge polishing process parameters may include at least one of the following: diameter parameter (DIA-2), first edge dimension parameter (A-2), angle parameter (ANG-2), bevel width parameter (BC-2), and thickness-related parameter (TR-2). DIA, A, ANG, BC, and TR are used to distinguish different process parameter types; the suffix "-1" indicates a parameter in the secondary chamfering process, and the suffix "-2" indicates a parameter in the edge polishing process. The parameter names and parameter codes mentioned above are only used to illustrate the correspondence between different process parameters, and do not limit this disclosure to using the above parameter names or parameter codes.

[0024] Historical process data may also include quality inspection data. Quality inspection data can include whether the product is qualified, the number of non-conforming products, the number of losses, whether parameters exceed specifications, the degree of parameter deviation, the category of inspection items, the quality grade, or the type of process anomaly. Quality inspection data can correspond to sample product 210, or to a specific process record of sample product 210. Quality inspection data is used to subsequently determine the parameter quality correspondence, so as to select the parameter combination with the better quality evaluation results from multiple candidate secondary chamfering process parameter combinations.

[0025] In step S220, the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters is determined based on historical process data.

[0026] Specifically, historical process data is first organized to ensure that the secondary chamfering process parameters, edge polishing process parameters, and quality inspection data for the same sample product 210 are combined into a single sample record. Each sample record may include the sample product identifier, product type, process batch number, secondary chamfering process parameters, edge polishing process parameters, quality inspection results, and inspection time. Then, based on multiple sample records, the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters is determined.

[0027] Parameter correspondence includes inter-parameter correspondence and parameter quality correspondence. Inter-parameter correspondence characterizes the relationship between parameters in the secondary chamfering process and parameters in the edge polishing process. For example, if a certain angle parameter in the secondary chamfering process falls within a first value range, the corresponding angle parameter in the edge polishing process may fall within a second value range. Parameter quality correspondence characterizes the relationship between parameters in the secondary chamfering process, parameters in the edge polishing process, and the quality inspection results represented by the quality inspection data. For example, if a combination of secondary chamfering and edge polishing process parameters results in a small number of defective products in historical samples, then this parameter combination can correspond to a better quality inspection result.

[0028] In some embodiments, the parameter correspondence can be determined through statistical tables. For example, the distribution of edge polishing process parameters corresponding to different ranges of secondary chamfering process parameters can be statistically analyzed to form a parameter mapping table. In other embodiments, the parameter correspondence can be determined through a data correlation model. The data correlation model can take historical process data as input and output the correspondence between secondary chamfering process parameters and edge polishing process parameters, or it can output the predicted values ​​or quality evaluation results of edge polishing process parameters corresponding to candidate parameter combinations.

[0029] In step S230, based on the range of edge polishing process parameters and the corresponding relationship between parameters to be satisfied in the edge polishing process of the target product, the parameter range of the secondary chamfering process parameters that makes the edge polishing process parameters fall within the range of edge polishing process parameters is determined.

[0030] Specifically, the range of edge polishing process parameters that the target product 200 needs to meet in the edge polishing process can be determined by the product specification system, process card, customer specifications, edge polishing site inspection specifications, or engineer input. The range of edge polishing process parameters can include the upper and lower limits of a single parameter, or it can include target ranges corresponding to multiple parameters. For example, the target product 200 may need to meet the angle parameter range, diameter parameter range, and chamfer width parameter range in the edge polishing process.

[0031] The secondary chamfering process parameter determination device 100 determines which secondary chamfering process parameters will cause the subsequent edge polishing process parameters to fall within the target range based on parameter correspondence. If the parameter correspondence is represented by a historical sample table, the device filters sample records in the historical sample table where the edge polishing process parameters fall within the target range, and statistically analyzes the value ranges of the secondary chamfering process parameters corresponding to these sample records. If the parameter correspondence is represented by a model, the device can traverse, search, or solve backwards in the candidate space of secondary chamfering process parameters to filter the secondary chamfering process parameter values ​​output by the model that fall within the target range. If the parameter correspondence is represented by rules, the range of secondary chamfering process parameters can be directly determined according to the rules.

[0032] Through the above steps, the range of parameters for the secondary chamfering process of the target product 200 is not determined solely based on manual experience, but rather based on the correspondence between parameters of preceding and subsequent processes in historical process data, as well as the range of edge polishing process parameters that the target product 200 needs to meet in the edge polishing process.

[0033] like Figure 3 As shown, Figure 3 A process for removing outliers from historical process data is illustrated. This process can be executed before determining parameter correspondences based on historical process data, or it can be invoked and executed by the relationship determination module 120 after the original historical process data is written to the parameter database. The process may include steps S310 to S340.

[0034] S310, Obtain historical process data to be preprocessed.

[0035] Specifically, the historical process data to be preprocessed may include parameter data for the secondary chamfering process, parameter data for the edge polishing process, and quality inspection data. Because equipment malfunctions, inspection anomalies, recording errors, product flow abnormalities, or manual data entry errors may occur during production, the historical process data may contain outliers. Directly determining parameter correspondences based on historical process data containing outliers could lead to deviations from actual process rules. Therefore, outlier removal processing is performed on the historical process data before determining parameter correspondences.

[0036] S320, determine the mean and standard deviation of data for the same process parameter.

