Method, device and computer-readable medium for setting processing parameters of production equipment

Abstract

The invention relates to the field of industrial big data technology, and in particular to a workpiece data processing method, device and computer-readable medium, which are used to accurately determine the relationship between the processing parameters of production equipment, ambient environment condition data and the quality inspection results of the workpiece, and then determine the optimal processing parameters based on the relationship. A workpiece data method includes: obtaining processing condition data, quality attribute values ​​and quality inspection result data of each of a plurality of workpieces processed by a production equipment, wherein the processing condition data of a workpiece includes: the processing parameters used by the production equipment when processing the workpiece and the ambient environment condition data of the production equipment when processing the workpiece; determining a first relationship between the quality attribute value of the workpiece processed by the production equipment and the ambient environment condition data of the production equipment when processing the workpiece and the processing parameters of the production equipment; and determining a second relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production equipment.

Description

Technical Field

[0001] The present invention relates to industrial big data technology, and in particular to a workpiece data processing method, device and computer-readable medium. Background Art

[0002] The quality of the workpiece processed by a production equipment depends largely on the parameter settings of the production equipment. Taking CNC machine tools (CNC) as an example, the settings of CNC processing parameters, such as changes in cutting position, cutting angle, axis height and clamping position, will lead to changes in the quality attribute values ​​of the processed workpiece. A workpiece that passes the quality inspection must pass the quality inspection, that is, its quality attribute values ​​need to pass the quality inspection (usually, different types of workpieces need to be inspected for different quality attributes, such as: for worm gears in the automotive industry, the average torque, transmission ratio, peak speed, etc. need to be inspected). Therefore, the setting of production equipment processing parameters is crucial for workpiece manufacturing, especially for the optimization of workpiece quality during the workpiece manufacturing process in the finishing industry.

[0003] At present, the processing parameters of production equipment such as CNC are usually set based on human experience, and the optimal setting of the parameters may be affected by the acceptable range of various quality attributes of the workpiece. Therefore, it is often difficult to achieve the optimal setting of the processing parameters by relying solely on human experience. In addition, the quality of the workpiece is easily affected by the surrounding environmental conditions, such as the influence of temperature and humidity during the workpiece manufacturing process. Therefore, when setting the processing parameters, the changes in the surrounding environmental conditions must also be taken into account, which further increases the difficulty of setting the processing parameters.

[0004] As mentioned above, workpiece quality is related to the setting of production equipment processing parameters, and it is difficult to optimize workpiece quality by currently only relying on human experience to set processing parameters. Workpiece quality is often measured by quality inspection, so how to set production equipment processing parameters to improve the pass rate of workpiece quality inspection is an urgent problem to be solved.

[0005] One possible method is to achieve automatic optimal setting of the processing parameters of the production equipment by determining the relationship between the processing parameters of the production equipment and the pass rate of the workpiece quality inspection. The relationship between the processing parameters of the workpiece production equipment and the pass rate of the workpiece quality inspection can be determined by a machine learning algorithm. One way is to use a supervised learning algorithm to establish a model and determine the relationship by learning a large amount of labeled data. However, for the manufacturing industry, data collection and labeling are often difficult and costly, especially in factories that are not digitized. If you want to determine the relationship between the processing parameters of a production equipment during the manufacturing process and the pass rate of the quality inspection of the workpiece it processes, it often requires a large amount of data. Assume that you want to determine the pass rate of the worm gear under the following CNC processing parameter settings: the y-axis cutting position is 91, the helix angle is -2.97, the z-axis height is 209.8, the ambient temperature is 7 degrees Celsius, and the humidity is 70%. To complete the above processing, it is usually necessary to collect hundreds of worm gear samples to ensure statistical significance, so that the accurate pass rate of the worm gear quality inspection can be calculated under a single combination of CNC processing parameters and ambient conditions. However, in production practice, it is usually impossible to obtain sufficient data. For each combination of CNC machining parameters and ambient conditions, there are often only a few annotated data. It is difficult to obtain the accurate relationship between CNC machining parameters and workpiece quality inspection results under different ambient conditions with the help of traditional regression models. Summary of the invention

[0006] The embodiments of the present invention provide a workpiece data processing method, device and computer-readable medium for accurately determining the relationship between the processing parameters of production equipment, ambient environment condition data and the quality inspection results of the workpiece, and then determining the optimal processing parameters based on the relationship.

[0007] In a first aspect, a workpiece data processing method is provided, comprising: obtaining processing condition data, quality attribute values ​​and quality inspection result data of each of a plurality of workpieces processed by a production device, wherein the processing condition data of a workpiece comprises: processing parameters used by the production device when processing the workpiece and ambient environment condition data of the production device when processing the workpiece; determining a first relationship between the quality attribute value of the workpiece processed by the production device and the ambient environment condition data and the processing parameters of the production device when processing the workpiece based on the processing condition data and quality attribute values ​​of each of the plurality of workpieces; determining a second relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production device.

[0008] In a second aspect, a workpiece data processing device is provided, comprising: a workpiece data acquisition module, configured to acquire processing condition data, quality attribute values ​​and quality inspection result data of each of a plurality of workpieces processed by a production equipment, wherein the processing condition data of a workpiece include: processing parameters used by the production equipment when processing the workpiece and ambient environment condition data of the production equipment when processing the workpiece; a first relationship determination module, configured to determine a first relationship between the quality attribute value of the workpiece processed by the production equipment and the ambient environment condition data and the processing parameters of the production equipment when processing the workpiece based on the processing condition data and quality attribute values ​​of each of the plurality of workpieces; and a second relationship determination module, configured to determine a second relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production equipment based on the quality attribute value and quality inspection result data of each of the plurality of workpieces.

[0009] Different from the previous method of directly determining the relationship between the processing parameters of the production equipment and the pass rate of the workpiece quality inspection, in the embodiment of the present invention, by processing the workpiece data, a first relationship between the quality attribute value of the workpiece processed by the production equipment and the processing parameters of the production equipment and the surrounding environmental conditions when processing the workpiece, and a second relationship between the workpiece quality attribute value and the workpiece quality inspection result data are determined. By determining the first relationship and the second relationship, the relationship between the quality inspection result and the processing parameters of the production equipment can be described more accurately.

[0010] Optionally, when obtaining the processing condition data, quality attribute values, and quality inspection result data of each of the plurality of workpieces processed by a production device, the processing condition data, quality attribute values, and quality inspection result data of each of the plurality of workpieces processed by the production device may be obtained so that all the acquired data cover various combinations of ambient environment condition data and processing parameters as much as possible. In this way, for each combination of the processing parameters of the production device and the ambient environment conditions, only a small amount of data may be used to train the mathematical model of the first relationship.

[0011] Optionally, a conditional generative adversarial network (CGAN) model can be used to obtain the first relationship between the distribution of the processing parameters of the production equipment and the quality attributes of the workpiece under different ambient conditions. Specifically, the processing condition data of each of the plurality of workpieces can be used as an input vector of a generator in a CGAN model; the processing condition data of each of the plurality of workpieces can be used as an input vector of a discriminator in the CGAN model; the quality attribute value of each of the plurality of workpieces or the output vector of the generator of the CGAN model can be used as another input vector of the discriminator in the CGAN model; the CGAN model is trained, and the generator of the CGAN model obtained by training is used as the first relationship. For each combination of the processing parameters of the production equipment and the ambient conditions, only a small amount of data can be used to train the CGAN model. For any combination of processing parameters and ambient conditions, the trained CGAN model can generate any number of simulated workpiece samples. The method provided by the embodiment of the present invention can significantly reduce the data required to determine the relationship between the processing parameters and the workpiece quality inspection results.

