Sensor data time sequence similarity analysis method and device in wafer processing process

Abstract

The invention provides a time sequence similarity analysis method and device for sensor data in a wafer processing process, and the method comprises the steps: obtaining time sequence data collected by a sensor corresponding to a target parameter in the wafer processing process; based on the time sequence data, respectively determining a reference sequence and at least one comparison sequence; respectively carrying out dimensionless and normalized data preprocessing on the reference sequence and the at least one comparison sequence; for each comparison sequence, calculating a correlation coefficient of each data sampling point at the same position between the reference sequence and the comparison sequence based on a grey correlation analysis algorithm; and based on the correlation coefficient of each data sampling point, obtaining a correlation degree for quantitatively representing the correlation between the reference sequence and each comparison sequence. According to the method, the FDC system and the grey correlation analysis algorithm are combined, time sequence analysis can be automatically carried out on the time sequence data collected by the sensor, and the analysis efficiency and quality are improved.

Description

Technical Field

[0001] This application relates to the field of semiconductor technology, and particularly to a method and device for analyzing the temporal similarity of sensor data in the wafer processing process. Background Art

[0002] In the wafer processing process, sensors of semiconductor machines play a crucial role. They can capture and record a large amount of process data in real time. These data are then processed, filtered, and displayed through various logics in a Fault Detection and Classification (FDC) system for engineers to analyze.

[0003] When a certain sensor parameter deviates from the set process range, the FDC system will issue a warning. After receiving the warning, engineers analyze the reason for the parameter data. The existing method is to display the relevant parameters within the same time period in the form of tables and charts, and draw conclusions by observation. For example, if a certain parameter is particularly important, engineers trace its historical data through the FDC system and observe and compare it visually to understand which other parameters are highly correlated with this parameter within a specific time period. However, in the highly dynamic environment of wafer processing, many parameters are constantly changing over time, and the applicability of static data analysis methods is limited. Dynamic data can only be analyzed based on the experience of engineers. Therefore, the existing data analysis method highly depends on the personal observation and judgment of engineers, and the entire analysis process is inefficient, resulting in uncertainty and inefficiency of the analysis results. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, this application provides a method and device for analyzing the temporal similarity of sensor data in the wafer processing process, so as to at least solve the problem of low efficiency in performing temporal correlation analysis on sensor data in the existing technology.

[0005] To achieve the above object and other advantages, this application is implemented by adopting the following technical solutions: In a first aspect, this application provides a method for analyzing the temporal similarity of sensor data in the wafer processing process, including: Obtain the temporal data collected by the sensor corresponding to the target parameter in the wafer processing process; Based on the temporal data, determine a reference sequence and at least one comparison sequence respectively; Perform dimensionless and normalized data preprocessing on the reference sequence and the at least one comparison sequence respectively; For each of the comparison sequences, calculate the correlation coefficients of each data sampling point at the same position between the reference sequence and the comparison sequence based on the grey relational analysis algorithm; Based on the correlation coefficients of each data sampling point, an association degree is obtained for quantitatively representing the association size between the reference sequence and each comparison sequence.

[0006] According to a method for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application, the step of respectively determining a reference sequence and at least one comparison sequence based on the temporal data includes: Select a representative data sequence from the temporal data as the reference sequence; Regard the remaining temporal data after removing the reference sequence as a data sequence for intercepting comparison sequences that match the length of the reference sequence; According to the set window size and step size, move the sliding window step by step on the data sequence. After each move, the data segment within the sliding window will be updated to form a new comparison sequence.

[0007] According to a method for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application, the step of calculating the correlation coefficients of each data sampling point at the same position between the reference sequence and the comparison sequence based on the grey relational analysis algorithm for each comparison sequence includes: Convert the preprocessed reference sequence and the at least one comparison sequence into a first slope sequence and at least one second slope sequence respectively; For each second slope sequence, calculate the correlation coefficients of each data sampling point at the same position between it and the first slope sequence.

[0008] According to a method for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application, the step of converting the preprocessed reference sequence and the at least one comparison sequence into a first slope sequence and at least one second slope sequence respectively includes: Set the slope value of the first data sampling point of the first slope sequence and the at least one second slope sequence to 0; For the second and subsequent data sampling points in the first slope sequence and the at least one second slope sequence, subtract the value of the previous data sampling point from the value of the data sampling point as the slope value of the data sampling point.

[0009] According to a method for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application, the step of calculating the correlation coefficients of each data sampling point at the same position between each second slope sequence and the first slope sequence includes: Based on the first slope sequence and the at least one second slope sequence, obtain the minimum value and the maximum value of the absolute differences of slopes between all data sampling points among different sequences; For each data sampling point of each of the second slope sequences, calculate the absolute difference of the slope value from the data sampling point at the same position in the first slope sequence; Based on the minimum value and the maximum value of the absolute differences of slopes, the absolute differences, and a preset resolution coefficient, determine the correlation coefficient between each of the data sampling points at the same position between the first slope sequence and the second slope sequence.