[0037] Specifically, the mean and standard deviation can be calculated for data under the same product type, the same process parameters, the same equipment, or the same process version. The mean represents the central tendency of the process parameter, and the standard deviation represents the dispersion of the data for that process parameter. The mean and standard deviation can be determined based on the secondary chamfering process parameter data, or they can be determined based on the edge polishing process parameter data.

[0038] S330, determine the upper and lower control limits based on the mean and standard deviation of the data.

[0039] In some embodiments, a 3σ control approach can be used to determine the upper and lower control limits. The upper control limit is the data mean plus three standard deviations, and the lower control limit is the data mean minus three standard deviations. For the same process parameter, if a certain process parameter data is greater than the upper control limit or less than the lower control limit, then that process parameter data is identified as outlier data.

[0040] S340, Remove abnormal data to obtain the preprocessing process dataset.

[0041] Specifically, abnormal data can be deleted from historical process data, or abnormal data can be marked as data that does not participate in the determination of parameter correspondence. For data records containing multiple parameters, if one of the key process parameters is identified as abnormal data, the entire data record can be removed. For non-key parameter anomalies that do not affect the determination of parameter correspondence, only the non-key parameter data can be removed, or missing value processing can be performed on the non-key parameter data. The preprocessed data can form a preprocessed process dataset 310.

[0042] In some embodiments, outlier removal can also be combined with product specification data. For example, if a process parameter exceeds the physically feasible or technologically permissible range of the corresponding parameter in the product specification system, that process parameter can be identified as outlier data. Outlier removal can also be combined with the status of testing equipment. For example, if a testing device is in a calibration malfunction or maintenance state during a specific time period, the testing data acquired during that time period can be marked as low-confidence data. Through outlier removal, the impact of outlier data on parameter correspondences can be reduced.

[0043] like Figure 4 As shown, Figure 4 This paper illustrates a process for determining the correspondence between historical process data and parameters. This process follows... Figure 3 The data preprocessing process is used to transform historical process data or preprocessed process dataset 310 into parameter correspondences and parameter quality correspondences that can be used subsequently. This process may include steps S410 to S450.

[0044] S410, organize the sample data in the historical process dataset.

[0045] Specifically, the historical process dataset 300 may include data records corresponding to multiple sample products 210. Each sample product 210 corresponds to at least one set of secondary chamfering process parameters and at least one set of edge polishing process parameters. If a sample product 210 uses a certain set of secondary chamfering process parameters in the secondary chamfering process and forms a certain set of edge polishing process parameters after the edge polishing process, then a corresponding data can be formed between the set of secondary chamfering process parameters and the set of edge polishing process parameters. During data processing, data of the same sample product 210 in the preceding and following processes are grouped into the same sample record based on the sample product identifier, batch identifier, process flow identifier, or recording time.

[0046] S420 associates quality inspection data with the corresponding sample records.

[0047] Specifically, the historical process dataset 300 may also include quality inspection data. Quality inspection data may include pass / fail indicators, fail / unqualified indicators, number of fail / unqualified products, number of losers, rework indicators, scrap indicators, parameter out-of-specification types, or quality grades. For the same sample product 210, if the secondary chamfering process parameters and the edge polishing process parameters both correspond to the quality inspection results of that sample product 210, then the quality inspection data is associated with the preceding and following process parameter data for that sample product 210.

[0048] S430 determines the correspondence between parameters based on the pairing data of parameters from the preceding and following processes.

[0049] Specifically, the correspondence between parameters is formed through pairing parameters from preceding and following processes in historical process data. For example, the secondary chamfering process parameter is used as the preceding parameter, and the edge polishing process parameter is used as the following parameter, and the value correspondence between the preceding and following parameters is statistically analyzed. When multiple secondary chamfering process parameters simultaneously affect the same edge polishing process parameter, the correspondence between parameters can be a multiple-input single-output relationship. When multiple secondary chamfering process parameters simultaneously affect multiple edge polishing process parameters, the correspondence between parameters can be a multiple-input multiple-output relationship.

[0050] S440 determines the correspondence between parameters and quality based on parameters of the preceding and following processes and quality inspection data.

[0051] Specifically, the parameter-quality correspondence is formed using upstream and downstream parameters and quality inspection data from historical process data. For example, the number of defective products under different combinations of secondary chamfering process parameters can be counted, as can the pass / fail probability corresponding to different combinations of secondary chamfering and edge polishing process parameters. The parameter-quality correspondence is used to evaluate the quality risk corresponding to candidate secondary chamfering process parameter combinations. The quality inspection results are those actually obtained from historical samples, while the quality evaluation results are those obtained by evaluating candidate parameter combinations based on the parameter-quality correspondence. The quality inspection results are the input facts from historical data, and the quality evaluation results are the processing results used for parameter combination selection.

[0052] S450 forms a parameter database for the secondary chamfering and edge polishing processes.

[0053] In some embodiments, the historical process dataset 300 may also include product specification data. The product specification data may come from a product specification system. The product specification data may include the range of edge polishing process parameters that the target product 200 needs to meet in the edge polishing process, and may also include the basic allowable range of parameters for the secondary chamfering process. By associating and storing the product specification data with the actual production data of the secondary chamfering process and the actual production data of the edge polishing process, a parameter database for the secondary chamfering process and the edge polishing process is formed.

