Wafer data processing method and device, chip and computer readable storage medium

Abstract

The invention relates to a wafer data processing method and device, a chip and a computer readable storage medium. The method comprises the following steps: acquiring a test data set of a wafer, wherein the test data set comprises a test value in the wafer; partitioning the wafer according to at least one feature dimension to obtain a plurality of wafer partitions; determining characteristic parameters of each wafer partition according to the test data set and the wafer partitions; according to the characteristic parameters of each wafer partition, constructing a to-be-measured space vector of each wafer; and calculating the vector similarity of the wafers according to the to-be-measured space vector of each wafer, and determining the wafer similarity based on the vector similarity. According to the method, wafer identification and classification are carried out based on the wafer similarity, a convolutional neural network does not need to be used for execution, the requirement for computing power is reduced, and the data processing speed is increased.

Description

Technical Field

[0001] The present application relates to the field of semiconductor technology, and in particular to a wafer data processing method, device, chip and computer-readable storage medium. Background Art

[0002] Modern chips can only be obtained through a long semiconductor industry manufacturing process. For example, it is necessary to go through the process of silicon ingot-purification-cutting-grinding-polishing-lithography and etching-ion implantation and lamination-cutting separation to obtain wafer-testing-packaging-chip. Among them, after obtaining the wafer, the basic unit Die (particle) on it must be subjected to various types of tests (for example: current and voltage test, resistance test, etc.) to determine whether it is qualified. The analysis of test results is a very important part in the semiconductor industry. This importance is reflected in two aspects. On the one hand, Die is the basic unit of the chip. If a Die fails the test, it cannot be used to manufacture chips and can only be discarded; on the other hand, based on the analysis of the test results, the possible causes of the problems in the semiconductor process can be inferred. If the inferred cause is correct, it will help improve the process, thereby improving the yield of the semiconductor process and reducing the probability of defects in the wafer as much as possible.

[0003] Based on the above, it can be understood that data analysis of wafer test results is extremely important to the semiconductor industry, but data analysis of wafer test results is very difficult. On the one hand, there are many types of tests, and the types of data obtained are different; on the other hand, the amount of data obtained from the test is large, and it is more computationally intensive and time-consuming to perform data analysis.

[0004] One known prior art is to use deep learning methods to detect wafer test patterns, but deep learning methods are not suitable for this semiconductor business scenario. Since the semiconductor manufacturing process is very long and there are many types of tests, the "patterns / modes" generated are inexhaustible, and the model obtained by the deep learning method can only "recognize" the "patterns / modes" that have appeared in the training data, and cannot recognize the "patterns / modes" that are not in the training data, so the generalization of the deep learning solution is poor; finally, the image recognition algorithm using deep learning is generally executed using a convolutional neural network, and the operation of the convolutional neural network requires a lot of computing power, so the overall energy consumption and speed are not good enough. Summary of the invention

[0005] Based on this, it is necessary to provide a wafer data processing method, device, computer equipment, computer readable storage medium and computer program product to address the above technical problems.

[0006] In a first aspect, the present application provides a wafer data processing method, comprising:

[0007] Acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer;

[0008] Partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions;

[0009] Determine characteristic parameters of each wafer partition according to the test data set and the wafer partition to which each die belongs;

[0010] Constructing a space vector to be measured for each wafer according to the characteristic parameters of each wafer partition;

[0011] The vector similarity of the wafer is calculated according to the space vector to be measured of each wafer, and the wafer similarity is determined based on the vector similarity.

[0012] In one embodiment, the step of acquiring a test data set of a wafer, wherein the test data set includes test values ​​in the wafer, includes:

[0013] Acquiring test data of at least one electrical parameter of a plurality of wafers;

[0014] The test data of the at least one electrical parameter is normalized to obtain a test data set of each electrical parameter of the wafer.

[0015] In one of the embodiments, the feature dimension includes at least one of a direction feature, an electrical parameter type feature, a partition threshold feature, a process processing area feature, and a target pattern distribution feature.

[0016] In one embodiment, the directional feature includes at least one partition direction, and the wafer is partitioned according to at least one characteristic dimension to obtain a plurality of wafer partitions, including:

[0017] Based on at least one partitioning direction, separate partitioning or combined partitioning is performed to obtain a plurality of wafer partitions with the same area and / or the same number of dies.

[0018] In one embodiment, the at least one partitioning direction includes partitioning according to the radial and axial directions of the wafer.