[0012] Optionally, the quality attribute value of each of the plurality of first simulated workpieces can be generated according to the processing condition data, the quality attribute value of each of the plurality of workpieces processed by the production equipment, and the first relationship; the plurality of first simulated workpieces are compared with the obtained distribution of the quality attribute values ​​of the plurality of workpieces processed by the production equipment to determine the accuracy of the first relationship. By measuring the similarity between the real workpiece sample and the simulated workpiece sample, it can be ensured that the simulated sample can fully represent the real workpiece.

[0013] In a third aspect, a workpiece data processing device is provided, comprising: at least one memory for storing computer-readable codes; and at least one processor for executing the computer-readable codes stored in the at least one memory to execute the method described in the first aspect.

[0014] In a fourth aspect, a processing parameter setting method is provided, comprising: generating a set of processing parameters of the production equipment under given ambient environment condition data; for each element in the set, determining the quality attribute value of each of the plurality of second simulated workpieces processed according to the processing parameters represented by the element according to the given ambient environment condition data, the element and the first relationship, wherein the first relationship is the relationship between the quality attribute value of the workpiece processed by the production equipment and the ambient environment condition data of the production equipment when processing the workpiece and the processing parameters of the production equipment; for each element in the set, determining the quality inspection result data of the second simulated workpiece according to the second relationship and the determined quality attribute value of the second simulated workpiece for each of the plurality of second simulated workpieces, wherein the second relationship is the relationship between the quality inspection result data of the workpiece processed by the production equipment and the quality attribute value; for each element in the set, calculating the quality inspection pass rate according to the determined quality inspection result data of each of the plurality of second simulated workpieces; repeating the following process until a preset condition is met, wherein the preset condition comprises: iterative times When the number reaches the maximum number of iterations or the calculated quality inspection pass rate of the workpieces processed by the production equipment reaches the pass rate threshold, a new element is added to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; for the new element added to the set, according to the given ambient environment condition data, the new element and the first relationship, the quality attribute value of each of the plurality of third simulated workpieces processed according to the processing parameters represented by the new element is determined; for the new element added to the set, for each of the plurality of third simulated workpieces, according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces, the quality inspection result data of each of the plurality of third simulated workpieces is determined; for the new element added to the set, the quality inspection pass rate is calculated based on the determined quality inspection result data of each of the plurality of third simulated workpieces; the element corresponding to the maximum value of all quality inspection pass rates obtained by statistics is determined as the optimal value of the processing parameters of the production equipment under the given ambient environment condition data.

[0015] In a fifth aspect, a processing parameter setting device is provided, comprising: a processing parameter set establishment module, configured to generate a set of processing parameters of the production equipment under given ambient environment condition data; an adjustment module, configured to: for each element in the set, determine the quality attribute value of each of a plurality of second simulated workpieces processed according to the processing parameters represented by the element according to the given ambient environment condition data, the element and a first relationship, wherein the first relationship is the relationship between the quality attribute value of the workpiece processed by the production equipment and the ambient environment condition data of the production equipment when processing the workpiece and the processing parameters of the production equipment; for each element in the set, for each of the plurality of second simulated workpieces, determine the quality inspection result data of the second simulated workpiece according to the second relationship and the determined quality attribute value of the second simulated workpiece, wherein the second relationship is the relationship between the quality inspection result data of the workpiece processed by the production equipment and the quality attribute value; for each element in the set, calculate the quality inspection pass rate according to the determined quality inspection result data of each of the plurality of second simulated workpieces; repeat the following process until a preset condition is met, wherein the preset condition The method comprises: when the number of iterations reaches the maximum number of iterations or the calculated quality inspection pass rate of the workpieces processed by the production equipment reaches the pass rate threshold, a new element is added to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; for the new element added to the set, according to the given ambient environment condition data, the new element and the first relationship, the quality attribute value of each of the plurality of third simulated workpieces produced according to the processing parameters represented by the new element is determined; for the new element added to the set, for each of the plurality of third simulated workpieces, according to the second relationship and the determined quality attribute value of the plurality of third simulated workpieces, the quality inspection result data of each of the plurality of third simulated workpieces is determined; for the new element added to the set, the quality inspection pass rate is calculated according to the determined quality inspection result data of each of the plurality of third simulated workpieces; an optimal processing parameter determination module is used to determine the element corresponding to the maximum value of all quality inspection pass rates obtained by statistics as the optimal value of the processing parameter of the production equipment under the given ambient environment condition data.

[0016] By adopting the embodiment of the present invention, the optimal processing parameters under any ambient conditions can be automatically determined so that the quality inspection pass rate of the workpiece is the highest. Different from the previous direct determination of the relationship between the processing parameters of the production equipment and the quality inspection pass rate of the workpiece, in the embodiment of the present invention, by processing the workpiece data, the first relationship between the quality attribute value of the workpiece processed by the production equipment and the processing parameters of the production equipment and the ambient conditions when processing the workpiece, and the second relationship between the workpiece quality attribute value and the workpiece quality inspection result data are determined. A large number of workpiece quality attribute values ​​can be simulated according to the determined first relationship, and then a large number of workpiece quality inspection result data can be simulated according to the determined second relationship, and the quality inspection pass rate can be statistically calculated based on this. Since the simulated data meets the statistical significance, and the simulation of the data is based on the determined first relationship and the second relationship, the statistically obtained quality inspection pass rate can be considered to be relatively accurate. Then, based on the relatively accurate quality inspection pass rate, the optimal processing parameters can be obtained, and the optimal setting of the processing parameters of the production equipment is realized. On the one hand, by determining the first relationship and the second relationship, the relationship between the quality inspection result and the processing parameters of the production equipment can be described more accurately. On the other hand, based on the first relationship and the second relationship, a large amount of data is simulated, and a more accurate quality inspection pass rate is obtained by statistics, and the optimal processing parameter setting is obtained accordingly.

[0017] Optionally, when a new element is added to the set, a Gaussian process may be fitted according to each element in the set and the quality inspection pass rate corresponding to the element obtained from previous statistics, and then the new element may be calculated using the Gaussian process.

[0018] According to a sixth aspect, a computer-readable medium is provided, wherein the computer-readable medium stores a computer-readable code, and when the computer-readable code is executed by at least one processor, the method provided in the first aspect or the fourth aspect is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of an industrial system provided by an embodiment of the present invention.

[0020] Figure 2 A flowchart of a workpiece data processing method provided by an embodiment of the present invention.

[0021] Figure 3 A schematic diagram of a process for determining two relationships in a workpiece data processing method provided in an embodiment of the present invention.

[0022] Figure 4 A schematic diagram of using a CGAN model to determine a first relationship in a workpiece data processing method provided in an embodiment of the present invention.

[0023] Figure 5The present invention provides a flow chart of a verification method.