[0054] This parameter database may include historical process data tables, product specification tables, data preprocessing record tables, parameter correspondence tables, model version tables, parameter recommendation record tables, and equipment debugging record tables. The historical process data table stores parameter data for sample product 210 during the secondary chamfering and edge polishing processes. The product specification table stores the parameter range to be met for target product 200 during the edge polishing process. The data preprocessing record table stores the mean, standard deviation, upper control limit, lower control limit, outlier identifier, and preprocessing time during the outlier removal process. The parameter correspondence table stores the correspondence between parameters and parameter quality correspondence. The model version table stores the model version, training data range, training time, evaluation indicators, and applicable product types for the data correlation model. The parameter recommendation record table stores the results of each parameter range determination and parameter combination determination. The equipment debugging record table stores control commands, actual process parameters, deviations, parameter compensation amounts, alarm information, and debugging termination conditions.

[0055] The aforementioned parameter database allows for the traceability of the basis for determining the parameter range of the secondary chamfering process. For example, once the parameter range for the secondary chamfering process corresponding to a target product 200 is determined, the target sample dataset 320 corresponding to that range, the correspondence between parameters, the parameter range of the edge polishing process, and the version of the data correlation model can be traced. When subsequent products exhibit quality abnormalities, the combination of secondary chamfering process parameters, control instructions, and actual process parameters used at that time can also be traced.

[0056] like Figure 5 As shown, Figure 5 This paper illustrates a process for establishing a data correlation model and determining parameter correspondences. This process can be used as... Figure 4 One specific implementation method for determining the correspondence between parameters is as follows, which can also be found in... Figure 4 This process is executed after the parameter database is created. The process may include steps S510 to S550.

[0057] S510, Obtain historical process data or preprocessed process datasets for model building.

[0058] Specifically, the data correlation model uses historical process data, preprocessing process dataset 310, or data from the parameter database as training data. Training data may include parameters for the secondary chamfering process, edge polishing process, product type, equipment number, batch information, quality inspection data, or other process conditions.

[0059] S520 establishes a data correlation model based on training data.

[0060] Specifically, the data correlation model is used to characterize the correspondence between the parameters of the secondary chamfering process and the parameters of the edge polishing process. The data correlation model can output the parameter correspondence, the predicted values ​​of the edge polishing process parameters under given secondary chamfering process parameters, and the quality evaluation results corresponding to candidate parameter combinations.

[0061] In some embodiments, the data correlation model includes a random forest model. A random forest model may include multiple decision trees. Each decision tree may be built based on a subset of samples and features from historical process data. Input features may include secondary chamfering process parameters, product type, equipment number, batch information, or other process conditions. Output targets may include edge polishing process parameters, whether the edge polishing process parameters fall within a target range, quality inspection results, or the number of defective products. The output of the random forest model can be obtained through averaging, voting, or weighted statistics of the results from multiple decision trees.

[0062] For example, the secondary chamfering process parameters DIA-1, A-1, ANG-1, BC-1, and TR-1 can be used as input features, and the edge polishing process parameters DIA-2, A-2, ANG-2, BC-2, and TR-2 can be used as output targets to train a random forest model. After training, when a candidate combination of secondary chamfering process parameters is input, the random forest model can output the corresponding predicted value of the edge polishing process parameters. If the predicted value falls within the range of edge polishing process parameters that the target product 200 needs to satisfy in the edge polishing process, then the candidate combination of secondary chamfering process parameters can enter the subsequent quality evaluation or parameter recommendation process.

[0063] S530 determines the correspondence between parameters based on a data correlation model.

[0064] Specifically, the data correlation model can be used to determine the correspondence between parameters, as well as the correspondence between parameter quality. When the data correlation model takes the secondary chamfering process parameters as input and the edge polishing process parameters as output, its output can be used to form the correspondence between parameters. When the data correlation model takes the secondary chamfering process parameters, the edge polishing process parameters, or a combination of both as input and the quality inspection results as output, its output can be used to form the correspondence between parameter quality.

[0065] S540 utilizes a data correlation model to assist in determining the parameter range for the secondary chamfering process in reverse.

[0066] Specifically, multiple candidate parameter values ​​or combinations can be generated within the candidate space of the secondary chamfering process parameters, and these candidate parameter values ​​or combinations can be input into the data correlation model. The data correlation model outputs the corresponding predicted values ​​of the edge polishing process parameters. If the predicted value corresponding to a certain candidate parameter value or combination falls within the range of the edge polishing process parameters, then the secondary chamfering process parameter corresponding to that candidate parameter value or combination is included in the statistical basis of the secondary chamfering process parameter range.

[0067] S550 updates the data correlation model based on newly added data.

[0068] In some embodiments, the data correlation model can be used in conjunction with a parameter database. The parameter database stores historical process data, preprocessing process data, product specification data, model training data, model version, and model output results. When a new sample product 210 completes the secondary chamfering and edge polishing processes, the new data is written to the parameter database. The secondary chamfering process parameter determination device 100 updates the data correlation model based on the new data, and can also retrain the data correlation model when preset data volume conditions, preset time conditions, or model evaluation index change conditions are met.

[0069] In some embodiments, the need to update the data correlation model can be determined based on model evaluation metrics. Model evaluation metrics may include prediction error of edge polishing process parameters, prediction accuracy of quality inspection results, parameter range hit rate, or candidate parameter combination validation pass rate. When the model evaluation metrics do not meet the preset model conditions, the data correlation model is retrained or fine-tuned using newly added historical process data. In this way, the data correlation model is updated as production data accumulates.