[0019] In one embodiment, the wafer includes a target wafer and a wafer to be inspected, and the method further includes:

[0020] Calculating the vector similarity between the target wafer's spatial vector and the spatial vector to be tested of each wafer to be tested according to the target wafer's spatial vector;

[0021] Wafer similarity is determined based on the vector similarity, and multiple wafer similarities are sorted and output.

[0022] In one embodiment, the method further comprises:

[0023] Acquire corresponding production information according to the wafers to be inspected corresponding to the plurality of wafer similarities;

[0024] A fault analysis is performed on the wafer pattern of the wafer to be inspected according to the production information to determine the wafer production problem.

[0025] In one embodiment, determining the characteristic parameters of each wafer partition according to the test data set and the wafer partition includes:

[0026] According to the wafer partition to which the test value belongs, at least one test value corresponding to each wafer partition is obtained from the test data set; and statistical processing is performed on the at least one test value corresponding to each wafer partition to obtain characteristic parameters of each wafer partition.

[0027] In one embodiment, the statistical processing of at least one test value corresponding to each wafer partition to obtain a characteristic parameter of each wafer partition includes:

[0028] Determining a first standard deviation according to at least one test value corresponding to each of the wafer partitions;

[0029] A characteristic parameter of each of the wafer partitions is determined according to the first standard deviation.

[0030] In one embodiment, the statistical processing of at least one test value corresponding to each wafer partition to obtain a characteristic parameter of each wafer partition includes:

[0031] Determine a second standard deviation and an average value according to at least one test value corresponding to each of the wafer partitions;

[0032] The second standard deviation and the average value are weighted to obtain characteristic parameters of each wafer partition.

[0033] In one embodiment, the weighted processing of the second standard deviation and the average value to obtain the characteristic parameter of each wafer partition includes:

[0034] Determine a first weighting coefficient corresponding to the second standard deviation and a second weighting coefficient corresponding to the average value; the sum of the first weighting coefficient and the second weighting coefficient is a preset value;

[0035] Weighted processing is performed according to the first weighting coefficient, the second weighting coefficient, the second standard deviation and the average value to obtain characteristic parameters of each wafer partition.

[0036] In one embodiment, the calculating the vector similarity of the wafer according to the measured space vector of each wafer, and determining the wafer similarity based on the vector similarity, comprises:

[0037] According to the space vector to be measured of each of the wafers, similarity calculation is performed on any two wafers among the multiple wafers according to a preset similarity calculation method to obtain multiple vector similarities of the any two wafers; wherein the preset similarity calculation method includes at least any one of Euler distance, Pearson correlation coefficient and cosine similarity;

[0038] Wafer similarity is determined based on the vector similarity.

[0039] In a second aspect, the present application also provides a wafer data processing device, comprising:

[0040] A data acquisition module, used to acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer;

[0041] A partitioning module, used for partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions;

[0042] A vector determination module, used to determine the characteristic parameters of each wafer partition according to the test data set and the wafer partition; and to construct a space vector to be tested for each wafer according to the characteristic parameters of each wafer partition;

[0043] The similarity determination module is used to calculate the vector similarity of the wafer according to the space vector to be measured of each wafer, and determine the wafer similarity based on the vector similarity.

[0044] In a third aspect, the present application further provides a chip, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0045] Acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer;

[0046] Partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions;

[0047] Determining characteristic parameters of each wafer partition according to the test data set and the wafer partition;

[0048] Constructing a space vector to be measured for each wafer according to the characteristic parameters of each wafer partition;

[0049] The vector similarity of the wafer is calculated according to the space vector to be measured of each wafer, and the wafer similarity is determined based on the vector similarity.

[0050] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0051] Acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer;

[0052] Partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions;

[0053] Determining characteristic parameters of each wafer partition according to the test data set and the wafer partition;

[0054] Constructing a space vector to be measured for each wafer according to the characteristic parameters of each wafer partition;

[0055] The vector similarity of the wafer is calculated according to the space vector to be measured of each wafer, and the wafer similarity is determined based on the vector similarity.

[0056] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0057] Acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer;

[0058] Partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions;

[0059] Determining characteristic parameters of each wafer partition according to the test data set and the wafer partition;

[0060] Constructing a space vector to be measured for each wafer according to the characteristic parameters of each wafer partition;

[0061] The vector similarity of the wafer is calculated according to the space vector to be measured of each wafer, and the wafer similarity is determined based on the vector similarity.