[0024] Figure 6 A flowchart of a method for setting processing parameters provided by an embodiment of the present invention.

[0025] Figure 7 This is a schematic diagram of comparing simulated workpiece data generated by using the CGAN model with real workpiece data in an embodiment of the present invention.

[0026] Figure 8 A schematic diagram of the structure of a workpiece data processing device provided in an embodiment of the present invention.

[0027] Fig. 9 A schematic diagram of another structure of a workpiece data processing device provided in an embodiment of the present invention.

[0028] List of reference numerals:

[0029] 100: Industrial system 10: Production equipment 20: Workpiece processed by production equipment 10

[0030] 30: Workpiece data processing device

[0031] 301: Workpiece data processing device 302: Verification device 303: Processing parameter setting device

[0032] 200: Workpiece Data Processing Method

[0033] S201: Acquire workpiece data S202: Determine a first relationship S203: Determine a second relationship

[0034] 40: Processing parameters 50: Quality attribute values ​​60: Quality inspection results

[0035] 70: Generative model 80: Classifier 90: Surrounding environment condition data

[0036] 701: Discriminator 702: Generator

[0037] 701c1, 701c2: Input vector of discriminator 701

[0038] 701b: Hidden layer of the discriminator 701a: Output of the discriminator 701

[0039] 702c1: Noise input vector of generator 702 702c2: Generator Input vector 702b of 702: hidden layer of generator 702

[0040] 702a: output vector of generator 702

[0041] 500: Verification method S501: Generate workpiece data S502: Data comparison

[0042] 600: Processing parameter setting method S601: Generate a set of processing parameters S602: Generate quality attribute values

[0043] S603: Determine the quality inspection result data S604: Count the quality inspection pass rate S605: Add new elements to the set

[0044] S606: Generate mass for new elements S607: Determine mass for new elements S608: Count mass for new elements

[0045] Attribute value inspection result data volume inspection pass rate

[0046] S609: Determine whether the preset conditions are met S610: Determine the optimal value of the processing parameter

[0047] 301a: Workpiece data acquisition module 301b: First relationship determination module 301c: Second relationship determination module

[0048] 302a: Simulation workpiece data generation module 302b: Data comparison module

[0049] 303a: Processing parameter set establishment module 303b: Adjustment module 303c: Optimal processing parameter determination module

[0050] 304: Memory 305: Processor 306: I / O (input / output) interface

[0051] 307: Bus DETAILED DESCRIPTION

[0052] In the embodiment of the present invention, a data-driven method is used to automatically set the processing parameters of the production equipment, and the workpiece quality is optimized by analyzing the workpiece data. By using the embodiment of the present invention, the optimal processing parameters under any ambient conditions can be automatically determined to maximize the quality inspection pass rate of the workpiece.

[0053] Different from the previous direct determination of the relationship between the processing parameters of the production equipment and the pass rate of the workpiece quality inspection, in the embodiment of the present invention, by processing the workpiece data, the first relationship between the quality attribute value of the workpiece processed by the production equipment and the processing parameters of the production equipment and the surrounding environmental conditions when processing the workpiece, and the second relationship between the workpiece quality attribute value and the workpiece quality inspection result data are determined. A large number of workpiece quality attribute values ​​can be simulated according to the determined first relationship, and then a large number of workpiece quality inspection result data can be simulated according to the determined second relationship, and the quality inspection pass rate can be statistically calculated accordingly. Since the simulated data meets the statistical significance, and the simulation of the data is based on the determined first relationship and the second relationship, the statistically obtained quality inspection pass rate can be considered to be relatively accurate. Then, based on the relatively accurate quality inspection pass rate, the optimal processing parameters can be obtained, and the optimal setting of the processing parameters of the production equipment is realized. On the one hand, by determining the first relationship and the second relationship, the relationship between the quality inspection result and the processing parameters of the production equipment can be described more accurately. On the other hand, according to the first relationship and the second relationship, a large amount of data is simulated, and a more accurate quality inspection pass rate is statistically obtained, and the setting of the optimal processing parameters is obtained accordingly.

[0054] In some embodiments of the present invention, a CGAN model can be used to obtain the relationship between the distribution of production equipment processing parameters and workpiece quality attributes under different ambient conditions. For each combination of production equipment processing parameters and ambient conditions, only a small amount of data can be used to train the CGAN model. For any combination of processing parameters and ambient conditions, the trained CGAN model can generate any number of simulated workpiece samples. Using the method provided by the embodiment of the present invention, the data required to determine the relationship between processing parameters and workpiece quality inspection results can be significantly reduced.

[0055] In addition, in some embodiments of the present invention, a KS test (Kolmogorov-Smirnov test) may be used to measure the similarity between the real workpiece sample and the simulated workpiece sample to ensure that the simulated sample can fully represent the real workpiece.

[0056] The simulated workpiece samples can be used for virtual quality inspection at almost no cost. In particular, a classifier is trained that can predict the quality inspection results under the condition of given workpiece quality attributes, and the virtual quality inspection is realized by training the classifier.

[0057] In some embodiments, a Bayesian optimization algorithm may also be used to automatically determine optimal processing parameters that ensure optimal quality inspection results based on simulated workpiece samples generated by the CGAN model and virtual quality inspection results obtained for these simulated workpiece samples.

[0058] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail with reference to the accompanying drawings. The embodiments described below are only part of the embodiments of the present invention, but not all of them.

[0059] Figure 1 Schematic diagram of an industrial system provided by an embodiment of the present invention. Figure 1 As shown, the industrial system 100 includes a production device 10 and a plurality of workpieces 20 processed by the production device 10. In addition, the industrial system 100 may further include a workpiece data processing device 30 configured to process data of the workpieces 20.

[0060] Among them, the workpiece data processing device 30 may include: a workpiece data processing device 301, which is configured to obtain data of the workpiece 20, and determine the relationship between the quality attribute value of the workpiece processed by the production equipment 10 and the ambient environment condition data 90 of the production equipment 10 when processing the workpiece and the quality data value of the workpiece processed by the production equipment 10 through data processing, which is called the "first relationship"; and determine the relationship between the quality inspection result data 60 of the workpiece processed by the production equipment 10 and the quality attribute value 50, which is called the "second relationship".

[0061] The workpiece data processing device 30 may further include a verification device 302 configured to generate a plurality of simulated workpiece data according to the first relationship, and compare the generated simulated workpiece data with the data of the workpiece processed by the production equipment 10 to determine the accuracy of the first relationship.

[0062] In addition, the workpiece data processing device 30 may also include a processing parameter setting device 303, which is configured to determine the optimal value of the processing parameter 40 of the production equipment 10 under the given ambient environment condition data 90 based on the above-mentioned first relationship and second relationship.

[0063] Figure 2 Flow chart of a workpiece data processing method provided by an embodiment of the present invention. The method 200 can be executed by a workpiece data processing device 30, such as Figure 2 As shown, the following steps may be included:

[0064] S201: Acquire workpiece data.

