Model training method, device and equipment applied to wafer level acceptable test

Abstract

The application provides a model training method, device and equipment applied to wafer-level acceptable test. The method comprises the following steps: obtaining historical measurement data of a preset point on a wafer; the historical measurement data comprises historical online measurement values and wafer-level acceptable test data; determining online measurement value image data in a historical time according to the historical online measurement values, and determining wafer-level acceptable test data image data in the historical time according to the historical wafer-level acceptable test data; the online measurement value image data represents online measurement values of all position points, and the wafer-level acceptable test data image data represents wafer-level acceptable test data of all position points; inputting the online measurement value image data into an initial model to obtain output image; the output image represents predicted data of the wafer-level acceptable test data of all position points on the wafer; and if the output image satisfies a training completion condition, a prediction model of the wafer-level acceptable test data is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a model training method and device and equipment applied to wafer acceptance test. BACKGROUND

[0002] In semiconductor manufacturing, WAT (Wafer Acceptance Test) is a complete electrical property test process, which aims to ensure that the wafer has the highest quality and acceptable performance before leaving the manufacturing plant. It is a comprehensive and rigorous test process, which tests every device on the wafer to ensure that their quality and performance meet the expected level.

[0003] Therefore, the WAT test plays a crucial role in the entire wafer production and manufacturing process. The current WAT test needs to consume a large amount of manpower and time, and how to improve the efficiency of the WAT test becomes a problem to be solved. SUMMARY

[0004] The present application provides a model training method, device and equipment applied to wafer acceptance test, to improve the efficiency of the WAT test.

[0005] In a first aspect, the present application provides a model training method applied to wafer acceptance test, comprising:

[0006] Obtaining historical measurement data corresponding to a preset point on a wafer; wherein the historical measurement data comprises online measurement values obtained in a preset historical time and wafer acceptance test data, the online measurement values are used to represent the size information of the wafer obtained in the manufacturing process of the wafer, and the wafer acceptance test data is used to represent the data obtained by performing wafer acceptance test on the wafer; the preset point is used to represent a preset position point on the wafer;

[0007] According to the online measurement values in the historical measurement data, the image data of the online measurement values corresponding to the wafer in the historical time is determined, and according to the wafer acceptance test data in the historical measurement data, the image data of the wafer acceptance test data corresponding to the wafer in the historical time is determined; wherein the image data of the online measurement values is used to represent the online measurement values corresponding to all position points on the wafer, and the image data of the wafer acceptance test data is used to represent the wafer acceptance test data corresponding to all position points on the wafer;

[0008] inputting image data of the online measurement value corresponding to the wafer in the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer;

[0009] If it is determined according to the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time that the output image satisfies a preset training completion condition, a prediction model of wafer-level acceptable test data is obtained; wherein the prediction model of wafer-level acceptable test data is used to predict wafer-level acceptable test data of any position point on the wafer.

[0010] In a second aspect, the present application provides a prediction method of wafer-level acceptable test data, comprising:

[0011] obtaining current online measurement value of a wafer;

[0012] inputting the current online measurement value of the wafer into a prediction model of wafer-level acceptable test data to obtain predicted data of wafer-level acceptable test data corresponding to the current online measurement value of the wafer; wherein the prediction model of wafer-level acceptable test data is a model obtained based on the method of the first aspect.

[0013] In a third aspect, the present application provides a model training device applied to wafer-level acceptable test, comprising:

[0014] a data acquisition module, configured to acquire historical measurement data corresponding to a preset point on a wafer; wherein the historical measurement data comprises online measurement value and wafer-level acceptable test data obtained in a preset historical time, the online measurement value is used to represent size information of the wafer obtained in a manufacturing process of the wafer, and the wafer-level acceptable test data is used to represent data obtained by performing wafer-level acceptable test on the wafer; the preset point is used to represent a preset position point on the wafer;

[0015] an image determination module, configured to determine image data of the online measurement value corresponding to the wafer in the historical time according to the online measurement value in the historical measurement data, and determine image data of the wafer-level acceptable test data corresponding to the wafer in the historical time according to the wafer-level acceptable test data in the historical measurement data; wherein the image data of the online measurement value is used to represent online measurement value corresponding to all position points on the wafer, and the image data of the wafer-level acceptable test data is used to represent wafer-level acceptable test data corresponding to all position points on the wafer;

[0016] a model training module, configured to input image data of online measurement values corresponding to the wafer in the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer;

[0017] a training completion module, configured to, if it is determined according to image data of wafer-level acceptable test data corresponding to the wafer in the historical time that the output image satisfies a preset training completion condition, obtain a prediction model of wafer-level acceptable test data; wherein the prediction model of wafer-level acceptable test data is used to predict wafer-level acceptable test data of any position point on the wafer.

[0018] In a fourth aspect, the present application provides a prediction device of wafer-level acceptable test data, comprising:

[0019] a measurement value acquisition module, configured to acquire current online measurement values of a wafer; wherein the online measurement values are used to represent size information of the wafer obtained in a manufacturing process of the wafer;

[0020] a data prediction module, configured to input the current online measurement values of the wafer into a prediction model of wafer-level acceptable test data to obtain predicted data of wafer-level acceptable test data corresponding to the current online measurement values of the wafer; wherein the prediction model of wafer-level acceptable test data is a model obtained based on the device of the third aspect.

[0021] In a fifth aspect, the present application provides an electronic device, comprising a processor and a memory connected with the processor in communication;

[0022] the memory stores computer execution instructions;

[0023] the processor executes the computer execution instructions stored in the memory to implement the model training method applied to wafer-level acceptable test of the first aspect or the prediction method of wafer-level acceptable test data of the second aspect.

[0024] In a sixth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the model training method applied to wafer-level acceptable test of the first aspect or the prediction method of wafer-level acceptable test data of the second aspect.

[0025] In a seventh aspect, the present application provides a computer program product, comprising a computer program which, when executed by a processor, implements the model training method for wafer-level acceptable test or the prediction method for wafer-level acceptable test data according to the first aspect.

[0026] The model training method for wafer-level acceptable test, the device and the equipment provided by the present application obtain the online measurement value corresponding to the preset point on the wafer and the wafer-level acceptable test data in the historical time, determine the online measurement value of all points on the wafer and the wafer-level acceptable test data in the historical time, and thus obtain the image data of the online measurement value and the image data of the wafer-level acceptable test data. The image data of the online measurement value corresponding to the wafer in the historical time is input into the initial model constructed in advance, and the output image is obtained. The output image can represent the prediction data of the wafer-level acceptable test data of all position points on the wafer. Whether the output image meets the preset training completion condition is determined according to the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time. If yes, the prediction model of the wafer-level acceptable test data is obtained. The prediction model of the wafer-level acceptable test data can predict the wafer-level acceptable test data of any position point on the wafer. By expanding the online measurement value of the preset point and the wafer-level acceptable test data into the online measurement value of all points and the wafer-level acceptable test data, the amount of training data can be increased, and the training accuracy of the model can be improved. The prediction model can be used to predict the wafer-level acceptable test data before the wafer is completed, so as to avoid the situation that the wafer is found to be abnormal after the wafer is manufactured, and the foundry loses a large amount of manufacturing cost and time cost, and the efficiency of the wafer-level acceptable test is improved. BRIEF DESCRIPTION OF DRAWINGS

[0027] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the present application.

[0028] Figure 1 A flowchart of a model training method for wafer-level acceptable test provided by an embodiment of the present application is shown in the figure;

[0029] Figure 2 A schematic diagram of a preset point on a wafer provided by an embodiment of the present application is shown in the figure;

[0030] Figure 3 A schematic diagram of image data of an online measurement value provided by an embodiment of the present application is shown in the figure;

[0031] Figure 4 A flowchart of a model training method for wafer-level acceptable test provided by an embodiment of the present application is shown in the figure;

[0032] Figure 5 An initial image schematic diagram of historical online measurement values provided for an embodiment of the present application;

[0033] Figure 6 A flow schematic diagram of a wafer-level acceptable test data prediction method provided for an embodiment of the present application;

[0034] Figure 7 A structural block diagram of a model training device applied to wafer-level acceptable testing provided for an embodiment of the present application;

[0035] Figure 8 A structural block diagram of a model training device applied to wafer-level acceptable testing provided for an embodiment of the present application;

[0036] Figure 9 A structural block diagram of a wafer-level acceptable test data prediction device provided for an embodiment of the present application;

[0037] Figure 10 A structural block diagram of an electronic device provided for an embodiment of the present application;

[0038] Figure 11 A structural block diagram of an electronic device provided for an embodiment of the present application.