[0062] The above-mentioned wafer data processing method, device, chip, computer-readable storage medium and computer program product obtain a test data set of the wafer, the test data set contains test values ​​in the wafer; partition the wafer according to at least one characteristic dimension to obtain multiple wafer partitions; determine the characteristic parameters of each wafer partition according to the test data set and the wafer partition; construct a space vector to be tested for each wafer according to the characteristic parameters of each wafer partition; calculate the vector similarity of the wafer according to the space vector to be tested for each wafer, and determine the wafer similarity based on the vector similarity. By partitioning and constructing the space vector to be tested according to at least one characteristic dimension, more features are extracted from the wafer, and the feature expression of the wafer is enhanced, which can provide more refined similarity and improve the accuracy of wafer data processing. For any type of wafer pattern, wafer identification and classification can be performed based on wafer similarity, and there is no need to use a convolutional neural network to execute, which reduces the demand for computing power and improves the data processing speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 1 is a schematic flow chart of a wafer data processing method in one embodiment;

[0065] Figure 2 A schematic flow chart of a method for determining characteristic parameters of wafer partitions in one embodiment;

[0066] Figure 3 A schematic diagram of partitioning of a wafer to be inspected in one embodiment;

[0067] Figure 4 is a structural block diagram of a wafer data processing device in one embodiment;

[0068] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0070] In the description of the embodiments of the present application, “plurality” means more than two, unless otherwise clearly and specifically defined.

[0071] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0072] As the core component of modern electronic devices, chips are widely used in a wide range of scenarios. Chips are not only used in computers and servers, mobile devices, embedded systems and electronic products, but also in navigation systems, radars, communication equipment, etc.

[0073] Chips need to go through a long semiconductor industry manufacturing process to be obtained. For example, it is necessary to go through the process of silicon ingot-purification-cutting-grinding-polishing-lithography and etching-ion implantation and coating-cutting separation to obtain wafers-testing-packaging-chips. Among them, after obtaining the wafer, the basic unit Die (particle) on it must be subjected to various types of tests, such as current and voltage tests, resistance tests, etc., to determine whether it is qualified. The analysis of test results is a very important part in the semiconductor industry. This importance is reflected in two aspects. On the one hand, Die is the basic unit of the chip. If a Die test fails, it will not be used to manufacture chips, but can only be discarded; on the other hand, based on the analysis of the test results, the possible causes of the problems in the semiconductor process can be inferred. If the inferred cause is correct, it will help improve the process, thereby improving the yield of the semiconductor process and reducing the probability of defects in the wafer as much as possible. At present, with the improvement of semiconductor lithography accuracy, it is becoming increasingly difficult to improve product yield. In order to ensure the product yield, it is necessary to identify the faults in the wafer for analysis. Therefore, the wafer needs to be tested.

[0074] The existing method is to use deep learning methods to detect wafer test patterns. The model obtained by the deep learning method can only "recognize" the "patterns / modes" that have appeared in the training data, and cannot recognize the "patterns / modes" that are not in the training data. Therefore, the generalization of deep learning solutions is poor, and image recognition algorithms that require deep learning are generally executed using convolutional neural networks. The operation of convolutional neural networks requires a lot of computing power, so the overall energy consumption is high and the speed is slow.

[0075] Therefore, a wafer data processing method is proposed, which comprises the following steps: obtaining a test data set of the wafer, wherein the test data set includes test values ​​in the wafer; partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions; determining characteristic parameters of each wafer partition according to the test data set and the wafer partitions; constructing a space vector to be tested for each wafer according to the characteristic parameters of each wafer partition; calculating the vector similarity of the wafer according to the space vector to be tested for each wafer, and determining the wafer similarity based on the vector similarity.

[0076] In an exemplary embodiment, Figure 1 As shown, a wafer data processing method is provided. This embodiment uses the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes steps 102 to 110, wherein:

[0077] Step 102 , obtaining a test data set of a wafer, where the test data set includes test values ​​in the wafer.

[0078] Among them, the test value can be understood as obtained during the wafer test process, such as an electrical test value. The wafer test can be, but is not limited to, at least one electrical test of resistance test, voltage test, and current test. The electrical test value can be, but is not limited to, at least one of voltage, current, and resistance. The test data in the test data set can be normalized data or directly obtained raw test data. This example illustrates that the test data in the test data set can be normalized data.

[0079] Normalization is to normalize the data to the same preset numerical range. It is understandable that different data features correspond to different normalization methods. When the data feature of the test data is that the numerical range of the test value is greater than the preset numerical range, the data is normalized by using percentile values ​​to determine the range of data scaling. When the data feature of the test data is that the number of identical values ​​in the test value is greater than the preset number, the ranking of each data value can be divided by the maximum value of all rankings to achieve normalization.

[0080] Step 104 , partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions.

[0081] Among them, the feature dimensions include but are not limited to at least one of directional features, electrical parameter type features, partition threshold features, process processing area features, and target pattern distribution features.