[0065] In step S201, the following data of each of the plurality of workpieces 20 processed by a production device 10 is obtained:

[0066] 1) Processing condition data

[0067] Among them, the processing condition data includes:

[0068] processing parameters 40 used by the production equipment 10 when processing the workpiece 20, and

[0069] Environmental condition data 90 of the production equipment 10 when processing the workpiece 20;

[0070] 2) Quality attribute value 50

[0071] 3) Quality inspection result data 60, such as: passed quality inspection or failed quality inspection.

[0072] Here, the plurality of workpieces 20 may be part or all of the workpieces processed by the production equipment 10. In step S201, the acquired data covers the combination of various ambient environment condition data 90 and processing parameters 40 as much as possible. For each combination of ambient environment condition data 90 and processing parameters 40, the amount of acquired data may not be limited. That is, in the embodiment of the present invention, it is not necessary to acquire a large amount of workpiece data to meet statistical significance, as long as the combination of various ambient environment condition data 90 and processing parameters 40 is covered as much as possible.

[0073] After step S201, in the embodiment of the present invention, the relationship between the quality inspection result data 60 of the workpiece processed by the production equipment 10 and the ambient environment condition data 90 of the production equipment 10 when processing the workpiece and the processing parameters 40 of the production equipment 10 is not directly determined. This is because, if the relationship is directly determined, a relatively large amount of workpiece data is required to meet statistical significance. Assuming that the quality inspection result data 60 includes: passed quality inspection and failed quality inspection, then under the same combination of processing parameters 40 and quality inspection result data 60, due to the influence of some random factors, one workpiece 20 processed by the production equipment 10 may pass the quality inspection, while another workpiece 20 may fail the quality inspection. If the amount of data is small, that is, when the workpiece sample is small, the statistically obtained relationship is not accurate. Figure 3 As shown, in the embodiment of the present invention, when there is less workpiece data, a first relationship between the quality attribute value 50 of the workpiece processed by the production equipment 10 and the ambient environment condition data 90 when the production equipment 10 processes the workpiece and the processing parameters 40 of the production equipment 10 is determined by a method such as a generation model 70, and then a second relationship between the quality inspection result data 60 of the workpiece processed by the production equipment 10 and the quality attribute value 50 is determined by a method such as a classifier 80. Since the quality inspection result data 60 is determined by the quality attribute value 50, the result is less affected by random factors, and therefore, the second relationship between the determined quality attribute value 50 and the quality inspection result data 60 is relatively accurate.

[0074] S202: Determine a first relationship.

[0075] In step S202, based on the processing condition data and quality attribute value 50 of each workpiece 20 in the plurality of workpieces 20, a first relationship between the quality attribute value 50 of the workpiece processed by the production equipment 10 and the ambient environment condition data 90 and the processing parameters 40 of the production equipment 10 when processing the workpiece can be determined.

[0076] Optionally, the first relationship is determined by training the CGAN model 70. Specifically, Figure 4 As shown, the CGAN model 70 includes two sub-models: a generator 702 (Generator, referred to as "G") and a discriminator 701 (Discriminator, referred to as "D").

[0077] The generator 702 of the CGAN model 70 may include two input vectors (z, y) and one output vector ((G(z|y))).

[0078] Among them, the two input vectors include:

[0079] 1) A noise vector 702c1(z), including randomly generated floating point numbers with values ​​between 0 and 1.

[0080] 2) The condition vector 702c2(y) is the processing condition data (a combination of the processing parameters 40 and the surrounding environment condition data 90) of each of the plurality of workpieces 20.

[0081] The output vector (G(z|y)) of the generator 702 is the mass property value 50 of the simulated workpiece.

[0082] The goal of the generator 702 is to generate a simulated quality attribute value 50 so that the discriminator 701 cannot distinguish the simulated quality attribute value 50 from the real quality attribute value 50 (i.e., the quality attribute value 50 of the workpiece 20 processed by the production equipment 10). This goal can be mathematically expressed as:

[0083]

[0084] Here, p(z) is a mutually independent multivariate uniform distribution with values ​​ranging from 0 to 1.

[0085] The discriminator 701 of the CGAN model 70 also has two input vectors and one output, which is a scalar.

[0086] Among them, the two input vectors include:

[0087] 1) Conditional vector 701c2(y), which is the same as a conditional vector 702c2 of generator 702.

[0088] 2) The quality attribute value 50 (G(z|y)) of the simulated workpiece or the quality attribute value 50 of the real workpiece (ie, the quality attribute value 50 of the workpiece processed by the production equipment 10 obtained in step S201).

[0089] The output of the discriminator 701 of the CGAN model 70 is a scalar (D(x'|y)). The larger the value, the greater the possibility that the discriminator 701 believes that x' is the quality attribute value 50 of the simulated workpiece. The goal of the discriminator 701 is to distinguish between the real quality attribute value 50 and the simulated quality attribute value 50, which can be mathematically expressed as:

[0090]

[0091] Wherein, p(x)is the distribution of the real quality attribute value 50. In the above formula, the first part minimizes the output for the real quality attribute value 50; the second part maximizes the output for the simulated quality attribute value 50 output by the generator 702; the third part is the gradient penalty loss (see Gulrajani, Ishaan, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, and Aaron C. Courville, 2017, "Improved training of wasserstein gans" published on pages 5767 to 5777 in "Advances in Neural Information Processing Systems", where ω is a hyperparameter, usually taking a value of 10. p(x') is uniformly taken along the straight line between the real quality attribute value 50 and the simulated quality attribute value 50 to avoid gradient disappearance or surge during training.

[0092] The CGAN model 70 can be trained by alternately minimizing the objective functions of the generator 702 and the discriminator 701 for multiple iterations until the discriminator 701 cannot distinguish between the real quality attribute value 50 and the simulated quality attribute value 50. Then, the generator 702 can be used to generate a large number of simulated quality attribute values ​​50 under arbitrary processing parameters 40 and arbitrary surrounding environment condition data 90. Accordingly, the distribution of the simulated quality attribute values ​​50 is close to the distribution of the real quality attribute values ​​50. Figure 7The figure shows the distribution of the quality data values ​​50 of the actual workpiece collected from the worm gear manufacturing plant and the simulated quality attribute values ​​50 generated by the CGAN model under the same CNC machining parameters and ambient conditions. In the figure, the high-dimensional quality attribute values ​​are mapped to two dimensions for easy display through the dimensionality reduction method. Among them, the solid circles represent the real quality attribute values ​​50, while the hollow circles represent the simulated quality attribute values ​​50.

[0093] S203: Determine a second relationship.

[0094] In step S203 , a second relationship between the quality inspection result data 60 and the quality attribute value 50 of the workpiece processed by the production equipment 10 may be determined according to the quality attribute value and the quality inspection result data of each of the plurality of workpieces 20 .

[0095] If the quality inspection result data 60 is passed or failed, the classifier 80 can be used to determine the second relationship. In fact, as long as the best classification accuracy can be achieved, any classifier can be selected, such as random forest, support vector machine, and even a predefined acceptable range can be used to check the quality attribute value 50 in this step. Specifically, the input of the classifier 80 is the quality attribute value 50 of the workpiece 20 obtained in step S201, and the output is the quality inspection result data 60 of the workpiece 20. The classifier 80 is trained based on the real data obtained in step S201. After the training is completed, the classifier 80 can be used to classify whether the simulated quality attribute value 50 passes the quality inspection, which can be regarded as a virtual quality test.