[0070] In some embodiments, the data correlation model may further include at least one of a regression model, a gradient boosting tree model, and a neural network model. A regression model can be used to express linear or nonlinear functional relationships between parameters, a gradient boosting tree model can be used to handle nonlinear mapping relationships between multiple parameters, and a neural network model can be used to handle complex coupling relationships between multiple secondary chamfering process parameters.

[0071] like Figure 6 As shown, Figure 6 This paper presents a process for determining the parameter range of a secondary chamfering process based on the correspondence between parameters. This process... Figure 4 or Figure 5 After obtaining the parameter correspondence, this process is executed to reverse-determine the parameter range of the preceding secondary chamfering process based on the edge polishing process parameter range that the target product 200 needs to meet in the edge polishing process. This process may include steps S610 to S630.

[0072] S610, Obtain the range of edge polishing process parameters that the target product needs to meet in the edge polishing process.

[0073] Specifically, the secondary chamfering process parameter determination device 100 reads the edge polishing process parameter range from the product specification system based on the product type, product specification number, or process version number of the target product 200. This edge polishing process parameter range may include the target range for a single edge polishing process parameter, or it may include target ranges corresponding to multiple edge polishing process parameters respectively.

[0074] S620, based on the correspondence between parameters, determines the target sample data from historical process data where the edge polishing process parameters fall within the range of edge polishing process parameters.

[0075] Specifically, if the correspondence between parameters is represented by a historical data table, then data records in the historical process dataset whose edge polishing process parameters fall within the target range are directly selected. The selected data records can form the target sample dataset 320. If the correspondence between parameters is represented by a statistical model, then the historical sample interval corresponding to the target edge polishing process parameter range is determined according to the statistical model, and then the corresponding samples are selected from the historical process dataset. If the correspondence between parameters is represented by a machine learning model, then the secondary chamfering process parameters that make the edge polishing process parameters fall within the target range are selected according to the model output results, and the corresponding historical sample records or simulated sample records are used as the target sample dataset 320.

[0076] When the edge polishing process involves multiple parameters, an intersection filtering method can be used, where data records whose parameters for multiple edge polishing processes all meet the corresponding target range are included in the target sample dataset 320. Alternatively, a hierarchical filtering method can be used, where data records whose key edge polishing process parameters meet the target range are first filtered, and then sorted according to the similarity of other edge polishing process parameters. A weighted scoring method can also be used, where weights are assigned based on the importance of different edge polishing process parameters, and the target sample data is determined based on the weighted scores.

[0077] S630, determine the parameter range of the secondary chamfering process parameters based on the corresponding secondary chamfering process parameters in the target sample data.

[0078] Specifically, the distribution of the secondary chamfering process parameters in the target sample dataset 320 is statistically analyzed. For example, the minimum and maximum values ​​of a certain secondary chamfering process parameter in the target sample data are statistically analyzed, and the interval between the minimum and maximum values ​​is determined as the parameter range of that secondary chamfering process parameter. Alternatively, the concentrated distribution interval, quantile interval, or interval satisfying the stability condition of the secondary chamfering process parameter can be statistically analyzed, and this interval can be used as the parameter range of that secondary chamfering process parameter. For multiple secondary chamfering process parameters, the parameter range of each secondary chamfering process parameter can be determined separately, or a joint range of multiple parameters can be formed.

[0079] In some embodiments, the range of parameters for the secondary chamfering process can be corrected based on the quality inspection data of the target sample data. For example, if some samples in the target sample data have edge polishing process parameters within the target range but the quality inspection results show that they are unqualified, the secondary chamfering process parameters corresponding to these samples can be removed from the range statistics, or the weight of these samples in the statistical process can be reduced. In this way, the range of parameters for the secondary chamfering process not only corresponds to the range of parameters for the edge polishing process, but also takes into account historical quality inspection results.

[0080] In some embodiments, when the target product 200 is product X, and the target range of the angle parameter ANG-2 in the edge polishing process is 12 to 30, target sample data in which ANG-2 falls within the range of 12 to 30 can be selected from historical process data. Then, the value distribution of the angle parameter ANG-1 in the secondary chamfering process corresponding to these target sample data can be statistically analyzed. If the statistical results show that when ANG-1 is between 18 and 36, ANG-2 subsequently falls more often between 12 and 30, then 18 to 36 can be determined as the suggested range for ANG-1, and a target value of 28 can be further given as a reference value for parameter combination screening or equipment debugging. The above values ​​are example values ​​in the embodiments and do not mean that the parameter range of the secondary chamfering process of this disclosure must be limited to this value range.

[0081] like Figure 7 As shown, Figure 7 This paper presents a process for determining the parameter combination for a secondary chamfering process based on the correspondence between parameter quality. This process... Figure 6 After determining the parameter range for the secondary chamfering process, this process is executed to filter recommended or feasible combinations of secondary chamfering process parameters within that range. This process may include steps S710 to S740.

[0082] S710, determine the candidate parameter combination within the parameter range of the secondary chamfering process.