[0010] According to a method for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application, the step of obtaining the minimum value and the maximum value of the absolute differences of slopes between all data sampling points among different sequences based on the first slope sequence and the at least one second slope sequence includes: For each of the second slope sequences, calculate the first absolute difference of the slope value between each data sampling point of the second slope sequence and the data sampling point at the corresponding position in the first slope sequence, and determine a first minimum absolute difference of slopes and a first maximum absolute difference of slopes from among the multiple first absolute differences of slopes; When there are multiple second slope sequences, obtain multiple first minimum absolute differences of slopes and multiple first maximum absolute differences of slopes; Determine a second minimum absolute difference of slopes from among the multiple first minimum absolute differences of slopes, and determine a second maximum absolute difference of slopes from among the multiple first maximum absolute differences of slopes, to respectively serve as the minimum value and the maximum value of the absolute differences of slopes.

[0011] According to a method for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application, the step of obtaining, based on the correlation coefficients of each data sampling point, a correlation degree for quantitatively representing the degree of association between the reference sequence and each comparison sequence includes: For each of the comparison sequences, take the average of the correlation coefficients of all data sampling points to obtain a quantitative representation of the degree of association with the reference sequence, that is, the correlation degree.

[0012] In a second aspect, the present application provides an electronic device, which includes: One or more processors; and a memory storing computer program instructions, where the computer program instructions, when executed, cause the processor to execute the method for analyzing the temporal similarity of sensor data in a wafer processing process as described in any one of the above.

[0013] In a third aspect, the present application provides a computer-readable storage medium, on which a computer program / instructions are stored, and when the computer program / instructions are executed by a processor, the method for analyzing the temporal similarity of sensor data in a wafer processing process as described in any one of the above is implemented.

[0014] In a fourth aspect, the present application provides a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the method for analyzing the temporal similarity of sensor data in a wafer processing process as described in any one of the above is implemented.

[0015] A method and device for analyzing the temporal similarity of sensor data in a wafer processing process provided by the present application obtain temporal data collected by a sensor corresponding to a target parameter in the wafer processing process; respectively determine a reference sequence and at least one comparison sequence based on the temporal data; perform data preprocessing of dimensionless and normalization on the reference sequence and the at least one comparison sequence respectively; for each comparison sequence, calculate the correlation coefficient of each data sampling point at the same position between the reference sequence and the comparison sequence based on the grey relational analysis algorithm; obtain a correlation degree for quantitatively representing the magnitude of the correlation between the reference sequence and each comparison sequence based on the correlation coefficients of each data sampling point. The present application improves the grey relational analysis algorithm and combines it with the FDC system to realize automatic temporal analysis of a large amount of temporal data. It helps the engineers in the wafer processing factory to analyze and summarize the data law of the change trend of the sensor data of the target parameter over time. This analysis method reduces the usage threshold of the temporal analysis algorithm, can complete specific analysis tasks with high quality without relying on rich analysis experience, reduces the manual analysis cost, improves the analysis efficiency and quality of the wafer processing process, and brings significant economic benefits to the production and management of the enterprise. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other implementation manners can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic flowchart of the method for analyzing the temporal similarity of sensor data in a wafer processing process provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed Embodiments

[0018] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented in accordance with the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives preferred embodiments and, in conjunction with the accompanying drawings, details are described as follows.

[0019] It should be noted that what is explicitly and implicitly understood by those of ordinary skill in the art is that the embodiments described in this application can be combined with other embodiments without conflict. Unless otherwise defined, the technical terms or scientific terms involved in this application should be the ordinary meanings understood by those with ordinary skills in the technical field to which this application belongs. The words such as "a", "one", "a kind", "the" and the like involved in this application do not indicate a limitation in quantity and can represent a singular or plural number. The terms "including", "comprising", "having" and any variations thereof involved in this application are intended to cover non-exclusive inclusion; the terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0020] The grey relational analysis method is a method for quantitatively describing and comparing the development and change trend of a system. Its basic idea is to judge the closeness of its connection according to the similarity degree of the geometric shapes of the sequence curves. The more similar the curves are, the closer the connection between the corresponding sequences is and the greater the correlation degree is. The grey relational analysis method is different from the correlation analysis method of mathematical statistics. First, the theoretical bases are different. Grey relational analysis is based on the grey process of the grey system, while the correlation analysis of mathematical statistics is based on the stochastic process of probability theory. Second, their analysis methods are different. Grey relational analysis is used to compare time series, while the correlation analysis of mathematical statistics is used to compare arrays between factors. Finally, their requirements for the amount of data are different. Grey relational analysis does not require too much data, while the correlation analysis of mathematical statistics requires a certain amount of data. In addition, grey relational analysis mainly studies the dynamic behavior of the system, while the correlation analysis of mathematical statistics mainly focuses on the static study of the system.