[0096] Figure 5 Flow chart of the verification method provided by the embodiment of the present invention. The verification method 500 can be Figure 1 The verification device 302 in FIG. Figure 5 As shown, the following steps may be included:

[0097] S501: Generate workpiece data.

[0098] In step S501 , a plurality of quality attribute values ​​50 of first simulated workpieces 20 may be generated according to the processing condition data 40 , the quality attribute value 50 , and the first relationship of each of the plurality of workpieces 20 processed by the production equipment 10 .

[0099] Optionally, if the first relationship is obtained by training the CGAN model 70, the generator 702 in the CGAN model 70 may be used to generate a large number of simulated quality attribute values ​​50.

[0100] S502: Compare the first simulated workpiece 20 generated in step S501 with the acquired distribution of quality attribute values ​​of a plurality of workpieces 20 processed by the production equipment 10 to determine the accuracy of the first relationship.

[0101] Optionally, the Kolmogorov-Smirnov test can be used to quantitatively measure the similarity between the true quality attribute value 50 and the simulated quality attribute value 50 to ensure that the simulated quality attribute value 50 can fully represent the true quality attribute value 50. In statistics, the Kolmogorov-Smirnov test (ks test or ks test) is a nonparametric test that can be used to compare whether the continuous one-dimensional probability distributions of two samples are equal. Here, the multivariate generalization of the test is used to check whether the difference between the simulated quality attribute value 50 and the true quality attribute value 50 is statistically significant (see Jerome H. Friedman and Lawrence C. Rafsky, "Multivariate generalizations of the Wald-Wolfowitz and Smirnov two-sample tests" published in "The annals of Statistics" in 1979, pages 679 to 717). After determining that the simulated quality attribute value 50 is very close to the actual quality attribute value 50 , the simulated quality attribute value 50 may be further used to perform quality analysis.

[0102] After obtaining the above-mentioned first relationship and second relationship, the quality inspection pass rate of the workpiece can be evaluated under any ambient conditions and any set processing parameters. Among them, a brute force method for seeking the optimal setting of the processing parameters of the production equipment is to enumerate all possible parameter settings and select the parameters with the highest quality inspection pass rate. However, this method is too expensive and sometimes unacceptable, especially when the ambient conditions change rapidly and the optimal processing parameters need to be adjusted dynamically. Therefore, an embodiment of the present invention also provides a processing parameter setting method, such as Figure 6 shown.

[0103] Figure 6 Flow chart of the processing parameter setting method provided by the embodiment of the present invention. The processing parameter setting method 600 can be Figure 1 The processing parameter setting device 303 is used to execute. Figure 6 As shown, the method may include the following steps:

[0104] S601: Generate a set of processing parameters 40.

[0105] In step S601, under given ambient environment condition data, a set of processing parameters 40 of a production device 10 is generated, wherein the set of processing parameters 40 may be randomly generated. Taking CNC as an example, M processing parameters 40 are randomly generated, wherein M is a positive integer.

[0106] S602: Generate a quality attribute value of 50.

[0107] In step S602, for each element in the set generated in step S601, a plurality of quality attribute values ​​50 of the second simulated workpiece 20 are generated according to the given ambient environment condition data 90, the element and the first relationship. Still taking CNC as an example, given the ambient environment condition data 90, for each of the M processing parameters 40 randomly generated in the aforementioned step S601, a large number of quality attribute values ​​50 are generated respectively using the CGAN model 70.

[0108] S603: Determine the quality inspection result data 60.

[0109] In step S603, for each element in the set, the quality inspection result data 60 of each second simulated workpiece 20 is determined according to the second relationship and the quality attribute values ​​50 of the plurality of second simulated workpieces 20 generated. Still taking CNC as an example, for each of the M processing parameters 40 randomly generated in the aforementioned step S601, the aforementioned second relationship is used to generate the quality inspection result data 60 corresponding to each quality attribute value 50 according to the large number of quality attribute values ​​50 corresponding to the processing parameter 40 generated in step S602.

[0110] S604: Calculate the quality inspection pass rate.

[0111] In step S604, for each element in the set, the quality inspection pass rate corresponding to the element is counted according to the quality inspection result data 60 of the plurality of second simulated workpieces 20 corresponding to the element. Still taking CNC as an example, for each of the M processing parameters 40 randomly generated in step S601, the quality inspection pass rate of the workpiece processed by CNC using the processing parameter 40 is counted according to the large amount of quality inspection result data 60 corresponding to the processing parameter 40.

[0112] After step S604, steps S605 to S608 are repeated until a preset condition is met. The preset condition can be set according to actual conditions. For example: the number of iterations reaches the maximum number of iterations; another example: the calculated quality inspection pass rate of the workpiece processed by the production equipment 10 reaches the pass rate threshold; another example: the number of iterations reaches the maximum number of iterations or the calculated quality inspection pass rate of the workpiece processed by the production equipment 10 reaches the pass rate threshold. If one of these two conditions is met, the loop is exited.

[0113] S605: Add a new element to the collection.

[0114] In step S605, a new element is added to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past. Among them, a Gaussian process (GP) can be fitted according to each element in the set and the quality inspection pass rate corresponding to the element obtained by statistics in the past (see Rasmussen, Carl Edward, "Gaussian Process in Machine Learning" published on pages 63 to 71 in "Gaussian Process in Machine Learning" in 2004), and the new element is calculated using the Gaussian process. Still taking CNC as an example, a Gaussian process can be fitted, with all elements in the set, that is, each processing parameter 40 as the input of the Gaussian process, and the quality inspection pass rate corresponding to each processing parameter 40 as the output of the Gaussian process.

[0115] Specifically, Bayesian optimization (see Frazier, Peter I., A Tutorial on Bayesian Optimization, published on arXiv preprint arXiv: 1807.02811 in 2018) can be used to effectively search the candidate space of the processing parameters 40. Then, under each processing parameter 40 setting in a given set, a Gaussian process is used to estimate the posterior distribution of the workpiece quality inspection pass rate of all unsearched settings:

[0116] P(y|x,θ,D)

[0117] Where x is the unsearched setting, θ is the parameter of the Gaussian process, and D is the estimated value of the pass rate of the quality inspection of each search setting and its corresponding workpiece. Then, we can maximize the quality inspection by choosing The space of processing parameters 40 is effectively searched by setting the expected value of the rate. The calculation formula is as follows:

[0118]

[0119] in, Indicates the highest quality inspection pass rate in the searched settings. For a detailed explanation of the above formula, please refer to the aforementioned "Bayesian Optimization Tutorial".

[0120] By using the above-mentioned Bayesian optimization method, the optimal configuration of the processing parameters 40 is achieved under the specific ambient environment condition data 90 .

[0121] S606: Generate a quality attribute value for the new element.

[0122] In step S606 , the quality attribute 50 of each of the plurality of third simulated workpieces 20 produced according to the processing parameters represented by the new element is determined according to the given ambient environment condition data, the new element and the first relationship.

[0123] S607: Determine quality inspection result data for the new element.

[0124] In step S607 , for the new element, for each of the plurality of third simulated workpieces 20 , quality inspection result data 60 for each of the plurality of third simulated workpieces 20 is determined according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces 20 .