[0082] Exemplarily, according to actual needs, the wafer is divided into N zones according to a specified partitioning direction to obtain N wafer partitions.

[0083] Step 106 , determining characteristic parameters of each wafer partition according to the test data set and the wafer partition.

[0084] Among them, the test values ​​in the test data set can determine the wafer partition to which they belong based on information such as coordinates. In the case where the test data set is a normalized parameter set obtained after normalization processing, the characteristic parameters can be determined based on all normalized parameters of each wafer partition. Optionally, the normalized parameters of each wafer partition are statistically processed to obtain the characteristic parameters of each wafer partition. The statistical processing can be various statistical algorithms such as averaging, median, maximum value, minimum value, and any combination thereof.

[0085] When the test data set is the original test data, it is necessary to determine all the test data (test values) of the wafer partition from the test data set, normalize all the test data of each wafer partition to obtain normalized parameters, and then determine the characteristic parameters based on all the normalized parameters of each wafer partition.

[0086] Exemplarily, all test data of each wafer partition are determined from the test data set, all test data of each wafer partition are normalized to obtain corresponding normalized parameters, the normalized parameters of each wafer partition are statistically processed to obtain characteristic parameters of each wafer partition, and a space vector to be tested is constructed based on the characteristic parameters to ensure the accuracy of the similarity calculation.

[0087] Step 108 , constructing a space vector to be measured for each wafer according to the characteristic parameters of each wafer partition.

[0088] At this point, the data of each parameter on each wafer has been converted into an array (vector). The length of the array is the number of wafer partitions, and each value in the array represents the statistical value of the corresponding wafer partition. Based on the above array, the vector space distance between each set of data (i.e. different wafers and different parameters) is calculated to represent their similarity. The smaller the spatial distance, the higher the similarity of the two sets of data, that is, the more similar the patterns / modes of the two images are.

[0089] Step 110 , calculating the vector similarity of the wafers according to the space vector to be measured of each wafer, and determining the wafer similarity based on the vector similarity.

[0090] The vector similarity may be determined based on the vector space distance of the wafer. The vector space distance may be determined by any one of Euler distance, Pearson correlation coefficient, and cosine similarity.

[0091] Exemplarily, according to the space vector to be measured of each wafer, that is, the array of each wafer, the vector space distance between the wafers is calculated to obtain the vector similarity of the wafers. According to the vector similarity of the wafers, the similarity of the wafers is determined, and similar wafers can be determined from multiple wafers according to the wafer similarity. Among them, the method of determining similar wafers can be to classify multiple wafers according to the vector similarity of the wafers to obtain multiple wafer groups, and randomly select a preset number of similar wafers from each group of wafers to determine, or according to the vector similarity of the wafers, by sorting the wafer similarity, obtaining wafers corresponding to the preset number of wafer similarities arranged in order, and determining the wafer as a similar wafer.

[0092] In the above-mentioned wafer data processing method, a test data set of the wafer is obtained, and the test data set includes test values ​​in the wafer; the wafer is partitioned according to at least one characteristic dimension to obtain multiple wafer partitions; the characteristic parameters of each wafer partition are determined according to the test data set and the wafer partition; the space vector to be tested of each wafer is constructed according to the characteristic parameters of each wafer partition; the vector similarity of the wafer is calculated according to the space vector to be tested of each wafer, and the wafer similarity is determined based on the vector similarity. By partitioning and constructing the space vector to be tested according to at least one characteristic dimension, more features are extracted from the wafer, and the feature expression of the wafer is enhanced, so that more refined similarity can be provided, and the accuracy of wafer data processing is improved. For any type of wafer pattern, wafer identification and classification can be performed based on wafer similarity, and there is no need to use convolutional neural networks to execute, which reduces the demand for computing power and improves data processing speed.

[0093] In an exemplary embodiment, a method for obtaining a test data set of a wafer is provided, including: obtaining test data of at least one electrical parameter of multiple wafers; normalizing the test data of at least one electrical parameter to obtain a test data set for each electrical parameter of the wafer.

[0094] When the test data involves more than one electrical parameter, or involves test values ​​of more than one wafer manufacturing link, the value range and order of magnitude between different test data sets will be very different. Therefore, different normalization methods are selected for data unification based on the data characteristics and data types of the wafer test data sets.

[0095] Normalization is to normalize the data to the same preset numerical range. It is understandable that different data features correspond to different normalization methods. When the data feature of the test data is that the numerical range of the electrical test value is greater than the preset numerical range, the data is normalized by using percentile values ​​to determine the range of data scaling. When the data feature of the test data is that the number of identical values ​​in the electrical test value is greater than the preset number, normalization can be achieved by dividing the ranking of each data value by the maximum value of all rankings.