[0021] Refer to Figure 1 As shown, the embodiment of this application provides a method for analyzing the temporal similarity of sensor data in the wafer processing process, including: Step S1: Obtain the temporal data collected by the sensor corresponding to the target parameter in the wafer processing process.

[0022] Specifically, during the wafer processing, a large number of sensors are deployed to monitor various process parameters, such as temperature, pressure, flow rate, vibration, etc. These sensors collect data at certain time intervals to form time-series data. According to the specific process requirements of wafer processing, the target parameters to be monitored are determined. According to the characteristics of the target parameters, the monitoring data collected by the corresponding sensors are obtained. Taking the semiconductor furnace tube process as an example, the selected target parameter is set to GAS01, which refers to the concentration measurement of a certain process gas, and the time-series data of this target parameter collected by the corresponding sensor is obtained. Other target parameters can be FLOW01 (the flow measurement of a certain fluid), MFC01 (the mass measurement of a certain fluid), HEATPOWER01 (the measurement of the heating power of the machine).

[0023] Step S2: Based on the time-series data, determine the reference sequence and at least one comparison sequence respectively.

[0024] In this embodiment, step S2 specifically includes: Step S201: Select a representative data sequence from the time-series data as the reference sequence; Step S202: Consider the remaining time-series data after removing the reference sequence as a data sequence, and use it to intercept a comparison sequence that matches the length of the reference sequence; Step S203: According to the set window size and step size, gradually move the sliding window on the data sequence. After each move, the data segment within the sliding window will be updated to form a new comparison sequence.

[0025] Specifically, select a representative data sequence from the time-series data as the reference sequence. This data sequence can reflect the main characteristics of the overall data, such as trends, seasonality, periodicity, etc. The selection of the reference sequence can be determined based on historical data, expert experience, or specific analysis requirements. The reference sequence will be used as the benchmark for subsequent analysis, and will be compared with the comparison sequences to reveal the relationships or differences between them.

[0026] Remove the reference sequence from the original time-series data, and the remaining data constitutes a new data sequence. According to the length of the reference sequence, intercept a data segment with the same length from the remaining data sequence as the comparison sequence. This ensures that the comparison sequence is consistent with the reference sequence in length, facilitating comparative analysis.

[0027] The size of the sliding window determines the length of the data segment intercepted each time, and the step size determines the moving speed of the sliding window on the data sequence. By repeating the process of moving the sliding window, multiple comparison sequences for comparison with the reference sequence are generated. These comparison sequences cover different parts of the remaining data sequence, thus achieving a comprehensive analysis of the entire data sequence. This helps to reveal hidden patterns or trend changes in the time-series data, etc.

[0028] Step S3: Perform dimensionless and normalized data preprocessing on the reference sequence and at least one comparison sequence respectively.

[0029] Specifically, in a complex semiconductor processing and manufacturing system, there are often multiple different factors or variables. The physical meanings represented by these factors are different, resulting in different data dimensions, that is, the reference sequence x0 and the comparison sequence x i may have different dimensions. Data with different dimensions may vary greatly in numerical value, and their change ranges and trends may also be completely different. Direct comparison may lead to inaccurate or misleading results. For the convenience of analysis, before performing grey relational analysis, the reference sequence x0 and the comparison sequence x i are subjected to dimensionless and normalized data preprocessing.

[0030] The dimensionless and normalized data preprocessing includes: initialization processing, mean value processing or range transformation processing.

[0031] Initialization processing: Initialization processing is to divide all the data in a sequence by the first number in the sequence to obtain a new sequence. For a time series, this new sequence represents the multiples of the values at different times in the original sequence relative to the value at the first time. Obviously, this new sequence is dimensionless.

[0032] For example, let the original sequence be x = [x(1), x(2), …, x(n)], and after performing initialization processing on it, the new sequence obtained is as follows:

[0033] Mean value processing: Mean value processing is to divide all the data in a sequence by the average value of all the data in the sequence to obtain a new sequence. For a time series, this new sequence represents the multiples of the values at different times in the original sequence relative to the average value. Obviously, this new sequence is also dimensionless.

[0034] For example, let the original sequence be x = [x(1), x(2), …, x(n)], and the data average value is:

[0035] After performing mean value processing on the sequence x, the new sequence obtained is as follows:

[0036] Range transformation processing: Range transformation is a method for standardizing or normalizing data. It maps the original data to a specified interval, usually [0, 1] or [-1, 1]. Range transformation calculates the maximum and minimum values in the data sequence, and then subtracts the minimum value from each data point and divides by the data range (the maximum value minus the minimum value) to achieve data standardization. For spatial or time series, the range transformation method is often used for data processing.