[0125] S608: Count the quality inspection pass rate for the new element.

[0126] In step S608, for the new element added to the set, the quality inspection pass rate is calculated according to the quality inspection result data of each of the plurality of third simulated workpieces 20. S609: Determine whether the above preset conditions are met. If the above preset conditions are met, then jump out of the loop and execute step S610, otherwise return to step S605.

[0127] S610: Determine optimal values ​​of processing parameters.

[0128] The element corresponding to the maximum value of all quality inspection pass rates obtained by statistics is determined as the optimal value of the processing parameter of the production equipment 10 under the given ambient environment condition data.

[0129] exist Figure 6 In the process shown, the number of iterations is N, which is initially set to 0. The values ​​of N and the number of elements M in the set of processing parameters 40 generated in step S601 may be determined by the number of processing parameters 40 to be considered. Taking CNC as an example, the number of CNC processing parameters is usually less than 10, so here it can be considered to set M to 5 and N to 10. The values ​​of M and N can be adjusted according to experiments.

[0130] Figure 8The workpiece data processing device 30 may include: a workpiece data processing device 301 , and optionally, a verification device 302 and / or a processing parameter setting device 303 .

[0131] Among them, the workpiece data processing device 301 may include: a workpiece data acquisition module 301a, which is configured to acquire processing condition data, quality attribute values ​​and quality inspection result data of each workpiece 20 in a plurality of workpieces 20 processed by a production device 10, wherein the processing condition data of a workpiece 20 includes: processing parameters used by the production device 10 when processing the workpiece 20 and surrounding environment condition data when the production device 10 processes the workpiece 20; a first relationship determination module 301b, which is configured to determine the first relationship between the quality attribute value of the workpiece processed by the production device 10 and the surrounding environment condition data and the processing parameters of the production device 10 when processing the workpiece based on the processing condition data and quality attribute values ​​of each workpiece 20 in the plurality of workpieces 20; a second relationship determination module 301c, which is configured to determine the second relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production device 10 based on the quality attribute value and quality inspection result data of each workpiece 20 in the plurality of workpieces 20.

[0132] Optionally, the workpiece data acquisition module 301a is specifically configured to obtain processing condition data, quality attribute values ​​and quality inspection result data of each workpiece 20 among a plurality of workpieces 20 processed by a production device 10, so that all acquired data covers various combinations of ambient environment condition data and processing parameters as much as possible.

[0133] Optionally, the first relationship determination module 301b is specifically configured to: use the processing condition data of each workpiece 20 in the plurality of workpieces 20 as an input vector of the generator 702 in a CGAN model; use the processing condition data of each workpiece 20 in the plurality of workpieces 20 as an input vector of the discriminator 701 of the CGAN model; use the quality attribute value of each workpiece 20 in the plurality of workpieces 20 or the output vector 702a of the generator of the CGAN model as another input vector of the discriminator 701 in the CGAN model; train the CGAN model, and use the generator 701 of the trained CGAN model as the first relationship.

[0134] Among them, the verification device 302 may include: a simulated workpiece data generation module 302a, configured to generate a plurality of quality attribute values ​​of the first simulated workpieces 20 based on the processing condition data, quality attribute values, and the first relationship of each of the plurality of workpieces 20 processed by the production equipment 10; a data comparison module 302b, configured to compare the plurality of first simulated workpieces 20 with the distribution of the quality attribute values ​​of the plurality of workpieces 20 processed by the production equipment 10 to determine the accuracy of the first relationship.

[0135] The processing parameter setting device 303 may include:

[0136] A processing parameter set establishing module 303a is configured to generate a set of processing parameters of the production equipment 10 under given ambient environment condition data;

[0137] An adjustment module 303b is configured to determine, for each element in the set, the quality attribute value of each of the plurality of second simulated workpieces 20 processed according to the processing parameters represented by the element according to given ambient environment condition data, the element and a first relationship, wherein the first relationship is the relationship between the quality attribute value of the workpiece processed by the production equipment 10 and the ambient environment condition data and the processing parameters of the production equipment 10 when the workpiece is processed by the production equipment 10; for each element in the set, for each of the plurality of second simulated workpieces 20, determine the quality inspection result data of the second simulated workpiece 20 according to the second relationship and the determined quality attribute value of the second simulated workpiece 20, wherein the second relationship is the relationship between the quality inspection result data of the workpiece processed by the production equipment 10 and the quality attribute value; for each element in the set, calculate the quality inspection pass rate according to the determined quality inspection result data of each of the plurality of second simulated workpieces 20;

[0138] The adjustment module 303b is further configured to repeat the following process until a preset condition is satisfied, wherein the preset condition includes: the number of iterations reaches the maximum number of iterations or the calculated quality inspection pass rate of the workpieces processed by the production equipment 10 reaches the pass rate threshold: a new element is added to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; for the new element added to the set, according to the given ambient environment condition data, the new element and the first relationship, the quality attribute value of each of the plurality of third simulated workpieces 20 produced according to the processing parameters represented by the new element is determined; for the new element added to the set, for each of the plurality of third simulated workpieces 20, according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces 20, the quality inspection result data of each of the plurality of third simulated workpieces 20 is determined; for the new element added to the set, the quality inspection pass rate is calculated according to the determined quality inspection result data of each of the plurality of third simulated workpieces 20;

[0139] An optimal processing parameter determination module 303c is configured to determine the element corresponding to the maximum value of all quality inspection pass rates obtained by statistics as the optimal value of the processing parameter of the production equipment 10 under given ambient environment condition data.

[0140] Optionally, when the adjustment module 303b adds a new element to the set so that the expected value of the quality inspection pass rate obtained based on the statistics of the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by previous statistics, it is specifically configured as follows: fitting a Gaussian process according to each element in the set and the quality inspection pass rate corresponding to the element obtained by previous statistics; and calculating the new element using the Gaussian process.

[0141] Fig. 9 FIG. 3 is another structural diagram of a workpiece data processing device 30 provided in an embodiment of the present invention. Fig. 9 As shown, under this structure, the workpiece data processing device 30 may include at least one memory 304 for storing computer-readable codes; at least one processor 305 is configured to execute the computer-readable codes stored in at least one memory 304, execute the aforementioned method 200, and optionally, also execute the aforementioned method 500 and / or method 600. Figure 8The modules shown in the figure can be regarded as program modules written in computer readable codes stored in the memory 304. When these program modules are called by the processor 305, the aforementioned methods 200, 500 and / or 600 can be executed. The workpiece data processing device 30 may also include an I / O interface 306, which can be connected to external devices such as a mouse and a display. Optionally, at least one memory 304, at least one processor 305 and the I / O interface 306 can communicate with each other through a bus 307.

[0142] In addition, an embodiment of the present invention further provides a computer-readable medium, which stores a computer-readable code. When the computer-readable code is executed by at least one processor, the aforementioned methods 200, 500 and / or 600 are implemented.

[0143] In summary, the embodiments of the present invention provide a workpiece data processing method, device and computer-readable medium for accurately determining the relationship between the processing parameters of production equipment, ambient environment condition data and the quality inspection results of the workpiece, and then determining the optimal processing parameters based on the relationship.