[0039] The specific embodiments of the present application have been shown through the above-described drawings, and will be described in more detail hereinafter. These drawings and written descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0040] In order to make the purposes, technical solutions, and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0041] It should be clear that the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work, fall within the scope of protection of the present application.

[0042] The following description refers to the accompanying drawings. Unless otherwise indicated, same or similar elements in different drawings are denoted by same reference numerals. The implementations described in the following example embodiments are not meant to represent all implementations consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0043] In the description of the present application, it is understood that the terms "first", "second", "third" and the like are only used to distinguish similar objects, and do not necessarily describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, "a plurality of" means two or more, unless otherwise specified. The association between the associated objects is described, and it means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after it.

[0044] It should be noted that due to the limitation of the length of the present application, all optional embodiments are not enumerated in the specification, and those skilled in the art should be able to think of any combination of technical features as long as the technical features do not contradict each other, which can constitute an optional embodiment. The embodiments are described in detail below.

[0045] In semiconductor manufacturing, WAT is a complete set of electrical property test flow, which aims to ensure that the wafers have the highest quality and acceptable performance before leaving the manufacturing plant. WAT test is the last few checkpoints of wafer manufacturing, which is used to verify whether the electrical parameters of semiconductor devices meet the specification requirements. It is a comprehensive and rigorous test process that tests every device on the wafer to ensure that their quality and performance meet the expected level.

[0046] WAT test data can monitor whether the wafer manufacturing process is abnormal, determine whether the product meets the electrical specification requirements of the process technology platform, and associate WAT data and yield, check the most relevant parameters, find out the problematic process steps, and then test and analyze specific WAT test parameters, improve the process, or develop the next generation of process technology platform.

[0047] Traditional WAT test plays a crucial role in the entire wafer production and manufacturing process, but it still has some shortcomings and challenges. WAT test usually requires a lot of time and resources to test and analyze multiple devices on each wafer one by one. The test of key measurement parameters is located between BEOL (Back End of Line) and OQA (Outgoing Quality Assure), at which time the wafer has completed the process manufacturing. If the WAT test is abnormal, it will cause the foundry to lose a lot of manufacturing cost and time cost, and the test efficiency is low.

[0048] The application provides a model training method, device and equipment applied to wafer-level acceptable testing.

[0049] The technical solutions of the application and how the technical solutions solve the above technical problems will be described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes can not be described again in some examples. The embodiments of the application will be described below with reference to the drawings.

[0050] Figure 1 is a flowchart of a model training method applied to wafer-level acceptable testing according to an embodiment of the application. The method can be executed by a model training device applied to wafer-level acceptable testing. As shown in the figure, the method comprises the following steps: Figure 1

[0051] S101, obtaining historical measurement data corresponding to a preset point on a wafer; wherein the historical measurement data comprises online measurement values obtained within a preset historical time and wafer-level acceptable test data, the online measurement values are used to represent size information of the wafer obtained in the manufacturing process of the wafer, and the wafer-level acceptable test data is used to represent data obtained by performing wafer-level acceptable test on the wafer; and the preset point is used to represent a preset position point on the wafer.

[0052] Exemplarily, the wafer is a circular silicon wafer, and a plurality of position points can be preset on the plane of the wafer as the preset points. Figure 2 is a schematic diagram of the preset points on the wafer. Figure 2 Five preset points are arranged in the figure, which are respectively located at the upper, lower, left and right positions and the center position of the wafer. The measurement data of the preset points within the preset historical time is obtained as the historical measurement data. For example, the measurement data of the preset points one month ago can be obtained. The measurement data can include Inline data and WAT data, that is, the historical measurement data can include Inline data and WAT data obtained within the preset historical time, and the Inline data can refer to online measurement values.

[0053] The online measurement values can be used to represent size information of the wafer obtained in the manufacturing process of the wafer. The manufacturing process of the wafer can include a plurality of process passes, and Inline data of different position points can be obtained in each process pass. For example, the Inline data can include THK (thickness), OVL (deviation), CD (critical dimension) and OCD (optical critical dimension) and other data categories. The WAT data can be used to represent data obtained by performing wafer-level acceptable test on the wafer, and the WAT data can be an electrical parameter of the wafer. ​

[0054] The historical measurement data corresponding to the preset point on the wafer can be obtained, and the Inline data and the WAT data of the wafer that has undergone the WAT at the preset point can be obtained. The historical measurement data of the preset point under multiple process passes can be obtained, for example, there are three process passes, and the Inline data and the WAT data of each preset point of the wafer under the three process passes are obtained respectively for the wafer that has been manufactured.

[0055] S102, according to the online measurement value in the historical measurement data, determine the image data of the online measurement value corresponding to the wafer in the historical time, and according to the wafer-level acceptable test data in the historical measurement data, determine the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time; wherein the image data of the online measurement value is used to represent the online measurement value corresponding to all position points on the wafer, and the image data of the wafer-level acceptable test data is used to represent the wafer-level acceptable test data corresponding to all position points on the wafer.

[0056] Exemplarily, the obtained historical measurement data is the historical measurement data of part of the points on the wafer, rather than the historical measurement data of the entire wafer. The historical measurement data of all position points on the wafer can be determined according to the historical measurement data of part of the points. Avoiding storing the historical measurement data of all position points on the wafer effectively saves storage space.

[0057] The Inline data of the preset point in the historical measurement data can be determined, and the Inline data corresponding to all position points in the wafer in the historical time is calculated according to the determined Inline data of the preset point. For example, the average value of the Inline data of the preset point can be calculated, and the average value is determined as the Inline data of the position points other than the preset point. The image data of the Inline data is obtained according to the Inline data of all position points in the wafer. The image data of the Inline data can be used to represent the Inline data corresponding to all position points on the wafer, for example, the size of the Inline data can be represented by color on the image data. Figure 3 The image data of the online measurement value is shown in the diagram. Figure 3 One block in the diagram is one position point on the wafer, and all blocks constitute a complete wafer, that is, a circle, and the pixel value size of each block represents the size of the Inline data of the position point. Different image data can be generated for Inline data of different data categories, for example, image data of THK, image data of OVL, image data of CD and image data of OCD can be generated.

[0058] The WAT data of the preset point in the historical measurement data can be determined, and the WAT data corresponding to all position points in the wafer in the historical time can be calculated according to the determined WAT data of the preset point. For example, the average value of the WAT data of the preset point can be calculated, and the average value is determined as the WAT data of the position points other than the preset point. The image data of the WAT data can be obtained according to the WAT data of all position points in the wafer. The image data of the WAT data can be used to represent the WAT data corresponding to all position points on the wafer, for example, the size of the WAT data can be represented by color on the image data.

[0059] In S103, the image data of the online measurement value corresponding to the wafer in the historical time is input into the initial model constructed in advance for processing to obtain an output image. The output image is used to represent the predicted data of the wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer.

[0060] For example, an initial model for predicting WAT data is constructed in advance, and in this embodiment, an initial neural network model can be constructed for training. The image data of the Inline data corresponding to the wafer in the historical time determined in S102 is input into the initial model constructed in advance as input data. If image data of different data categories is generated, image data of different data categories can be input into the model as input data. The initial model is provided with network layers such as convolution layer, pooling layer and full connection layer, which can process the input data to obtain output data corresponding to the input data, i.e. the output image corresponding to the image data of the Inline data. The trained model is used to predict the WAT data of the wafer, and in the training, the input data represents the Inline data of all position points on the wafer, and the output image can be used to represent the predicted data of the WAT data obtained by performing WAT on all position points on the wafer.