[0096] Exemplarily, test data of at least one electrical parameter of each wafer among a plurality of wafers is obtained, and the test data of at least one electrical parameter is normalized to obtain a test data set of each electrical parameter of the wafer.

[0097] In an exemplary embodiment, the feature dimension includes but is not limited to at least one of a direction feature, an electrical parameter type feature, a partition threshold feature, a process processing area feature, and a target pattern distribution feature. The feature dimension of the partition can be determined based on different business scenarios to meet the needs of different scenarios.

[0098] The electrical parameter type feature can be the known electrical parameters to be analyzed, and the location where the known electrical parameters are prone to failure. The partition threshold feature can be to divide the wafer into N zones according to the specified partition direction according to actual needs. The process processing area feature can be understood as the area of ​​each processing in wafer manufacturing, and the partition is based on the size of the area processed each time. The target pattern distribution feature partition can be understood as a wafer image of a known wafer, and the wafer is partitioned according to the location of the wafer area where the wafer pattern is located and the size of the area. It can be understood that the partition direction and number of partitions for each wafer can be the same.

[0099] In an exemplary embodiment, a wafer is partitioned according to at least one characteristic dimension to obtain multiple wafer partitions, including: performing individual partitioning or combined partitioning based on at least one partitioning direction to obtain multiple wafer partitions with the same area and / or equal number of dies.

[0100] Among them, at least one partitioning direction includes partitioning according to the radial and / or axial direction of the wafer, and the axial direction can be understood as the direction along the axis of the wafer. The radial direction can be understood as the direction along the radius based on the center of the wafer. The number of partitions can be determined according to actual needs. It can be understood that the accuracy of the similarity is related to the number of partitions and the partition area. The number of partitions can be determined based on the physical size of the wafer and the preset segmentation area, and the number of partitions can also be determined based on the test value density on the wafer.

[0101] It should be noted that the number of partitions in each partition direction can be determined randomly, according to the wafer test value density, or according to the number and area of ​​the basic wafer units in the partition. Figure 3 FIG. 1 is a schematic diagram of partitioning of a wafer to be inspected in one embodiment. Figure 3The wafer partition shown is partitioned based on a combination of axial and radial partitions. The number of axial partitions is 4 and the number of radial partitions is 8, that is, the wafer data will be converted into a 4*8=32-dimensional vector, so that more accurate results can be obtained in subsequent similarity calculations. The wafer is divided into as many numbers as possible based on radial and axial partitions, thereby increasing the dimension of the vector space to obtain more accurate results.

[0102] Optionally, in an exemplary embodiment, the wafer includes a target wafer and a wafer to be detected, and the method further includes: calculating the vector similarity of the target wafer with the space vector to be detected of each wafer to be detected according to the space vector of the target wafer; determining the wafer similarity based on the vector similarity, and sorting and outputting the similarities of multiple wafers. The similarity calculation can quickly identify the wafer that is most similar or least similar to the target wafer, which is helpful for further detection or analysis in a targeted manner and effectively controlling product quality.

[0103] After determining the wafer similarity, problems in the wafer process can be analyzed based on the wafer similarity. In an exemplary embodiment, corresponding production information is obtained based on wafers to be inspected corresponding to multiple wafer similarities, and fault analysis is performed on wafer patterns of the wafers to be inspected based on the production information to determine wafer production problems.

[0104] The production information includes production information of each link of wafer production, such as process parameters and equipment parameters of production equipment.

[0105] By performing fault analysis on the wafer pattern of the target wafer according to the production information, a fault resolution strategy is obtained, and problems that may arise in the process are resolved based on the fault resolution strategy, thereby effectively improving the yield of semiconductor manufacturing.

[0106] The following provides a method for determining characteristic parameters of wafer partitions, such as Figure 2 As shown, in an exemplary embodiment, determining characteristic parameters of each wafer partition according to the test data set and the wafer partition includes the following steps:

[0107] Step 202 : acquiring at least one test value corresponding to each wafer partition from a test data set according to the wafer partition of each test value.

[0108] Step 204 , statistically process at least one test value corresponding to each wafer partition to obtain characteristic parameters of each wafer partition.

[0109] It is understandable that the way of re-statisticing the values ​​in each partition, including taking the average, median, maximum, minimum and other statistical algorithms, and using their combination can improve the accuracy of the data. In order to further improve the accuracy on this basis, two methods for determining characteristic parameters are proposed, including:

[0110] Method 1: Determine a first standard deviation according to at least one test value corresponding to each wafer partition; and determine a characteristic parameter of each wafer partition according to the first standard deviation.