[0083] Specifically, when Figure 6Once the parameter range for one or more secondary chamfering process parameters is determined, multiple candidate parameter combinations can be formed within that range. Candidate parameter combinations can include candidate values ​​for a single secondary chamfering process parameter, or combined values ​​for multiple secondary chamfering process parameters. Candidate parameter combinations can be determined through grid search, historical sample selection, segmented value selection, response surface methodology, orthogonal experimental design, or model-based recommendation methods.

[0084] Since secondary chamfering typically involves grinding, a change in one secondary chamfering parameter can affect other secondary chamfering parameters, and may also collectively affect edge polishing parameters. Therefore, candidate parameter combinations can be determined based on multiple secondary chamfering parameters. For example, in product X, candidate parameter combinations can be generated within at least some of the parameter ranges of DIA-1, A-1, ANG-1, BC-1, and TR-1. If the range of ANG-1 has been determined to be 18 to 36 based on the target sample data, multiple candidate parameter combinations can be generated within this range by combining the allowable ranges of DIA-1, A-1, BC-1, and TR-1.

[0085] S720, based on the parameter quality correspondence, determines the quality evaluation result corresponding to the candidate parameter combination.

[0086] Specifically, the parameter quality correspondence reflects the relationship between the parameters of the secondary chamfering process, the parameters of the edge polishing process, and the quality inspection results. For each candidate parameter combination, the quality evaluation result corresponding to that candidate parameter combination is determined based on the parameter quality correspondence. The quality evaluation result can be the predicted value of the quality inspection result, the pass probability, the fail probability, the predicted value of the number of non-conforming products, the predicted value of the number of losers, the quality grade, or whether the preset quality conditions are met.

[0087] In some embodiments, the parameter quality correspondence can be implemented using a table lookup method. For example, historical sample data that closely matches the candidate parameter combinations can be found in historical process data, and the quality evaluation result of the candidate parameter combination can be determined based on the quality inspection results in these historical sample data. In other embodiments, the parameter quality correspondence can be implemented using a model. For example, a quality evaluation model can be established that takes the candidate parameter combination as input and the predicted value of the quality inspection result as output, and the quality evaluation result corresponding to each candidate parameter combination can be determined based on the quality evaluation model.

[0088] S730, based on the quality evaluation results, determine the parameter combination for the secondary chamfering process from the candidate parameter combinations.

[0089] Specifically, a candidate parameter combination that meets preset quality conditions is selected from multiple candidate parameter combinations. Preset quality conditions may include a pass probability higher than a preset probability, a failure probability lower than a preset probability, a predicted number of non-conforming products lower than a preset number, a predicted number of losses lower than a preset number, or a quality level reaching a preset level. If multiple candidate parameter combinations meet the preset quality conditions, one of the candidate parameter combinations is further selected based on equipment commissioning costs, parameter variation range, historical stability, or similarity to the current equipment formula.

[0090] In some embodiments, candidate parameter combinations with a smaller number of non-conforming products corresponding to the quality evaluation results are determined as the parameter combinations for the secondary chamfering process. To avoid limiting the optimal parameter combination to a single mathematically optimal one, the parameter combinations for the secondary chamfering process in this embodiment are recommended parameter combinations determined under preset evaluation rules. These preset evaluation rules include at least one of the following: number of non-conforming products, quality inspection results, equipment adjustability, and parameter stability.

[0091] S740 outputs the parameter combination for the secondary chamfering process.

[0092] Specifically, the parameter combinations for the secondary chamfering process can be output to the process engineer for confirmation, or to the parameter debugging module 140 for automatic equipment debugging. The parameter combinations for the secondary chamfering process can be output in the form of a parameter list, equipment recipe, control command parameters, recommendation report, or database record. The output content can include the target value for each secondary chamfering process parameter, or it can include the allowable range for each secondary chamfering process parameter.

[0093] like Figure 8 As shown, Figure 8 This paper illustrates a device commissioning process for a combination of parameters in a secondary chamfering process. The process involves... Figure 7 After determining the parameter combination for the secondary chamfering process, this process is executed to send the parameter combination to the secondary chamfering processing equipment 500 and to dynamically adjust based on the actual execution results. This process may include steps S810 to S870.

[0094] S810, obtain the parameter combination for the secondary chamfering process.

[0095] Specifically, the parameter combination for the secondary chamfering process can be determined by... Figure 7 The process is defined as shown. This combination of parameters for the secondary chamfering process can include target values ​​for one or more secondary chamfering process parameters, or it can include target values ​​and allowable ranges. This combination of secondary chamfering process parameters can serve as the debugging target for the secondary chamfering processing equipment 500. If the parameter combination includes multiple parameters, these parameters can be organized according to the equipment recipe format.

[0096] S820 converts the combination of parameters for the secondary chamfering process into control commands that can be recognized by the secondary chamfering equipment.

[0097] Specifically, the parameter debugging module 140 can convert the combination of secondary chamfering process parameters into control instructions that the equipment can recognize, based on the communication protocol, equipment recipe format, or control system interface of the secondary chamfering processing equipment 500. These control instructions may include parameter writing instructions, recipe update instructions, equipment debugging instructions, or process execution instructions. The control instructions may include equipment number, parameter name, parameter target value, parameter allowable range, execution time, or instruction verification information.

[0098] S830 sends control commands to the secondary chamfering equipment.