[0061] In this embodiment, the image data of the online measurement value corresponding to the wafer in the historical time is input into the initial model constructed in advance for processing to obtain an output image, including: inputting the image data of the online measurement value corresponding to the wafer in the historical time into the initial model constructed in advance; wherein the initial model constructed in advance is a U-shaped network structure; performing size expansion processing on the image data of the online measurement value corresponding to the wafer in the historical time according to the deconvolution network layer preset in the initial model to obtain the size expansion processed image data; performing feature extraction on the size expansion processed image data according to the convolution network layer preset in the initial model to obtain the output image.

[0062] Specifically, a suitable convolutional neural network model is designed as an initial model for training, so as to predict the WAT data. The network structure can adopt the structure of UNet (U-shaped network), and a deconvolution network layer can be added to the network structure to expand the image size of the input data.

[0063] The image data of the Inline data corresponding to the wafer at the historical time is input into the initial model constructed in advance. The image data of the Inline data corresponding to the wafer at the historical time is subjected to size expansion processing according to the preset deconvolution network layer in the initial model, to obtain the image data after the size expansion processing. For example, the image size is expanded by 4 times through the deconvolution component, i.e., from 9x11 to 36x44, to avoid the image being too small after multiple convolutions. The image data after the size expansion processing is subjected to feature extraction and calculation through the convolution layer and other network layers, and finally the output image of the WAT is output.

[0064] The beneficial effects of such a setting are that the neural network model can be designed to automatically predict the WAT data before the wafer manufacturing is completed, thereby improving the efficiency of the WAT test. The deconvolution layer in the model can provide the neural network with more fine-grained high-dimensional features, thereby improving the accuracy of the WAT parameter prediction.

[0065] In S104, if it is determined that the output image meets the preset training completion condition according to the image data of the wafer-level acceptable test data corresponding to the wafer at the historical time, a wafer-level acceptable test data prediction model is obtained; the wafer-level acceptable test data prediction model is used to predict the wafer-level acceptable test data at any position point on the wafer.

[0066] Exemplarily, the training completion condition is preset, and the initial model is subjected to multiple iterative training. After each output image is obtained, it is determined whether the output image meets the preset training completion condition. If yes, it is determined that the model training is completed, and the wafer-level acceptable test data prediction model is obtained; if no, it is determined that the model training is not completed, and the training needs to be continued. The wafer-level acceptable test data prediction model after the training can be used to predict the wafer-level acceptable test data at any position point on the wafer.

[0067] A loss function can be designed in advance, which is used to calculate the difference between the output image and the image data of the wafer-level acceptable test data corresponding to the wafer at the historical time, and to determine whether the calculated difference meets the preset training completion condition. If not, the parameters of the network model are iteratively updated by using the optimizer through the back propagation, and finally the training of the prediction model is completed.

[0068] This application provides a model training method for wafer-level acceptable testing. By acquiring online measurement values ​​and wafer-level acceptable test data corresponding to preset points on the wafer within a historical time period, the method determines the online measurement values ​​and wafer-level acceptable test data for all points on the wafer within that historical time period, thereby obtaining image data of the online measurement values ​​and image data of the wafer-level acceptable test data. The image data of the online measurement values ​​corresponding to the wafer within the historical time period is input into a pre-constructed initial model, and an output image is obtained. The output image can represent the predicted wafer-level acceptable test data for all locations on the wafer. Based on the image data of the wafer-level acceptable test data corresponding to the wafer within the historical time period, it is determined whether the output image meets preset training completion conditions. If so, a prediction model for wafer-level acceptable test data is determined, which can predict the wafer-level acceptable test data for any location on the wafer. By expanding the online measurement values ​​and wafer-level acceptable test data from preset locations to all locations, the amount of training data can be increased, thereby improving the training accuracy of the model. Predictive models can predict wafer-level acceptable test data before wafer fabrication is complete, avoiding significant manufacturing and time losses for foundries due to anomalies discovered after wafer manufacturing is finished, thus improving the efficiency of wafer-level acceptable testing.

[0069] Figure 4 This is a flowchart illustrating a model training method for wafer-level acceptable testing, provided as an embodiment of this application. This embodiment is an optional embodiment based on the above embodiments.

[0070] In this embodiment, obtaining historical measurement data corresponding to preset points on the wafer can be further refined as follows: obtaining online measurement values ​​corresponding to a preset first point on the wafer within a preset historical time period, which are historical online measurement values; and obtaining wafer-level acceptable test data corresponding to a preset second point on the wafer within a preset historical time period, which are historical wafer-level acceptable test data; wherein, the first point includes at least one location point, and the second point includes at least one location point; preprocessing the historical online measurement values ​​and historical wafer-level acceptable test data to determine the target point among the first and second points, thereby obtaining the historical online measurement values ​​and historical wafer-level acceptable test data corresponding to the target point.

[0071] like Figure 4 As shown, the method includes the following steps:

[0072] S401, obtaining, as the historical inline measurement value, the inline measurement value corresponding to the preset first point on the wafer obtained within the preset historical time, and obtaining, as the historical wafer-level acceptable test data, the wafer-level acceptable test data corresponding to the preset second point on the wafer obtained within the preset historical time; wherein the first point includes at least one position point, and the second point includes at least one position point.

[0073] Exemplarily, a plurality of position points can be preset as the first point. The first point is a position point for obtaining Inline data, that is, from the Inline data obtained within the historical time, the inline measurement value corresponding to the preset first point on the wafer is obtained as the historical inline measurement value.

[0074] A plurality of position points can also be preset as the second point. The position points of the second point can be completely consistent with the position points of the first point, or can be inconsistent. For example, the first point has 13 position points, and the second point has 21 position points, wherein the positions of 9 position points in the first point and the second point are consistent. The second point is a position point for obtaining WAT data, that is, from the WAT data obtained within the historical time, the WAT data corresponding to the preset second point on the wafer is obtained as the historical wafer-level acceptable test data. In the manufacturing process of the wafer, the Inline data of each first point and the WAT data of the second point can be stored in real time, so as to facilitate the acquisition of historical measurement data. Moreover, it is not necessary to store the Inline data and the WAT data of each position point on the wafer, thereby effectively saving storage space.

[0075] In the embodiment, the manufacturing process of the wafer includes at least two process passes; after obtaining, as the historical inline measurement value, the inline measurement value corresponding to the preset first point on the wafer obtained within the preset historical time, the method further includes: determining the historical inline measurement value corresponding to each process pass obtained within the preset historical time; determining the correlation between the historical inline measurement values corresponding to the two process passes; and if the correlation meets a preset rejection condition, retaining the historical inline measurement value corresponding to one process pass from the historical inline measurement values corresponding to the two process passes.

[0076] Specifically, the manufacturing process of the wafer can include a plurality of process passes, and the finished product of the wafer can be obtained after the plurality of process passes. In each process pass, the Inline data of the wafer can change. Therefore, when the historical inline measurement value is obtained, the historical inline measurement values in the plurality of process passes can be obtained. That is, the inline measurement values of each first point in different process passes occurring within the preset historical time are obtained.

[0077] From the obtained historical online measurement values, historical online measurement values corresponding to each process pass are determined. For example, if there are three process passes, historical online measurement values of each first site under the three process passes can be obtained. Correlations between historical online measurement values corresponding to two process passes are determined. Correlations between historical online measurement values under each two process passes can be determined, or a process pass can be determined as a main link and other process passes as secondary links, and historical online measurement values of the main link are compared with historical online measurement values of each secondary link to determine correlations between the main link and each secondary link. For example, three process passes can be compared in pairs to determine correlations between historical online measurement values of a first process pass and a second process pass, correlations between historical online measurement values of the first process pass and a third process pass, and correlations between historical online measurement values of the second process pass and the third process pass. For another example, a first process pass is determined as a main link, and correlations between historical online measurement values of the first process pass and a second process pass and correlations between historical online measurement values of the first process pass and a third process pass are determined.