[0144] The workpiece data processing method provided by the embodiment of the present invention is purely data-driven, and compared with the method driven by human experience, it significantly improves the accuracy, efficiency and versatility of the process of setting optimal processing parameters for workpiece processing.

[0145] The solution provided by the embodiment of the present invention is cost-effective and reduces the amount of data required to construct a mathematical model for predicting the pass rate of a workpiece produced under given processing parameters and ambient conditions. The solution provided by the embodiment of the present invention only requires searching a small portion of the processing parameter candidate space to find the optimal processing parameters.

[0146] In the embodiment of the present invention, the CGAN model is used to simulate the quality attribute value of the workpiece under given processing parameters, and then the classifier is used to perform virtual quality inspection to determine the quality inspection pass rate of the workpiece. The solution provided by the embodiment of the present invention can be used to determine the relationship between the processing parameters of the production equipment and the quality inspection result data with only a small amount of workpiece data. Compared with the method that relies solely on human experience, the determined relationship is more accurate.

[0147] In the embodiment of the present invention, the Bayesian optimization method can be used to automatically and effectively find the optimal settings of the processing parameters of the production equipment, so as to obtain the highest workpiece quality inspection pass rate under any ambient environment conditions.

[0148] It should be noted that not all steps and modules in the above-mentioned processes and system structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by at least two physical entities, or some components in at least two independent devices can be implemented together.

[0149] In the above embodiments, the hardware unit can be realized by mechanical or electrical means. For example, a hardware unit can include permanent dedicated circuits or logic (such as special processors, field programmable gate arrays (Field-Programmable Grate Array, FPGA) or application specific integrated circuits (Application Specific Integrated Circuits, ASIC) etc.) to complete the corresponding operation. The hardware unit can also include programmable logic or circuits (such as general-purpose processors or other programmable processors), which can be temporarily set by software to complete the corresponding operation. Specific implementation (mechanical means or dedicated permanent circuits or temporarily set circuits) can be determined based on cost and time considerations.

[0150] The embodiments of the present invention are shown and described in detail above through the accompanying drawings and preferred embodiments. However, the embodiments of the present invention are not limited to these disclosed embodiments. Based on the above embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present invention, and these embodiments are also within the protection scope of the embodiments of the present invention.

Claims

1. A workpiece data processing method (200), characterized in that: include: Acquiring processing condition data, quality attribute values, and quality inspection result data of each of a plurality of workpieces (20) processed by a production device (10), wherein the processing condition data of one of the workpieces (20) includes: processing parameters used by the production device (10) when processing the workpiece (20) and ambient environment condition data of the production device (10) when processing the workpiece (20); Determining, based on processing condition data and quality attribute values ​​of each of the plurality of workpieces (20), a first relationship between the quality attribute values ​​of the workpieces processed by the production equipment (10), the ambient environment condition data of the production equipment (10) when processing the workpieces, and the processing parameters of the production equipment (10); Determining a second relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production equipment (10) according to the quality attribute value and the quality inspection result data of each of the plurality of workpieces (20); Among them, it also includes: Under given ambient environment condition data, generating a set of processing parameters of the production equipment (10); For each element in the set, determining, based on the given ambient environment condition data, the element and the first relationship, a quality attribute value of each of a plurality of second simulated workpieces processed according to the processing parameters represented by the element; For each element in the set, for each of the plurality of second simulated workpieces, determining quality inspection result data of the second simulated workpiece according to the second relationship and the determined quality attribute value of the second simulated workpiece; For each element in the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of second simulated workpieces determined; Repeat the following process until a preset condition is met, wherein the preset condition includes: the number of iterations reaches a maximum number of iterations or the calculated quality inspection pass rate of the workpiece processed by the production equipment (10) reaches a pass rate threshold, Adding a new element to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; For the new element added to the set, determining, based on the given ambient environment condition data, the new element and the first relationship, a quality attribute value of each of a plurality of third simulated workpieces processed according to the processing parameters represented by the new element; For each of the plurality of third simulated workpieces, for the new element added to the set, according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces, determine quality inspection result data of each of the plurality of third simulated workpieces; For the new element added to the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of third simulated workpieces; The element corresponding to the maximum value of all quality inspection pass rates obtained by statistics is determined as the optimal value of the processing parameters of the production equipment (10) under the given ambient environment condition data.

2. The method (200) according to claim 1, characterized in that Obtaining processing condition data, quality attribute values, and quality inspection result data of each of a plurality of workpieces (20) processed by a production device (10), including: The processing condition data, quality attribute value and quality inspection result data of each of the plurality of workpieces (20) processed by the production equipment (10) are obtained, so that all the obtained data cover the combination of various ambient environment condition data and processing parameters as much as possible.

3. The method (200) as described in 1 or 2, characterized in that: Determining the first relationship includes: Using the processing condition data of each of the plurality of workpieces (20) as an input vector of a generator (702) in a CGAN model; Using the processing condition data of each of the plurality of workpieces (20) as an input vector of the discriminator (701) of the CGAN model; Using the quality attribute value of each of the plurality of workpieces (20) or the output vector (702a) of the generator of the CGAN model as another input vector of the discriminator (701) in the CGAN model; The CGAN model is trained, and the generator (702) of the CGAN model obtained by training is used as the first relationship.

4. The method (200) according to any one of claims 1 to 3, characterized in that: Also includes: generating a quality attribute value for each of a plurality of first simulated workpieces according to processing condition data, a quality attribute value of each of a plurality of workpieces (20) processed by the production equipment (10), and the first relationship; The plurality of first simulated workpieces are compared with the obtained distribution of quality attribute values ​​of the plurality of workpieces (20) processed by the production equipment (10) to determine the accuracy of the first relationship.

5. The method (200) according to claim 1, characterized in that Adding a new element to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past, includes: Fitting a Gaussian process according to each element in the set and the quality inspection pass rate corresponding to the element obtained from previous statistics; The new element is obtained by calculation using the Gaussian process.

6. A workpiece data processing device (30), characterized in that: include: A workpiece data acquisition module (301a) is configured to acquire processing condition data, quality attribute values ​​and quality inspection result data of each of a plurality of workpieces (20) processed by a production device (10), wherein the processing condition data of one of the workpieces (20) includes: processing parameters used by the production device (10) when processing the workpiece (20) and ambient environment condition data of the production device (10) when processing the workpiece (20); a first relationship determination module (301b) configured to determine, based on the processing condition data and the quality attribute value of each of the plurality of workpieces (20), a first relationship between the quality attribute value of the workpiece processed by the production equipment (10) and the surrounding environment condition data of the production equipment (10) when processing the workpiece and the processing parameters of the production equipment (10); a second relationship determination module (301c) configured to determine a second relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production equipment (10) based on the quality attribute value and the quality inspection result data of each of the plurality of workpieces (20); Among them, it also includes: a processing parameter set establishing module (303a) configured to randomly generate a set of processing parameters of the production equipment (10) under given ambient environment condition data; An adjustment module (303b) is configured to: For each element in the set, determining, based on the given ambient environment condition data, the element and the first relationship, a quality attribute value of each of a plurality of second simulated workpieces processed according to the processing parameters represented by the element; For each element in the set, for each of the plurality of second simulated workpieces, determining quality inspection result data of the second simulated workpiece according to the second relationship and the determined quality attribute value of the second simulated workpiece; For each element in the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of second simulated workpieces determined; Repeat the following process until a preset condition is met, wherein the preset condition includes: the number of iterations reaches a maximum number of iterations or the calculated quality inspection pass rate of the workpiece processed by the production equipment (10) reaches a pass rate threshold, Adding a new element to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; For the new element added to the set, determining, based on the given ambient environment condition data, the new element and the first relationship, a quality attribute value of each of a plurality of third simulated workpieces processed according to the processing parameters represented by the new element; For each of the plurality of third simulated workpieces, for the new element added to the set, according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces, determine quality inspection result data of each of the plurality of third simulated workpieces; For the new element added to the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of third simulated workpieces; An optimal processing parameter determination module (303c) is configured to determine the element corresponding to the maximum value of all quality inspection pass rates obtained by statistics as the optimal value of the processing parameter of the production equipment (10) under the given ambient environment condition data.