[0099] Specifically, the parameter debugging module 140 can send control commands to the secondary chamfering processing equipment 500 via an industrial network, equipment communication interface, manufacturing execution system interface, or equipment controller interface. Upon receiving the control commands, the secondary chamfering processing equipment 500 can update its equipment formula or enter parameter debugging mode. If manual confirmation is required, the control commands can be sent to the equipment control interface for operator confirmation before execution.

[0100] S840: Obtain the actual process parameters after the secondary chamfering processing equipment executes the control command.

[0101] Specifically, after the secondary chamfering processing equipment 500 executes control commands, it can generate actual process parameters. These actual process parameters can come from equipment sensors, equipment logs, process detection results, or debugging detection results. The parameter debugging module 140 can acquire the actual process parameters in real time or periodically. In some embodiments, the feedback process can be continuously performed with a sampling period of 1 second. This 1-second sampling period is a sampling period in some embodiments and does not constitute a fixed limitation on the sampling period of this disclosure.

[0102] S850 determines whether the deviation between the actual process parameters and the target process parameters meets the preset deviation conditions.

[0103] Specifically, the target process parameters can be determined based on a combination of secondary chamfering process parameters. The deviation between the actual process parameters and the target process parameters can be the absolute value of their difference or the rate of change of the deviation. The rate of change of the deviation can be determined based on the ratio of the difference between the actual process parameters and the target process parameters to the target process parameters. In some embodiments, dynamic fine-tuning can be triggered when the rate of change of the deviation between the actual value and the target value is greater than 5%. This 5% is a triggering threshold in the embodiments and does not constitute a fixed limitation on the preset deviation conditions of this disclosure.

[0104] In some embodiments, the preset deviation condition may include the deviation reaching the dynamic fine-tuning trigger threshold and not exceeding the automatically adjustable range of the secondary chamfering processing equipment 500; if the deviation meets the preset deviation condition, step S860 is executed; if the deviation does not reach the dynamic fine-tuning trigger threshold, step S870 may be executed to determine whether the actual process parameter has stabilized within the stable range corresponding to the target process parameter; if the deviation exceeds the automatically adjustable range or the equipment returns to an abnormal state, alarm information may be output and manual processing may be initiated.

[0105] S860 determines the parameter compensation amount when the deviation meets the preset deviation conditions, and updates the control command according to the parameter compensation amount.

[0106] Specifically, the parameter compensation amount can include the compensation magnitude and the compensation direction. The compensation magnitude is determined based on the difference between the actual process parameter and the target process parameter. The compensation direction is determined based on whether the actual process parameter is larger or smaller than the target process parameter. For example, when the actual process parameter is larger than the target process parameter, the compensation direction can be used to decrease the corresponding equipment setting value; when the actual process parameter is smaller than the target process parameter, the compensation direction can be used to increase the corresponding equipment setting value. After updating the control command according to the parameter compensation amount, the parameter adjustment module 140 can send the updated control command to the secondary chamfering processing equipment 500.

[0107] S870 determines whether the actual process parameters are stable within the stable range corresponding to the target process parameters.

[0108] Specifically, after updating the control instructions, the parameter debugging module 140 continues to acquire the actual process parameters fed back by the secondary chamfering processing equipment 500, and determines whether the actual process parameters fall within the preset stable range corresponding to the target process parameters. The preset stable range is determined based on the allowable deviation of the target process parameters, for example, a range of ±1% of the target value; this value is only an example and does not constitute a limitation on the preset stable range of this disclosure. If the actual process parameters fall within the preset stable range and remain stable within a preset duration, the current parameter debugging process ends. If the actual process parameters do not fall within the preset stable range, the process returns to execution S840 to continue acquiring the actual process parameters, and deviation judgment and compensation updates are performed according to S850 to S860; if the actual process parameters exceed the automatically adjustable range of the secondary chamfering processing equipment 500, or the secondary chamfering processing equipment 500 returns to an abnormal state, an alarm message is output and manual processing is initiated.

[0109] In some embodiments, when the actual process parameters exceed the automatically adjustable range of the secondary chamfering processing equipment 500, or when the secondary chamfering processing equipment 500 cannot complete the automatic adjustment according to the updated control instructions, the parameter debugging module 140 can output alarm information. The alarm information may include the target product identifier, the secondary chamfering processing equipment 500 identifier, the abnormal parameter name, the target process parameter, the actual process parameter, the deviation, the parameter compensation amount, the equipment return status, and the suggested handling method. The parameter debugging module 140 can also provide a manual interactive interface for operators to input manual adjustment parameters, confirm continued automatic debugging, pause debugging, or end debugging. Manually input data, alarm information, and subsequent debugging results can be written to the parameter database for subsequent process traceability and model updates.

[0110] like Figure 9 As shown, Figure 9 The structure of a secondary chamfering process parameter determination device 100 is shown. The secondary chamfering process parameter determination device 100 may include a data acquisition module 110, a relationship determination module 120, and a parameter determination module 130. In some embodiments, the secondary chamfering process parameter determination device 100 may further include a parameter adjustment module 140.

[0111] The data acquisition module 110 is used to acquire historical process data of the target product. The historical process data includes secondary chamfering process parameters and corresponding edge polishing process parameters. The data acquisition module 110 can acquire data from the secondary chamfering processing equipment 500, edge polishing process inspection equipment, quality inspection system, product specification system, production database, or data storage module. The data acquisition module 110 can also acquire the range of edge polishing process parameters that the target product 200 needs to meet in the edge polishing process.