[0078] A calculation formula of the correlation can be preset, and the correlation can represent a similarity between historical online measurement values. For example, an average value of historical online measurement values of a first site under each process pass can be determined, and a difference between average values corresponding to two process passes is determined. The smaller the difference is, the greater the correlation between historical online measurement values of the two process passes is. Online measurement values can include multiple data categories, such as CD, OCD, THK, and OVL. When the correlation is determined, online measurement values of different data categories can be respectively subjected to correlation calculation. For example, correlations of CD data under a first process pass and a second process pass are determined, and correlations of OCD data under the first process pass and the second process pass are determined.

[0079] A rejection condition is preset, and the rejection condition can be used to determine whether to reject the historical inline measurement value of the process pass. It is determined whether the correlation between the historical inline measurement values corresponding to two process passes meets the preset rejection condition, for example, the preset correlation threshold is 0.9, and it is determined whether the calculated correlation is greater than 0.9. If yes, one of the historical inline measurement values corresponding to the two process passes is retained, and the historical inline measurement value corresponding to the other process pass is rejected. If no, the historical inline measurement values of the two process passes are both retained. When rejecting, one of the two process passes can be randomly selected for data rejection. A rejection rule can also be preset, for example, the process pass after the flow can be selected for data rejection, or the data amount of the two process passes can be determined, and the process pass with smaller data amount can be selected for data rejection.

[0080] The beneficial effect of such a setting is that the data in different process passes is traversed for data rejection, the data amount for subsequent calculation is reduced, the dimensionality of the data is reduced, and the efficiency of model training is improved.

[0081] In the embodiment, after obtaining the inline measurement value corresponding to the preset first point on the wafer within the preset historical time, the method further includes: if the number of the first points is not the preset point number threshold, deleting the inline measurement value corresponding to the preset first point on the wafer.

[0082] Specifically, the number of the first points and the number of the second points can be preset by the foundry of the wafer, and the number of the first points preset by different foundries can be different. A first point number threshold can be preset to determine whether the number of the first points of the wafer obtained is the preset point number threshold. If yes, subsequent data processing is continued; if no, the historical measurement data of the wafer can be deleted, that is, the data of the wafer is not used for training. For example, the preset point number threshold is 13, and the Inline data with a first point total number other than 13 is removed.

[0083] The beneficial effect of such a setting is that the historical measurement data is uniformly processed, and the different data amounts of different wafers do not affect the training efficiency and accuracy.

[0084] In this embodiment, the wafer-level acceptable test data corresponding to the preset second point on the wafer obtained within the preset historical time is obtained as the historical wafer-level acceptable test data, including: obtaining all wafer-level acceptable test data corresponding to the preset second point on the wafer obtained within the preset historical time; selecting wafer-level acceptable test data of a preset data category from all wafer-level acceptable test data as the historical wafer-level acceptable test data.

[0085] Specifically, in addition to screening the historical online measurement values of the first point, the historical wafer-level acceptable test data of the second point can also be screened. Different data categories can also be included in the wafer-level acceptable test data, and one or more data categories are set in advance as the data category to be retained. All wafer-level acceptable test data corresponding to the second point on the wafer obtained within the preset historical time is obtained, and wafer-level acceptable test data of a preset data category is selected from all wafer-level acceptable test data as the historical wafer-level acceptable test data to be retained. That is, information irrelevant to WAT prediction is excluded from all WAT data.

[0086] The beneficial effect of such setting is that important WAT information is retained, information redundancy is avoided to affect model training, and the efficiency and accuracy of model training are improved.

[0087] S402, pre-process the historical online measurement values and the historical wafer-level acceptable test data, determine the target point in the first point and the second point, and obtain the historical online measurement values and the historical wafer-level acceptable test data corresponding to the target point.

[0088] Exemplarily, the number and position of the first point and the second point can be different, and therefore, after obtaining the historical online measurement values of the first point and the historical wafer-level acceptable test data of the second point, the historical online measurement values and the historical wafer-level acceptable test data can be pre-processed, and the pre-processing can be alignment processing of the historical online measurement values and the historical wafer-level acceptable test data. One or more target points are determined from the first point and the second point. The target point in the first point is consistent with the target point in the second point, that is, the point repeated in the first point and the second point can be determined as the target point. The historical online measurement values corresponding to the target point and the historical wafer-level acceptable test data corresponding to the target point are determined.

[0089] In the embodiment, the historical online measurement values and the historical wafer-level acceptable test data are preprocessed to determine target points in the first points and the second points, including: adding the first points to a preset spatial coordinate system to obtain an initial image of the historical online measurement values; in the initial image of the historical online measurement values, a coordinate point of the first point is used to represent a position of the first point in the wafer, and a pixel value of the first point is used to represent the historical online measurement value of the first point; adding the second points to the preset spatial coordinate system to obtain an initial image of the historical wafer-level acceptable test data; in the initial image of the historical wafer-level acceptable test data, a coordinate point of the second point is used to represent a position of the second point in the wafer, and a pixel value of the second point is used to represent the historical wafer-level acceptable test data of the second point; and overlapping the initial image of the historical online measurement values and the initial image of the historical wafer-level acceptable test data to determine points that coincide in the initial image of the historical online measurement values and the initial image of the historical wafer-level acceptable test data as target points.

[0090] Specifically, according to the historical online measurement value of the first point, the data in the form of an image of the historical online measurement value is obtained as an initial image. For example, the first point can be added to a preset spatial coordinate. In the initial image, the position of the first point in the wafer and the size of the historical online measurement value at the first point can be represented. That is, the coordinate point of the first point can be used to represent the position of the first point in the wafer, and the pixel value size can be used to represent the size of the historical online measurement value. For different data categories in the historical online measurement value, an initial image corresponding to each data category can be generated. That is, for the CD data of a certain process pass, an initial image can be generated; and for the OCD data of a certain process pass, an initial image can also be generated. Figure 5 An initial image of the historical online measurement value is shown in the figure. Figure 5 There are 21 first points in the figure, which are distributed on the wafer. The pixel value of each block can represent the historical online measurement value at the position.

[0091] According to the historical wafer-level acceptable test data of the second point, the data in the form of an image of the historical wafer-level acceptable test data is obtained as an initial image. For example, the second point can be added to a preset spatial coordinate. In the initial image, the position of the second point in the wafer and the size of the historical wafer-level acceptable test data at the second point can be represented. That is, the coordinate point of the second point can be used to represent the position of the second point in the wafer, and the pixel size can be used to represent the size of the historical wafer-level acceptable test data. For different data categories in the historical wafer-level acceptable test data, an initial image corresponding to each data category can be generated.

[0092] After obtaining the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data, the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data can be overlapped and aligned. The points in the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data that are coincident are determined as target points.

[0093] The beneficial effect of such arrangement is that the historical measurement data is visualized in the form of images based on image processing technology, making data analysis and problem identification more intuitive and efficient, and improving the efficiency of data processing. Through alignment of the initial images, data redundancy can be reduced, further improving the efficiency of model training.

[0094] S403, according to the online measurement value in the historical measurement data, determine the image data of the online measurement value corresponding to the wafer in the historical time, and according to the wafer-level acceptable test data in the historical measurement data, determine the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time; wherein the image data of the online measurement value is used to represent the online measurement value corresponding to all position points on the wafer, and the image data of the wafer-level acceptable test data is used to represent the wafer-level acceptable test data corresponding to all position points on the wafer.

[0095] Exemplarily, the historical online measurement value and the historical wafer-level acceptable test data are only Inline data and WAT data at the target point of the wafer, and there are many position points on the entire wafer that do not have Inline data and WAT data. After obtaining the historical online measurement value and the historical wafer-level acceptable test data, the historical online measurement value and the historical wafer-level acceptable test data can be supplemented. According to the historical online measurement value and the historical wafer-level acceptable test data of the target point, the historical online measurement value and the historical wafer-level acceptable test data of all position points on the wafer are determined. The historical online measurement value of all position points on the wafer is determined as the image data of the historical online measurement value, and the historical wafer-level acceptable test data of all position points on the wafer is determined as the image data of the historical wafer-level acceptable test data. The image data of the historical online measurement value can be determined based on the initial image of the historical online measurement value, and the image data of the historical wafer-level acceptable test data can be determined based on the initial image of the historical wafer-level acceptable test data.