[0111] Wherein, determining the characteristic parameter of each wafer partition according to the first standard deviation may be taking the first standard deviation as the characteristic parameter of each wafer partition. The first standard deviation may be determined by an existing method, which will not be described in detail here.

[0112] Method 2: Determine the second standard deviation and the average value according to at least one test value corresponding to each wafer partition; perform weighted processing on the second standard deviation and the average value to obtain characteristic parameters of each wafer partition.

[0113] Among them, the weighted processing can be: determining a first weighting coefficient corresponding to the second standard deviation, and a second weighting coefficient corresponding to the average value; the sum of the first weighting coefficient and the second weighting coefficient is a preset value; weighted processing is performed according to the first weighting coefficient, the second weighting coefficient, the second standard deviation and the average value to obtain the characteristic parameters of each wafer partition.

[0114] For example, the first weighting coefficient a*the second standard deviation STD+the second weighting coefficient b*the average value, a+b=1, a is between 0.2-0.7, and b is between 0.5-0.8. Further, in an exemplary embodiment, a=b=0.5, and the result obtained by weighting with STD+average value is the best, which greatly improves the accuracy and reliability of similarity calculation.

[0115] Optionally, in an exemplary embodiment, calculating the vector similarity of the wafers according to the measured space vector of each wafer, and determining the wafer similarity based on the vector similarity includes:

[0116] According to the spatial vector to be tested of each wafer, similarity calculation is performed on any two wafers among the multiple wafers according to a preset similarity calculation method to obtain vector similarities of multiple arbitrary two wafers; wherein the preset similarity calculation method includes at least any one of Euler distance, Pearson correlation coefficient and cosine similarity; wafer similarity is determined based on vector similarity.

[0117] Furthermore, according to the space vector of each wafer, the Euler formula is used to calculate the similarity of any two wafers in the multiple wafers to obtain the Euler distance, and then the Euler distance is weighted to obtain the vector space distance between any two wafers. Among them, the weighted calculation can be implemented by the existing method, and the weighted calculation fully considers the electrical performance parameters of the wafer, ensuring the reliability of the vector similarity.

[0118] After determining the wafer similarity, in order to intuitively obtain the defect differences between the wafers, in an exemplary embodiment, similar wafers are displayed in response to a wafer pattern display instruction.

[0119] In an exemplary embodiment, test data of at least one electrical parameter of multiple wafers are obtained; the test data of at least one electrical parameter are normalized to obtain a test data set for each electrical parameter of the wafer, the wafer is partitioned according to radial and axial directions of the wafer to obtain multiple wafer partitions, and characteristic parameters of each wafer partition are determined based on the test data set and the wafer partitions; a space vector to be tested of each wafer is constructed based on the characteristic parameters of each wafer partition; vector similarity of the wafers is calculated based on the space vector to be tested of each wafer, and wafer similarity is determined based on the vector similarity.

[0120] Furthermore, the wafer includes a target wafer and a wafer to be detected. According to the space vector of the target wafer, the vector similarity with the space vector to be detected of each wafer to be detected is calculated; the wafer similarity is determined based on the vector similarity; and the multiple wafer similarities are sorted and output. And according to the wafers to be detected corresponding to the multiple wafer similarities, the corresponding production information is obtained; according to the production information, the wafer pattern of the wafer to be detected is analyzed for faults to determine the wafer production problem.

[0121] In the above example, a mixed axial + radial partitioning scheme is adopted for the wafer. Under the premise of ensuring that the number of each partition is the same, as many partitions as possible are obtained, so that in the subsequent calculation of vector space similarity, the dimension of the vector space is increased, thereby achieving higher precision in the calculation of vector space similarity and improving the accuracy of wafer data processing.

[0122] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0123] Based on the same inventive concept, the embodiment of the present application also provides a wafer data processing device for implementing the wafer data processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more wafer data processing device embodiments provided below can refer to the limitations of the wafer data processing method above, and will not be repeated here.

[0124] In an exemplary embodiment, Figure 4 As shown, a wafer data processing device is provided, including: a data acquisition module 402, a partitioning module 404, a vector determination module 406 and a similarity determination module 408, wherein:

[0125] The data acquisition module 402 is used to acquire a test data set of a wafer, where the test data set includes test values ​​in the wafer.

[0126] The partitioning module 404 is used to partition the wafer according to at least one characteristic dimension to obtain multiple wafer partitions.