[0112] The relationship determination module 120 is used to determine the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on historical process data. The relationship determination module 120 can perform data preprocessing, parameter correspondence determination, data correlation model establishment, model invocation, and model updating. The relationship determination module 120 can divide the parameter correspondence into inter-parameter correspondence and parameter quality correspondence. Inter-parameter correspondence is used to support parameter range determination, while parameter quality correspondence is used to support parameter combination filtering.

[0113] The parameter determination module 130 is used to determine the parameter range of the secondary chamfering process parameters that make the edge polishing process parameters fall within the edge polishing process parameter range, based on the parameter range and parameter correspondence that the target product 200 needs to satisfy in the edge polishing process. The parameter determination module 130 can determine the target sample dataset 320 from historical process data based on the parameter correspondence, and determine the parameter range based on the corresponding secondary chamfering process parameters in the target sample dataset 320. The parameter determination module 130 can also generate candidate parameter combinations within this parameter range and determine the quality evaluation result corresponding to the candidate parameter combinations based on the parameter quality correspondence.

[0114] The parameter debugging module 140 is used to determine the combination of secondary chamfering process parameters within the parameter range of the secondary chamfering process parameters, convert the combination of secondary chamfering process parameters into control commands that can be recognized by the secondary chamfering processing equipment 500, and send the control commands to the secondary chamfering processing equipment 500. The parameter debugging module 140 can also obtain the actual process parameters after the secondary chamfering processing equipment 500 executes the control commands, and when the deviation between the actual process parameters and the target process parameters meets the preset deviation conditions, determine the parameter compensation amount, and update the control commands according to the parameter compensation amount.

[0115] The modules in the secondary chamfering process parameter determination device 100 can be implemented by software functional units, hardware circuits, or by a processor executing a computer program. The data acquisition module 110, relationship determination module 120, parameter determination module 130, and parameter debugging module 140 can be integrated into the same electronic device or distributed across multiple devices to work collaboratively. For example, the data acquisition module 110 can be located in a production line data acquisition server, the relationship determination module 120 and parameter determination module 130 can be located in a process parameter analysis server, and the parameter debugging module 140 can be located in an equipment control server or the controller of the secondary chamfering processing equipment 500.

[0116] like Figure 10 As shown, Figure 10 The structure of an electronic device 1000 is shown. The electronic device 1000 may include a processor 1010 and a memory 1020. The processor 1010 and the memory 1020 may be connected via a bus, a communication interface, or other data path.

[0117] The memory 1020 stores a computer program that can be executed by the processor 1010. When the processor 1010 executes the computer program, it can implement the above-described method for determining the parameters of the secondary chamfering process. For example, the processor 1010 can perform processes such as acquiring historical process data, removing outliers from the historical process data, determining parameter correspondences, establishing a data correlation model, determining the range of secondary chamfering process parameters, determining candidate parameter combinations, determining the combination of secondary chamfering process parameters, generating control instructions, acquiring actual process parameters, judging deviation conditions, determining parameter compensation amounts, and updating control instructions.

[0118] Electronic device 1000 can be a server, industrial computer, process control equipment, production line data processing equipment, control unit of secondary chamfering equipment 500, or edge computing device. Electronic device 1000 can communicate with secondary chamfering equipment 500, and also with edge polishing process inspection equipment, quality inspection system, product specification system, or manufacturing execution system. Electronic device 1000 can acquire data and send control commands via wired network, industrial Ethernet, serial communication interface, fieldbus, or other communication methods.

[0119] The memory 1020 may include a read-only memory, random access memory, flash memory, solid-state drive, disk, or other storage components capable of storing computer programs and data. The memory 1020 may store historical process datasets 300, preprocessing process datasets 310, target sample datasets 320, data correlation models, parameter databases, product specification data, model training records, parameter recommendation records, equipment debugging records, and alarm records.

[0120] In some embodiments, the electronic device 1000 can periodically perform model update tasks. For example, when the number of newly added sample products in the parameter database reaches a preset number, or the model evaluation index does not meet the preset model conditions, or the preset update cycle is reached, the processor 1010 can update the data correlation model using newly added historical process data. The updated data correlation model can be used to determine the parameter range and parameter combination selection for the secondary chamfering process of the subsequent target product.

[0121] In some embodiments, the electronic device 1000 can also generate a parameter debugging report. The parameter debugging report may include the target product type, the parameter range for the edge polishing process, the parameter range for the secondary chamfering process, candidate parameter combinations, the parameter combination for the secondary chamfering process, quality evaluation results, control instructions, actual process parameters, parameter compensation amounts, number of debugging attempts, whether an alarm was triggered, and the final debugging results. The parameter debugging report can be stored in a parameter database for subsequent process traceability and process optimization.

[0122] Computer programs can also be stored in computer-readable storage media. When the computer program is executed by the processor 1010, the method for determining the parameters of the secondary chamfering process described above can be implemented. The computer-readable storage media can be non-transient storage media, including read-only memory, flash memory, solid-state memory, magnetic disk, or optical disk.