[0096] In the embodiment, according to the online measurement value in the historical measurement data, the image data of the online measurement value corresponding to the wafer in the historical time is determined, including: obtaining the target point image of the historical online measurement value according to the overlapping result of the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data; wherein the target point image of the historical online measurement value is used to represent the part of the initial image of the historical online measurement value which overlaps with the initial image of the historical wafer-level acceptable test data; and performing supplementary processing on the target point image of the historical online measurement value according to a preset spline interpolation algorithm to obtain the image data of the online measurement value corresponding to the wafer in the historical time.

[0097] Specifically, the initial image of the historical online measurement value is overlapped with the initial image of the historical wafer-level acceptable test data to obtain an overlapping result, which can be an image of the overlapping part of the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data. The preset point in the image of the overlapping part is the target point, that is, according to the overlapping result, the part of the initial image of the historical online measurement value which overlaps with the initial image of the historical wafer-level acceptable test data is determined as the target point image of the historical online measurement value.

[0098] The spline interpolation algorithm is preset, and there are blank position points in the target point image of the historical online measurement value. According to the spline interpolation algorithm, the historical online measurement value of the blank position point can be calculated to supplement the target point image of the historical online measurement value. For example, Figure 5 the image before filling, Figure 3 the image after filling, and 9*11 image data can be obtained after filling. In the embodiment, the specific calculation process of the spline interpolation algorithm is not specifically limited, for example, a cubic interpolation algorithm can be used.

[0099] The beneficial effect of such setting is that by filling the blank position points, the data amount of the historical online measurement value is improved, thereby providing a large amount of data set for training and improving the accuracy of model training.

[0100] In the embodiment, according to wafer-level acceptable test data in historical measurement data, the image data of wafer-level acceptable test data corresponding to the wafer in the historical time is determined, including: obtaining a target point image of the historical wafer-level acceptable test data according to an overlapping result of an initial image of the historical online measurement value and an initial image of the historical wafer-level acceptable test data; wherein the target point image of the historical wafer-level acceptable test data is used to represent a part of the initial image of the historical wafer-level acceptable test data that overlaps with the initial image of the historical online measurement value; and performing supplementary processing on the target point image of the historical wafer-level acceptable test data according to a preset spline interpolation algorithm to obtain the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time.

[0101] Specifically, the initial image of the historical online measurement value is overlapped with the initial image of the historical wafer-level acceptable test data to obtain an overlapping result, which can be an image of an overlapping part of the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data. The points in the image of the overlapping part are target points, that is, according to the overlapping result, the part of the initial image of the historical wafer-level acceptable test data that overlaps with the initial image of the historical online measurement value is determined as the target point image of the historical wafer-level acceptable test data. The positions and quantities of the target points in the target point image of the historical online measurement value and the target point image of the historical wafer-level acceptable test data are the same, but the meanings of the pixel values of the target points are different. The pixel value of a target point in the target point image of the historical online measurement value represents the historical online measurement value of the point, and the pixel value of a target point in the target point image of the historical wafer-level acceptable test data represents the historical wafer-level acceptable test data of the point.

[0102] The spline interpolation algorithm is preset, and the blank position points exist in the target point image of the historical wafer-level acceptable test data. According to the spline interpolation algorithm, the historical wafer-level acceptable test data of the blank position points can be calculated to supplement the target point image of the historical wafer-level acceptable test data. For example, Figure 5 the image before filling, Figure 3 the image after filling, and 9*11 image data can be obtained after filling. In the embodiment, the specific calculation process of the spline interpolation algorithm is not specifically limited, for example, a cubic interpolation algorithm can be used.

[0103] The beneficial effect of such setting is that by filling the blank position points, the data amount of the historical wafer-level acceptable test data is improved, thereby providing a large amount of data set for training and improving the accuracy of model training.

[0104] S404, input the image data of the online measurement value corresponding to the wafer in the historical time into the initial model constructed in advance for processing to obtain an output image; wherein the output image is used to represent the predicted data of the wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer.

[0105] Exemplarily, this step can refer to the above step S103, and will not be repeated here.

[0106] S405, if it is determined that the output image meets the preset training completion condition according to the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time, a prediction model of the wafer-level acceptable test data is obtained; wherein the prediction model of the wafer-level acceptable test data is used to predict the wafer-level acceptable test data of any position point on the wafer.

[0107] Exemplarily, this step can refer to the above step S104, and will not be repeated here.

[0108] The model training method applied to wafer-level acceptable test provided by the embodiments of the present application obtains the online measurement value and the wafer-level acceptable test data corresponding to the preset point on the wafer in the historical time, determines the online measurement value and the wafer-level acceptable test data of all points on the wafer in the historical time, and thus obtains the image data of the online measurement value and the image data of the wafer-level acceptable test data. The image data of the online measurement value corresponding to the wafer in the historical time is input into the initial model constructed in advance, and an output image is obtained by output. The output image can represent the predicted data of the wafer-level acceptable test data of all position points on the wafer. Whether the output image meets the preset training completion condition is determined according to the image data of the wafer-level acceptable test data corresponding to the wafer in the historical time. If yes, a prediction model of the wafer-level acceptable test data is obtained, which can predict the wafer-level acceptable test data of any position point on the wafer. By expanding the online measurement value and the wafer-level acceptable test data of the preset point to the online measurement value and the wafer-level acceptable test data of all points, the amount of training data can be increased, and the training accuracy of the model can be improved. Through the prediction model, the wafer-level acceptable test data can be predicted before the wafer is completed, avoiding the loss of a large amount of manufacturing cost and time cost of the foundry due to the test of abnormality after the wafer has been manufactured, and improving the efficiency of wafer-level acceptable test.

[0109] Figure 6 A flowchart of a wafer-level acceptable test data prediction method according to an embodiment of the present application is shown. The method can be executed by a wafer-level acceptable test data prediction device. As shown in the figure, the method comprises the following steps: Figure 6 ​

[0110] S601, acquire current inline measurement value of wafer.

[0111] Exemplarily, after the prediction model of wafer-level acceptable test data is trained, the WAT data can be predicted in the wafer manufacturing process, so that the electrical parameters of the wafer are analyzed in advance according to the prediction data.

[0112] The current Inline data of the wafer can be acquired in real time or at a fixed time during the wafer manufacturing process. The current Inline data can be the Inline data of the process pass that has been experienced. For example, there are three process passes in the wafer manufacturing process, and the second process pass has been completed. The current Inline data acquired can be the Inline data of the first process pass and the Inline data of the second process pass.

[0113] The acquired Inline data can be the Inline data of the preset point in the wafer. The preset point can be a pre-set position point in part of the wafer, or can be a position point in the whole wafer. The position and number of the preset point can be set according to actual needs.

[0114] The acquired inline measurement value can be in the form of an image. The acquired inline measurement value of the preset point can be added to the preset spatial coordinate system to obtain an image. The coordinate point of the preset point in the spatial coordinate system represents the position of the preset point in the wafer, and the pixel value of the preset point represents the inline measurement value of the preset point. The inline measurement value can include multiple data categories. For each process pass, an image can be generated for each data category. For example, there are three process passes, and there are four data categories. Therefore, 12 images of the inline measurement value can be obtained.

[0115] S602, input the current inline measurement value of the wafer into the prediction model of wafer-level acceptable test data to obtain the prediction data of the wafer-level acceptable test data corresponding to the current inline measurement value of the wafer.

[0116] Exemplarily, the current inline measurement value is taken as the input data, for example, an image corresponding to the inline measurement value can be taken as the input data. The current inline measurement value of the wafer is input into the trained wafer-level acceptable test data prediction model, and the input data is processed through network layers such as deconvolution layers and convolution layers in the model to obtain output data of the model. The output data can be in the form of an image, and the coordinates of the preset sites can be represented in the image of the output data, and the pixel value at the preset site represents the predicted value of the WAT data at the site. In the wafer-level acceptable test data prediction model, a calculation formula of the WAT data can be set, for example, the WAT data can be calculated as the predicted data according to the Inline data and the data of the preset recipe (recipe) and other manufacturing processes.