7. The device (30) according to claim 6, characterized in that The workpiece data acquisition module (301a) is specifically configured to acquire processing condition data, quality attribute values ​​and quality inspection result data of each of the plurality of workpieces (20) processed by the production equipment (10), so that all acquired data covers as many combinations of various ambient environment condition data and processing parameters as possible.

8. The device (30) according to claim 6 or 7, characterized in that The first relationship determination module (301b) is specifically configured to: Using the processing condition data of each of the plurality of workpieces (20) as an input vector of a generator (702) in a CGAN model; Using the processing condition data of each of the plurality of workpieces (20) as an input vector of the discriminator (701) of the CGAN model; Using the quality attribute value of each of the plurality of workpieces (20) or the output vector (702a) of the generator of the CGAN model as another input vector of the discriminator (701) in the CGAN model; The CGAN model is trained, and the generator (702) of the CGAN model obtained by training is used as the first relationship.

9. The device (30) according to any one of claims 6 to 8, characterized in that Also includes: a simulated workpiece data generating module (302a) configured to generate a quality attribute value for each of a plurality of first simulated workpieces based on processing condition data, a quality attribute value of each of a plurality of workpieces (20) processed by the production equipment (10), and the first relationship; A data comparison module (302b) is configured to compare the first simulated workpieces with the obtained distribution of quality attribute values ​​of the workpieces (20) processed by the production equipment (10) to determine the accuracy of the first relationship.

10. The device (30) according to claim 6, characterized in that The adjustment module (303b) is specifically configured as follows when a new element is added to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past: Fitting a Gaussian process according to each element in the set and the quality inspection pass rate corresponding to the element obtained from previous statistics; The new element is obtained by calculation using the Gaussian process.

11. A workpiece data processing device (30), characterized in that: include: at least one memory (304) for storing computer readable code; At least one processor (305) is used to execute the computer-readable code stored in the at least one memory (304) to execute the method according to any one of claims 1 to 5.

12. A processing parameter setting method (600), characterized in that: include: Under given ambient environment condition data, a set of processing parameters of a production device (10) is generated; For each element in the set, determining the quality attribute value of each of the plurality of second simulated workpieces processed according to the processing parameters represented by the element according to the given ambient environment condition data, the element and a first relationship, wherein the first relationship is a relationship between the quality attribute value of the workpiece processed by the production equipment (10) and the ambient environment condition data when the production equipment (10) processes the workpiece and the processing parameters of the production equipment (10); For each element in the set, for each of the plurality of second simulated workpieces, quality inspection result data of the second simulated workpiece is determined according to a second relationship and the determined quality attribute value of the second simulated workpiece, wherein the second relationship is a relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production equipment (10); For each element in the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of second simulated workpieces determined; Repeat the following process until a preset condition is met, wherein the preset condition includes: the number of iterations reaches a maximum number of iterations or the calculated quality inspection pass rate of the workpiece processed by the production equipment (10) reaches a pass rate threshold, Adding a new element to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; For the new element added to the set, determining, based on the given ambient environment condition data, the new element and the first relationship, a quality attribute value of each of a plurality of third simulated workpieces processed according to the processing parameters represented by the new element; For each of the plurality of third simulated workpieces, for the new element added to the set, according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces, determine quality inspection result data of each of the plurality of third simulated workpieces; For the new element added to the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of third simulated workpieces; The element corresponding to the maximum value of all quality inspection pass rates obtained by statistics is determined as the optimal value of the processing parameters of the production equipment (10) under the given ambient environment condition data.

13. The method (600) of claim 12, characterized in that Adding a new element to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past, includes: Fitting a Gaussian process according to each element in the set and the quality inspection pass rate corresponding to the element obtained from previous statistics; The new element is obtained by calculation using the Gaussian process.

14. A processing parameter setting device (303), characterized in that: include: A processing parameter set establishing module (303a) is configured to generate a set of processing parameters of a production device (10) under given ambient environment condition data; An adjustment module (303b) is configured to: For each element in the set, determining the quality attribute value of each of the plurality of second simulated workpieces processed according to the processing parameters represented by the element according to the given ambient environment condition data, the element and a first relationship, wherein the first relationship is a relationship between the quality attribute value of the workpiece processed by the production equipment (10) and the ambient environment condition data when the production equipment (10) processes the workpiece and the processing parameters of the production equipment (10); For each element in the set, for each of the plurality of second simulated workpieces, determining quality inspection result data of the second simulated workpiece (20) according to a second relationship and the determined quality attribute value of the second simulated workpiece, wherein the second relationship is a relationship between the quality inspection result data and the quality attribute value of the workpiece processed by the production equipment (10); For each element in the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of second simulated workpieces determined; Repeat the following process until a preset condition is met, wherein the preset condition includes: the number of iterations reaches a maximum number of iterations or the calculated quality inspection pass rate of the workpiece processed by the production equipment (10) reaches a pass rate threshold, Adding a new element to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past; For the new element added to the set, determining, based on the given ambient environment condition data, the new element and the first relationship, a quality attribute value of each of a plurality of third simulated workpieces produced according to the processing parameters represented by the new element; For each of the plurality of third simulated workpieces, for the new element added to the set, determining quality inspection result data of each of the plurality of third simulated workpieces according to the second relationship and the determined quality attribute values ​​of the plurality of third simulated workpieces; For the new element added to the set, calculating a quality inspection pass rate according to the quality inspection result data of each of the plurality of third simulated workpieces; An optimal processing parameter determination module (303c) determines the element corresponding to the maximum value of all quality inspection pass rates obtained by statistics as the optimal value of the processing parameter of the production equipment (10) under the given ambient environment condition data.

15. The device (303) according to claim 14, characterized in that The adjustment module (303b) is specifically configured as follows when a new element is added to the set so that the expected value of the quality inspection pass rate obtained by statistics based on the new element has the largest increase compared with the maximum value of the quality inspection pass rate obtained by statistics in the past: Fitting a Gaussian process according to each element in the set and the quality inspection pass rate corresponding to the element obtained from previous statistics; The new element is obtained by calculation using the Gaussian process.

16. A computer readable medium, characterized in that The computer-readable medium stores a computer-readable code, and when the computer-readable code is executed by at least one processor, the method according to any one of claims 1 to 5 or 12 and 13 is executed.

Images