[0127] The vector determination module 406 is used to determine the characteristic parameters of each wafer partition according to the test data set and the wafer partition; and to construct the space vector to be tested of each wafer according to the characteristic parameters of each wafer partition.

[0128] The similarity determination module 408 is used to calculate the vector similarity of the wafer according to the space vector to be measured of each wafer, and determine the wafer similarity based on the vector similarity.

[0129] The above-mentioned wafer data processing device obtains a test data set of the wafer, and the test data set includes test values ​​in the wafer; partitions the wafer according to at least one characteristic dimension to obtain multiple wafer partitions; determines the characteristic parameters of each wafer partition according to the test data set and the wafer partition; constructs a space vector to be tested for each wafer according to the characteristic parameters of each wafer partition; calculates the vector similarity of the wafer according to the space vector to be tested for each wafer, and determines the wafer similarity based on the vector similarity. By partitioning and constructing the space vector to be tested according to at least one characteristic dimension, more features are extracted from the wafer, and the feature expression of the wafer is enhanced, so that more refined similarity can be provided, and the accuracy of wafer data processing is improved. For any type of wafer pattern, wafer identification and classification can be performed based on wafer similarity, and generalization is good, and there is no need to use a convolutional neural network to execute, which reduces the demand for computing power and improves the data processing speed.

[0130] In an exemplary embodiment, the data acquisition module 402 is further used to acquire test data of at least one electrical parameter of a plurality of wafers;

[0131] The test data of at least one electrical parameter is normalized to obtain a test data set of each electrical parameter of the wafer.

[0132] In an exemplary embodiment, the feature dimension includes at least one of a direction feature, an electrical parameter type feature, a partition threshold feature, a process processing area feature, and a target pattern distribution feature.

[0133] In an exemplary embodiment, the partitioning module 404 is further configured to perform separate partitioning or combined partitioning based on at least one partitioning direction to obtain a plurality of wafer partitions with the same area and / or the same number of dies.

[0134] In an exemplary embodiment, at least one partitioning direction includes partitioning according to a radial direction and an axial direction of the wafer.

[0135] In an exemplary embodiment, the wafer data processing device further includes an output module, which is used to calculate the vector similarity between the target wafer's spatial vector and the spatial vector to be tested of each wafer to be tested, and to sort and output multiple vector similarities.

[0136] In an exemplary embodiment, the above-mentioned wafer data processing device also includes a fault analysis module, which is used to obtain corresponding production information based on the wafers to be inspected corresponding to multiple vector similarities; perform fault analysis on the wafer pattern of the wafer to be inspected based on the production information to determine the wafer production problem.

[0137] In an exemplary embodiment, the vector determination module 406 is also used to obtain at least one test value corresponding to each wafer partition from the test data set according to the wafer partition of each test value; and perform statistical processing on the at least one test value corresponding to each wafer partition to obtain characteristic parameters of each wafer partition.

[0138] In an exemplary embodiment, the wafer data processing device further includes a statistical processing module, and the statistical processing module is used to determine a first standard deviation according to at least one test value corresponding to each wafer partition.

[0139] In an exemplary embodiment, the vector determination module 406 is further configured to determine a characteristic parameter of each wafer partition according to the first standard deviation.

[0140] In an exemplary embodiment, the statistical processing module is used to determine a second standard deviation and an average value according to at least one test value corresponding to each wafer partition.

[0141] In an exemplary embodiment, the vector determination module 406 is further configured to perform weighted processing on the second standard deviation and the average value to obtain characteristic parameters of each wafer partition.

[0142] In an exemplary embodiment, the vector determination module 406 is further used to determine a first weighting coefficient corresponding to the second standard deviation and a second weighting coefficient corresponding to the average value; the sum of the first weighting coefficient and the second weighting coefficient is a preset value;

[0143] Weighted processing is performed according to the first weighting coefficient, the second weighting coefficient, the second standard deviation and the average value to obtain characteristic parameters of each wafer partition.

[0144] In an exemplary embodiment, the similarity determination module 408 is further used to perform similarity calculation on any two wafers among the plurality of wafers according to the measured spatial vector of each wafer according to a preset similarity calculation method, so as to obtain vector similarities of the plurality of arbitrary two wafers; wherein the preset similarity calculation method includes at least any one of Euler distance, Pearson correlation coefficient and cosine similarity;

[0145] Wafer similarity is determined based on vector similarity.