[0117] The embodiment of the present application provides a wafer-level acceptable test data prediction method. By obtaining the current inline measurement value of the wafer, the WAT data can be predicted before the wafer manufacturing is completed. The inline measurement value can be taken as the input data and input into the wafer-level acceptable test data prediction model to automatically obtain the predicted data of the WAT data, thereby improving the efficiency of the WAT test. A large amount of time and resources are saved, the electrical parameters of the wafer are analyzed before the wafer manufacturing is completed, potential problems are predicted and identified, and strong support is provided for quality control and optimization in the semiconductor manufacturing process.

[0118] Figure 7 A structure block diagram of a model training device applied to wafer-level acceptable test is provided in the embodiment of the present application. For ease of illustration, only parts related to the embodiment of the present disclosure are shown. Referring to Figure 7 , the device comprises a data acquisition module 701, an image determination module 702, a model training module 703 and a training completion module 704.

[0119] The data acquisition module 701 is configured to acquire historical measurement data corresponding to a preset site on a wafer. The historical measurement data comprises inline measurement values obtained within a preset historical time and wafer-level acceptable test data. The inline measurement values are used to represent size information of the wafer obtained in a wafer manufacturing process, and the wafer-level acceptable test data is used to represent data obtained by performing wafer-level acceptable test on the wafer. The preset site is used to represent a preset position point on the wafer.

[0120] The image determination module 702 is configured to determine image data of online measurement values corresponding to the wafer in the historical time according to the online measurement values in the historical measurement data, and determine image data of wafer-level acceptable test data corresponding to the wafer in the historical time according to wafer-level acceptable test data in the historical measurement data; wherein the image data of online measurement values is used to represent online measurement values corresponding to all position points on the wafer, and the image data of wafer-level acceptable test data is used to represent wafer-level acceptable test data corresponding to all position points on the wafer.

[0121] The model training module 703 is configured to input the image data of online measurement values corresponding to the wafer in the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer.

[0122] The training completion module 704 is configured to obtain a wafer-level acceptable test data prediction model if it is determined that the output image satisfies a preset training completion condition according to the image data of wafer-level acceptable test data corresponding to the wafer in the historical time; wherein the wafer-level acceptable test data prediction model is used to predict wafer-level acceptable test data of any position point on the wafer.

[0123] Figure 8 A structural block diagram of a model training device for wafer-level acceptable test provided by an embodiment of the present application is shown in Figure 7 On the basis of the embodiment shown in Figure 8 The data acquisition module 701 includes an acquisition unit 7011 and a preprocessing unit 7012.

[0124] The acquisition unit 7011 is configured to acquire online measurement values corresponding to a preset first point on the wafer obtained in a preset historical time as historical online measurement values, and acquire wafer-level acceptable test data corresponding to a preset second point on the wafer obtained in the preset historical time as historical wafer-level acceptable test data; wherein the first point includes at least one position point, and the second point includes at least one position point.

[0125] The preprocessing unit 7012 is configured to preprocess the historical online measurement values and the historical wafer-level acceptable test data, determine a target point in the first point and the second point, and obtain historical online measurement values and historical wafer-level acceptable test data corresponding to the target point.

[0126] In one example, the preprocessing unit 7012 is specifically configured to:

[0127] add the first point to the preset spatial coordinate system to obtain an initial image of the historical online measurement value; wherein in the initial image of the historical online measurement value, a coordinate point of the first point is used to represent a position of the first point in the wafer, and a pixel value of the first point is used to represent a historical online measurement value of the first point;

[0128] add the second point to the preset spatial coordinate system to obtain an initial image of the historical wafer-level acceptable test data; wherein in the initial image of the historical wafer-level acceptable test data, a coordinate point of the second point is used to represent a position of the second point in the wafer, and a pixel value of the second point is used to represent historical wafer-level acceptable test data of the second point;

[0129] overlap the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data to determine a point that coincides in the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data as the target point.

[0130] In one example, the process of manufacturing the wafer includes at least two process passes;

[0131] The apparatus further includes:

[0132] The data elimination module is configured to, after obtaining the online measurement value corresponding to the preset first point on the wafer obtained within the preset historical time, determine the historical online measurement value corresponding to each process pass obtained within the preset historical time as the historical online measurement value.

[0133] determine the correlation between the historical online measurement values corresponding to the two process passes;

[0134] If the correlation meets the preset elimination condition, retain the historical online measurement value corresponding to one process pass from the historical online measurement values corresponding to the two process passes.

[0135] In one example, the apparatus further includes:

[0136] The data deletion module is configured to, after obtaining the online measurement value corresponding to the preset first point on the wafer obtained within the preset historical time, if the number of the first points is not the preset point number threshold, delete the online measurement value corresponding to the preset first point on the wafer.

[0137] In one example, the obtaining unit 7011 includes:

[0138] a WAT data obtaining subunit, configured to obtain all wafer-level acceptable test data corresponding to a preset second point on the wafer obtained within a preset historical time;

[0139] a WAT data screening subunit, configured to select wafer-level acceptable test data of a preset data category from the all wafer-level acceptable test data, as the historical wafer-level acceptable test data.

[0140] In one example, the image determination module 702 includes:

[0141] a first obtaining unit, configured to obtain a target point image of the historical online metrology value according to an overlap result of an initial image of the historical online metrology value and an initial image of the historical wafer-level acceptable test data; the target point image of the historical online metrology value is used to represent a portion of the initial image of the historical online metrology value that overlaps with the initial image of the historical wafer-level acceptable test data;

[0142] a first supplement unit, configured to perform supplement processing on the target point image of the historical online metrology value according to a preset spline interpolation algorithm, to obtain image data of the online metrology value corresponding to the wafer within the historical time.

[0143] In one example, the image determination module 702 includes:

[0144] a second obtaining unit, configured to obtain a target point image of the historical wafer-level acceptable test data according to an overlap result of an initial image of the historical online metrology value and an initial image of the historical wafer-level acceptable test data; the target point image of the historical wafer-level acceptable test data is used to represent a portion of the initial image of the historical wafer-level acceptable test data that overlaps with the initial image of the historical online metrology value;

[0145] a second supplement unit, configured to perform supplement processing on the target point image of the historical wafer-level acceptable test data according to a preset spline interpolation algorithm, to obtain image data of the wafer-level acceptable test data corresponding to the wafer within the historical time.

[0146] In one example, the model training module 703 is specifically configured to:

[0147] input the image data of the online metrology value corresponding to the wafer within the historical time into a pre-constructed initial model; the pre-constructed initial model is a U-shaped network structure.

[0148] According to a preset deconvolution network layer in the initial model, the image data of the online measurement value corresponding to the wafer in the history time is subjected to size expansion processing to obtain size expansion processed image data.

[0149] According to a preset convolution network layer in the initial model, the size expansion processed image data is subjected to feature extraction to obtain the output image.

[0150] Figure 9 A structural block diagram of a wafer-level acceptable test data prediction device provided by an embodiment of the present application is shown in the figure, only parts related to the embodiments of the present disclosure are shown for the convenience of description. Figure 9 The device comprises a measurement value acquisition module 901 and a data prediction module 902.

[0151] The measurement value acquisition module 901 is configured to acquire the current online measurement value of the wafer, wherein the online measurement value is used to represent the size information of the wafer obtained in the wafer manufacturing process.

[0152] The data prediction module 902 is configured to input the current online measurement value of the wafer into a wafer-level acceptable test data prediction model to obtain prediction data of wafer-level acceptable test data corresponding to the current online measurement value of the wafer, wherein the wafer-level acceptable test data prediction model is a model obtained based on the device of any embodiment of the present application.