[0123] It should be noted that the execution order of each process step in the above embodiments can be adjusted according to actual processing needs. For example, outliers in historical process data can be removed first, and then a parameter database can be established; alternatively, the original historical process data can be written into the parameter database first, and then the preprocessing process can be called in the relationship determination module 120. Furthermore, the data correlation model can be pre-trained before the target product 200 enters the parameter determination process, or it can be established immediately based on the historical process data of the corresponding product type after the target product 200 enters the parameter determination process. These changes do not affect the technical essence of this disclosure, which determines the correspondence between process parameters before and after based on historical process data, and determines the range of secondary chamfering process parameters based on the range of edge polishing process parameters.

Claims

1. A method for determining parameters of a secondary chamfering process, characterized in that, include: Obtain historical process data for the target product, including secondary chamfering process parameters and edge polishing process parameters corresponding to the secondary chamfering process parameters; Based on the historical process data, determine the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters; Based on the range of edge polishing process parameters to be satisfied by the target product in the edge polishing process and the corresponding relationship of the parameters, determine the range of secondary chamfering process parameters that will make the edge polishing process parameters fall within the range of edge polishing process parameters.

2. The method for determining parameters of the secondary chamfering process according to claim 1, characterized in that, The historical process data also includes quality inspection data; The parameter correspondence includes the correspondence between parameters and the correspondence between parameter quality. The correspondence between parameters is used to characterize the correspondence between the parameters of the secondary chamfering process and the parameters of the edge polishing process. The parameter quality correspondence is used to characterize the correspondence between the secondary chamfering process parameters, the edge polishing process parameters, and the quality inspection results characterized by the quality inspection data.

3. The method for determining parameters of the secondary chamfering process according to claim 2, characterized in that, The step of determining the range of parameters for the secondary chamfering process that ensures the edge polishing process parameters fall within the range of edge polishing process parameters, based on the range of edge polishing process parameters to be satisfied by the target product in the edge polishing process and the corresponding parameter relationship, includes: Based on the correspondence between the parameters, target sample data in which the edge polishing process parameters fall within the range of the edge polishing process parameters are determined from the historical process data; Based on the corresponding secondary chamfering process parameters in the target sample data, determine the parameter range of the secondary chamfering process parameters.

4. The method for determining parameters of the secondary chamfering process according to claim 2, characterized in that, After determining the parameter range for the secondary chamfering process, the method further includes: Candidate parameter combinations are determined within the parameter range of the secondary chamfering process; Based on the parameter quality correspondence, the quality evaluation result corresponding to the candidate parameter combination is determined; Based on the quality evaluation results, the parameter combination for the secondary chamfering process is determined from the candidate parameter combinations.

5. The method for determining parameters of the secondary chamfering process according to claim 4, characterized in that, After determining the combination of parameters for the secondary chamfering process, the method further includes: The combination of parameters for the secondary chamfering process is converted into control commands that can be recognized by the secondary chamfering equipment, and the control commands are sent to the secondary chamfering equipment. Obtain the actual process parameters of the secondary chamfering equipment after executing the control command; If the deviation between the actual process parameters and the target process parameters meets the preset deviation condition, the parameter compensation amount is determined, and the control command is updated according to the parameter compensation amount. The parameter compensation amount includes the compensation magnitude and compensation direction determined based on the difference between the actual process parameters and the target process parameters.

6. The method for determining parameters of the secondary chamfering process according to claim 1, characterized in that, Before determining the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process data, the method further includes: The historical process data is subjected to outlier removal processing to obtain preprocessed process data; The step of determining the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process data includes: The parameter correspondence is determined based on the preprocessing process data; The outlier removal process includes determining the upper and lower control limits of the same process parameter based on the data mean and standard deviation of the same process parameter, and identifying process parameters that are greater than the upper control limit or less than the lower control limit as outlier data.

7. The method for determining parameters of the secondary chamfering process according to claim 1, characterized in that, The step of determining the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process data includes: A data correlation model is established based on the historical process data. The data correlation model is used to characterize the correspondence between the secondary chamfering process parameters and the edge polishing process parameters. The parameter correspondence is determined based on the data correlation model; The data correlation model includes at least one of the following: random forest model, regression model, gradient boosting tree model, and neural network model.

8. A device for determining parameters in a secondary chamfering process, characterized in that, include: The data acquisition module is configured to acquire historical process data of the target product, wherein the historical process data includes secondary chamfering process parameters and edge polishing process parameters corresponding to the secondary chamfering process parameters; The relationship determination module is configured to determine the parameter correspondence between the secondary chamfering process parameters and the edge polishing process parameters based on the historical process data; The parameter determination module is configured to determine the parameter range of the secondary chamfering process parameters that make the edge polishing process parameters fall within the edge polishing process parameter range, based on the range of edge polishing process parameters to be satisfied by the target product in the edge polishing process and the parameter correspondence.

9. The device for determining parameters of the secondary chamfering process according to claim 8, characterized in that, The device further includes: The parameter debugging module is configured to determine the combination of secondary chamfering process parameters within the parameter range of the secondary chamfering process parameters, convert the combination of secondary chamfering process parameters into control commands that can be recognized by the secondary chamfering processing equipment, and send the control commands to the secondary chamfering processing equipment.

10. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores computer programs that can be executed by the processor; When the processor executes the computer program, it implements the method for determining the parameters of the secondary chamfering process as described in any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program; When the computer program is executed by the processor, it implements the method for determining the parameters of the secondary chamfering process as described in any one of claims 1 to 7.

Images