[0146] Each module in the above-mentioned wafer data processing device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0147] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (Near Field Communication, NFC) or other technologies. When the computer program is executed by the processor, a wafer data processing method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0148] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0149] In one embodiment, a chip is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0150] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0151] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0152] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0153] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0154] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0155] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A wafer data processing method, characterized in that: The method comprises: Acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer; Partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions; Determining characteristic parameters of each wafer partition according to the test data set and the wafer partition; Constructing a space vector to be measured for each wafer according to the characteristic parameters of each wafer partition; The vector similarity of the wafer is calculated according to the space vector to be measured of each wafer, and the wafer similarity is determined based on the vector similarity.

2. The method according to claim 1, characterized in that The step of acquiring a test data set of a wafer, wherein the test data set includes test values ​​in the wafer, includes: Acquiring test data of at least one electrical parameter of a plurality of wafers; The test data of the at least one electrical parameter is normalized to obtain a test data set of each electrical parameter of the wafer.

3. The method according to claim 2, characterized in that The feature dimension includes at least one of a direction feature, an electrical parameter type feature, a partition threshold feature, a process processing area feature, and a target pattern distribution feature.

4. The method according to claim 3, characterized in that Partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions includes: Based on the at least one partitioning direction, separate partitioning or combined partitioning is performed to obtain a plurality of wafer partitions with the same area and / or the same number of dies.

5. The method according to claim 4, characterized in that The at least one partitioning direction includes partitioning according to the radial direction and the axial direction of the wafer.

6. The method according to claim 1, characterized in that The wafer includes a target wafer and a wafer to be detected, and the method further includes: Calculating the vector similarity between the target wafer's spatial vector and the spatial vector to be tested of each wafer to be tested according to the target wafer's spatial vector; Wafer similarity is determined based on the vector similarity, and multiple wafer similarities are sorted and output.

7. The method according to claim 6, characterized in that The method further comprises: Acquire corresponding production information according to the wafers to be inspected corresponding to the plurality of wafer similarities; A fault analysis is performed on the wafer pattern of the wafer to be inspected according to the production information to determine the wafer production problem.

8. The method according to claim 1, characterized in that Determining characteristic parameters of each wafer partition according to the test data set and the wafer partition includes: According to the wafer partition to which each test value belongs, at least one test value corresponding to each wafer partition is obtained from the test data set; and statistical processing is performed on the at least one test value corresponding to each wafer partition to obtain characteristic parameters of each wafer partition.

9. The method according to claim 8, characterized in that The statistical processing of at least one electrical property test value corresponding to each wafer partition to obtain characteristic parameters of each wafer partition includes: Determining a first standard deviation according to at least one test value corresponding to each of the wafer partitions; A characteristic parameter of each of the wafer partitions is determined according to the first standard deviation.

10. The method according to claim 8, characterized in that The statistical processing of at least one test value corresponding to each of the wafer partitions to obtain characteristic parameters of each of the wafer partitions includes: Determine a second standard deviation and an average value according to at least one test value corresponding to each of the wafer partitions; The second standard deviation and the average value are weighted to obtain characteristic parameters of each wafer partition.

11. The method according to claim 10, characterized in that The step of performing weighted processing on the second standard deviation and the average value to obtain characteristic parameters of each wafer partition includes: Determine a first weighting coefficient corresponding to the second standard deviation and a second weighting coefficient corresponding to the average value; the sum of the first weighting coefficient and the second weighting coefficient is a preset value; Weighted processing is performed according to the first weighting coefficient, the second weighting coefficient, the second standard deviation and the average value to obtain characteristic parameters of each wafer partition.

12. The method according to claim 1, characterized in that The step of calculating the vector similarity of the wafer according to the measured spatial vector of each wafer, and determining the wafer similarity based on the vector similarity includes: According to the space vector to be measured of each of the wafers, similarity calculation is performed on any two wafers among the multiple wafers according to a preset similarity calculation method to obtain multiple vector similarities of the any two wafers; wherein the preset similarity calculation method includes at least any one of Euler distance, Pearson correlation coefficient and cosine similarity; Wafer similarity is determined based on the vector similarity.

13. A wafer data processing device, characterized in that: The device comprises: A data acquisition module, used to acquire a test data set of a wafer, wherein the test data set includes test values ​​in the wafer; A partitioning module, used for partitioning the wafer according to at least one characteristic dimension to obtain a plurality of wafer partitions; A vector determination module, used to determine the characteristic parameters of each wafer partition according to the test data set and the wafer partition; and to construct a space vector to be tested for each wafer according to the characteristic parameters of each wafer partition; The similarity determination module is used to calculate the vector similarity of the wafer according to the space vector to be measured of each wafer, and determine the wafer similarity based on the vector similarity.

14. A chip comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 12 are implemented.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

16. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

Images