[0153] Figure 10 A structural block diagram of an electronic device provided by an embodiment of the present application is shown in the figure. Figure 10 The electronic device comprises a memory 1001 and a processor 1002; the memory 1001 is configured to store the memory of the processor 1002 executable instructions.

[0154] The processor 1002 is configured to execute the method provided by the above-mentioned embodiments.

[0155] The electronic device further comprises a receiver 1003 and a transmitter 1004. The receiver 1003 is configured to receive the instructions and data sent by other devices, and the transmitter 1004 is configured to send instructions and data to external devices.

[0156] Figure 11 A block diagram of an electronic device according to an exemplary embodiment is shown, which can be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0157] Device 1100 can include one or more of the following components: a processing component 1102, a memory 1104, a power supply component 1106, a multimedia component 1108, an audio component 1110, an input / output (I / O) interface 1112, a sensor component 1114, and a communication component 1116.

[0158] The processing component 1102 usually controls overall operations of the device 1100, such as operations associated with displaying, making phone calls, data communications, camera operations and recording operations. The processing component 1102 can include one or more processors 1120 to execute instructions to complete all or part of steps of the methods described above. In addition, the processing component 1102 can include one or more modules to facilitate

[0159] The memory 1104 is configured to store various types of data to support operations of the device 1100. Examples of these data include instructions for any application or method operating on the device 1100, contact data, phonebook data, messages, pictures, videos, and the like. The memory 1104 can be implemented by any type of volatile or non-volatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic storage, flash memory, magnetic or optical disk.

[0160] The power supply component 1106 provides power for the various components of the device 1100. The power supply component 1106 can include a power supply management system, one or more power sources, and other components associated with generating, managing and distributing power for the device 1100.

[0161] The multimedia component 1108 includes a screen providing an output interface between the device 1100 and a user. In some embodiments, the screen includes a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, swiping, and gestures on the touch panel. The touch sensors can not only sense a boundary of a touching or swiping action, but also detect duration and pressure related to the touching or swiping action. In some embodiments, the multimedia component 1108 includes a front camera and / or a rear camera. When the device 1100 is in an operation mode, such as a camera mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front and rear camera can be a fixed optical lens system or have a focal length and optical zooming capability.

[0162] The audio component 1110 is configured to output and / or input audio signals. For example, the audio component 1110 includes a microphone (MIC) configured to receive external audio signals when the device 1100 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 1104 or transmitted via the communication component 1116. In some embodiments, the audio component 1110 also includes a speaker for outputting audio signals.

[0163] The I / O interface 1112 provides an interface between the processing component 1102 and peripheral interface modules, such as a keypad, click wheel, button, and the like. The buttons can include, but are not limited to, a home button, a volume button, a start button, and a lock button.

[0164] The sensor component 1114 includes one or more sensors to provide various state assessments for the device 1100. For example, the sensor component 1114 can detect an open / closed state of the device 1100, relative positioning of components, such as a display and a keypad of the device 1100, a change in position of the device 1100 or a component of the device 1100, presence or absence of user contact with the device 1100, an orientation or acceleration / deceleration of the device 1100, and a temperature change of the device 1100. The sensor component 1114 can include a proximity sensor configured to detect presence of a nearby object without any physical touch. The sensor component 1114 can further include a light sensor, such as a CMOS or CCD image sensor, for use in an imaging application. In some embodiments, the sensor component 1114 can further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0165] The communication component 1116 is configured to facilitate wired or wireless communication between the device 1100 and other devices. The device 1100 can access a wireless network based on a communication standard, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 1116 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 1116 further includes a Near Field Communication (NFC) module to facilitate close proximity communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0166] In an exemplary embodiment, the device 1100 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic elements, for performing the above-described methods.

[0167] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions, such as the memory 1104 including instructions, is also provided, which can be executed by the processor 1120 of the device 1100 to complete the above-described methods. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disc, and the like.

[0168] A non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of a terminal device, enables the terminal device to perform the above-described model training method and wafer-level acceptable test data prediction method of the terminal device.

[0169] The present application also discloses a computer program product, comprising a computer program which, when executed by a processor, implements the method as described in the embodiments.

[0170] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, specially designed application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0171] Program code for carrying out methods of the present application can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, fully on a machine and partially on a remote machine or entirely on a remote machine or electronic device.

[0172] In the context of the present application, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), compact disc read-only memories (CD-ROMs), magnetic storage devices, or any suitable combination of the foregoing.

[0173] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0174] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0175] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as cloud computing server or cloud host, which is a host product in the cloud computing service system, and solves the defects of large management difficulty and weak business scalability in traditional physical host and VPS (Virtual Private Server, or VPS for short) services. The server can also be a distributed system server or a server combined with a blockchain. It should be understood that the various forms of procedures shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved, and this document does not limit this.

[0176] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0177] It is to be understood that the application is not limited to the precise construction herein disclosed and shown in the drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application is limited only by the claims that follow.

Claims

1. A method for training a model applied to wafer-level acceptable test, characterized in that, The method comprises: acquiring historical measurement data corresponding to preset point positions on a wafer; wherein the historical measurement data comprises online measurement values obtained within a preset historical time and wafer-level acceptable test data, the online measurement values being used to represent size information of the wafer obtained in a manufacturing process of the wafer, and the wafer-level acceptable test data being used to represent data obtained by performing wafer-level acceptable test on the wafer; and the preset point positions being used to represent preset position points on the wafer; determining image data of the online measurement values corresponding to the wafer within the historical time according to the online measurement values in the historical measurement data, and determining image data of the wafer-level acceptable test data corresponding to the wafer within the historical time according to the wafer-level acceptable test data in the historical measurement data; wherein the image data of the online measurement values is used to represent online measurement values corresponding to all position points on the wafer, and the image data of the wafer-level acceptable test data is used to represent wafer-level acceptable test data corresponding to all position points on the wafer; inputting the image data of the online measurement values corresponding to the wafer within the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of the wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer; if it is determined that the output image satisfies a preset training completion condition according to the image data of the wafer-level acceptable test data corresponding to the wafer within the historical time, obtaining a prediction model of the wafer-level acceptable test data; wherein the prediction model of the wafer-level acceptable test data is used to predict wafer-level acceptable test data of any position point on the wafer.

2. The method of claim 1, wherein, The method comprises: acquiring historical measurement data corresponding to preset point positions on a wafer; wherein the historical measurement data comprises online measurement values obtained within a preset historical time and wafer-level acceptable test data, the online measurement values being used to represent size information of the wafer obtained in a manufacturing process of the wafer, and the wafer-level acceptable test data being used to represent data obtained by performing wafer-level acceptable test on the wafer; and the preset point positions being used to represent preset position points on the wafer; determining image data of the online measurement values corresponding to the wafer within the historical time according to the online measurement values in the historical measurement data, and determining image data of the wafer-level acceptable test data corresponding to the wafer within the historical time according to the wafer-level acceptable test data in the historical measurement data; wherein the image data of the online measurement values is used to represent online measurement values corresponding to all position points on the wafer, and the image data of the wafer-level acceptable test data is used to represent wafer-level acceptable test data corresponding to all position points on the wafer; 3. The method of claim 2, wherein, inputting the image data of the online measurement values corresponding to the wafer within the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of the wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer; if it is determined that the output image satisfies a preset training completion condition according to the image data of the wafer-level acceptable test data corresponding to the wafer within the historical time, obtaining a prediction model of the wafer-level acceptable test data; wherein the prediction model of the wafer-level acceptable test data is used to predict wafer-level acceptable test data of any position point on the wafer. The method comprises: acquiring historical measurement data corresponding to preset point positions on a wafer; wherein the historical measurement data comprises online measurement values obtained within a preset historical time and wafer-level acceptable test data, the online measurement values being used to represent size information of the wafer obtained in a manufacturing process of the wafer, and the wafer-level acceptable test data being used to represent data obtained by performing wafer-level acceptable test on the wafer; and the preset point positions being used to represent preset position points on the wafer; determining image data of the online measurement values corresponding to the wafer within the historical time according to the online measurement values in the historical measurement data, and determining image data of the wafer-level acceptable test data corresponding to the wafer within the historical time according to the wafer-level acceptable test data in the historical measurement data; wherein the image data of the online measurement values is used to represent online measurement values corresponding to all position points on the wafer, and the image data of the wafer-level acceptable test data is used to represent wafer-level acceptable test data corresponding to all position points on the wafer; inputting the image data of the online measurement values corresponding to the wafer within the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of the wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer; if it is determined that the output image satisfies a preset training completion condition according to the image data of the wafer-level acceptable test data corresponding to the wafer within the historical time, obtaining a prediction model of the wafer-level acceptable test data; wherein the prediction model of the wafer-level acceptable test data is used to predict wafer-level acceptable test data of any position point on the wafer. acquiring historical measurement data corresponding to preset point positions on a wafer, comprising: acquiring online measurement values corresponding to a preset first point position on the wafer obtained within a preset historical time as historical online measurement values, and acquiring wafer-level acceptable test data corresponding to a preset second point position on the wafer obtained within the preset historical time as historical wafer-level acceptable test data; wherein the first point position comprises at least one position point, and the second point position comprises at least one position point; preprocessing the historical online measurement values and the historical wafer-level acceptable test data to determine target point positions in the first point position and the second point position, and obtaining historical online measurement values and historical wafer-level acceptable test data corresponding to the target point positions. The preprocessing the historical online measurement values and the historical wafer-level acceptable test data to determine target point positions in the first point position and the second point position comprises: adding the first point position to a preset spatial coordinate system to obtain an initial image of the historical online measurement values; wherein a coordinate point of the first point position in the initial image of the historical online measurement values is used to represent a position of the first point position in the wafer, and a pixel value of the first point position is used to represent the historical online measurement values of the first point position; adding the second point to a preset spatial coordinate system to obtain an initial image of the historical wafer-level acceptable test data; wherein in the initial image of the historical wafer-level acceptable test data, a coordinate point of the second point is used to represent a position of the second point in the wafer, and a pixel value of the second point is used to represent historical wafer-level acceptable test data of the second point; overlapping the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data to determine a point that is overlapped in the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data as the target point.

4. The method of claim 2, wherein, The process of manufacturing the wafer includes at least two process passes. After obtaining the online measurement value corresponding to the preset first point on the wafer within a preset historical time, the method further includes: determining the historical online measurement value corresponding to each process pass within the preset historical time; determining the correlation between the historical online measurement values corresponding to the two process passes; if the correlation meets a preset rejection condition, retaining the historical online measurement value corresponding to one process pass from the historical online measurement values corresponding to the two process passes.

5. The method of claim 2, wherein, After obtaining the online measurement value corresponding to the preset first point on the wafer within a preset historical time, the method further includes: if the number of the first points is not a preset point number threshold, deleting the online measurement value corresponding to the preset first point on the wafer.

6. The method of claim 2, wherein, obtaining wafer-level acceptable test data corresponding to a preset second point on the wafer within a preset historical time as historical wafer-level acceptable test data, including: obtaining all wafer-level acceptable test data corresponding to the preset second point on the wafer within the preset historical time; selecting wafer-level acceptable test data of a preset data category from the all wafer-level acceptable test data as the historical wafer-level acceptable test data.

7. The method of claim 3, wherein, According to the online measurement value in the historical measurement data, determining image data of the online measurement value corresponding to the wafer within the historical time, including: According to the overlapping result of the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data, obtaining a target point image of the historical online measurement value; wherein the target point image of the historical online measurement value is used to represent a part of the initial image of the historical online measurement value that is overlapped with the initial image of the historical wafer-level acceptable test data; According to a preset spline interpolation algorithm, supplementing the target point image of the historical online measurement value to obtain the image data of the online measurement value corresponding to the wafer within the historical time.

8. The method of claim 3, wherein, According to the wafer-level acceptable test data in the historical measurement data, determining image data of the wafer-level acceptable test data corresponding to the wafer within the historical time, including: According to an overlapping result of the initial image of the historical online measurement value and the initial image of the historical wafer-level acceptable test data, a target point image of the historical wafer-level acceptable test data is obtained; wherein the target point image of the historical wafer-level acceptable test data is used to represent a part of the initial image of the historical wafer-level acceptable test data that overlaps with the initial image of the historical online measurement value; According to a preset spline interpolation algorithm, the target point image of the historical wafer-level acceptable test data is supplemented to obtain image data of wafer-level acceptable test data corresponding to the wafer in the historical time.

9. The method according to any one of claims 1-8, characterized in that, The image data of the online measurement value corresponding to the wafer in the historical time is input into a pre-constructed initial model for processing to obtain an output image, including: The image data of the online measurement value corresponding to the wafer in the historical time is input into a pre-constructed initial model; wherein the pre-constructed initial model is a U-shaped network structure; According to a preset deconvolution network layer in the initial model, the image data of the online measurement value corresponding to the wafer in the historical time is subjected to size expansion processing to obtain size-expanded image data; According to a preset convolution network layer in the initial model, the size-expanded image data is subjected to feature extraction to obtain the output image.

10. A method for predicting wafer-level acceptable test data, characterized in that, Including: Obtaining current online measurement value of a wafer; Inputting the current online measurement value of the wafer into a wafer-level acceptable test data prediction model to obtain prediction data of wafer-level acceptable test data corresponding to the current online measurement value of the wafer; wherein the wafer-level acceptable test data prediction model is a model obtained based on the method in any one of claims 1-9.

11. A model training device applied to wafer-level acceptable test, characterized in that, Including: A data acquisition module is configured to acquire historical measurement data corresponding to preset points on a wafer; wherein the historical measurement data includes online measurement values and wafer-level acceptable test data obtained in a preset historical time, the online measurement values are used to represent size information of the wafer obtained in a wafer manufacturing process, and the wafer-level acceptable test data is used to represent data obtained by performing wafer-level acceptable test on the wafer; the preset points are used to represent preset position points on the wafer; An image determination module is configured to determine image data of online measurement values corresponding to the wafer in the historical time according to the online measurement values in the historical measurement data, and determine image data of wafer-level acceptable test data corresponding to the wafer in the historical time according to the wafer-level acceptable test data in the historical measurement data; wherein the image data of online measurement values is used to represent online measurement values corresponding to all position points on the wafer, and the image data of wafer-level acceptable test data is used to represent wafer-level acceptable test data corresponding to all position points on the wafer; A model training module is configured to input image data of online measurement values corresponding to the wafer in the historical time into a pre-constructed initial model for processing to obtain an output image; wherein the output image is used to represent predicted data of wafer-level acceptable test data obtained by performing wafer-level acceptable test on all position points on the wafer. A training completion module is configured to obtain a prediction model of wafer-level acceptable test data if it is determined that the output image meets a preset training completion condition according to image data of wafer-level acceptable test data corresponding to the wafer in the historical time; wherein the prediction model of wafer-level acceptable test data is used to predict wafer-level acceptable test data of any position point on the wafer.

12. A wafer-level acceptable test data prediction apparatus, comprising: The method comprises the following steps: A measurement value acquisition module is configured to acquire current online measurement values of a wafer; wherein the online measurement values are used to represent size information of the wafer obtained in a manufacturing process of the wafer. A data prediction module is configured to input the current online measurement values of the wafer into a prediction model of wafer-level acceptable test data to obtain predicted data of wafer-level acceptable test data corresponding to the current online measurement values of the wafer; wherein the prediction model of wafer-level acceptable test data is a model obtained based on the device of claim 11.

13. An electronic device, comprising: The device comprises: A processor and a memory connected to the processor in communication; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the model training method for wafer-level acceptable test or the prediction method of wafer-level acceptable test data according to any one of claims 1-9.

14. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the model training method for wafer-level acceptable test or the prediction method of wafer-level acceptable test data according to any one of claims 1-9.

15. A computer program product, characterised in that, The computer program is executed by the processor to implement the model training method for wafer-level acceptable test or the prediction method of wafer-level acceptable test data according to any one of claims 1-9.

Images