[0037] For example, let the original sequence be x = [x(1), x(2), …, x(n)], and the new sequence obtained after range transformation processing is y. The i-th data in y is:

[0038] Step S4: For each comparison sequence, based on the grey relational analysis algorithm, calculate the correlation coefficients of each data sampling point at the same position between the reference sequence and the comparison sequence.

[0039] The existing grey relational analysis algorithm is used to calculate the differences between each sampling point of two sequences. After summing up and averaging the differences of all sampling points, it is the correlation coefficient. The disadvantage of this method is that it cannot measure whether the patterns or trends between sequences are the same. The analysis method provided in this application introduces a slope value on the basis of the original grey relational analysis method to re-measure the similarity between sequences, which can fully consider the consistency of trends between different sequences. Therefore, by adding a corresponding grey relational analysis module to the existing FDC system, engineers using this system can easily perform similarity analysis between sequences without additional learning costs.

[0040] In this embodiment, step S4 specifically includes: Step S401: Convert the reference sequence and at least one comparison sequence after data preprocessing into a first slope sequence and at least one second slope sequence respectively.

[0041] In this embodiment, step S401 specifically includes: Step S4011: Set the slope value of the first data sampling point of the first slope sequence and at least one second slope sequence to 0; Step S4012: For the second and subsequent data sampling points in the first slope sequence and at least one second slope sequence, subtract the value of the previous data sampling point from the value of the data sampling point to be used as the slope value of the data sampling point.

[0042] First, set the slope values of the first data sampling points of the first slope sequence (after the reference sequence is converted) and all second slope sequences (after the comparison sequences are converted) to 0.

[0043] Set the slope value of the k-th data sampling point according to the following calculation formula:

[0044] The slope value actually calculates the rate of change of each data point relative to its previous data point. This slope value can be positive or negative, representing the trend direction of the sequence. The converted slope sequence directly reflects the rate of change of the data value over time, which helps to more intuitively analyze the dynamic change characteristics of the data.

[0045] Step S402: For each second slope sequence, calculate the correlation coefficient of each data sampling point at the same position between the second slope sequence and the first slope sequence.

[0046] The correlation coefficient is used to measure the similarity or difference between the data points at the same position in two sequences. In the spatial dimension, the correlation essentially reflects the degree of difference in the geometric shapes between the sequence curves. Therefore, the degree of difference in curve shapes can be used as a scale to measure the degree of correlation. When the curve shapes of two sequences are very similar, their correlation coefficients will be relatively high; conversely, when the curve shapes are quite different, the correlation coefficients will be relatively low.

[0047] In this embodiment, step S402 specifically includes: Step S4021: Based on the first slope sequence and at least one second slope sequence, obtain the minimum and maximum values of the absolute differences of the slopes between all data sampling points in different sequences.

[0048] In this embodiment, step S4021 specifically includes: Step S40211: For each second slope sequence, calculate the first absolute slope difference between each data sampling point of the second slope sequence and the data sampling point at the corresponding position in the first slope sequence, and determine a first minimum absolute slope difference and a first maximum absolute slope difference from among the multiple first absolute slope differences; Step S40212: When there are multiple second slope sequences, obtain multiple first minimum absolute slope differences and multiple first maximum absolute slope differences; Step S40213: Determine a second minimum absolute slope difference from among the multiple first minimum absolute slope differences, and determine a second maximum absolute slope difference from among the multiple first maximum absolute slope differences, to be used as the minimum and maximum values of the multiple absolute differences of the slopes, respectively.

[0049] Specifically, let the first slope sequence after conversion of the reference sequence be g0 = [g0(1), g0(2), …, g0(n)], and the second slope sequence after conversion of the i-th comparison sequence for comparing the degree of correlation with the reference sequence be g i = [g i (1), gi (2), …, g i (n)], i = 1, 2, …, p. p is the number of comparison sequences, and the comparison sequences can be one or more than one.

[0050] For each second slope sequence, calculate the first slope absolute difference between each data sampling point of the second slope sequence and the data sampling point at the corresponding position in the first slope sequence . Using the calculation formula Determine a first minimum slope absolute difference from among multiple first slope absolute differences, and the calculation formula Determine a first maximum slope absolute difference from among multiple first slope absolute differences.

[0051] When there are multiple second slope sequences, that is, when p > 1, multiple first minimum slope absolute differences and multiple first maximum slope absolute differences will be obtained. Using the calculation formula Determine a second minimum slope absolute difference from among multiple first minimum slope absolute differences , and using the calculation formula Determine a second maximum slope absolute difference from among multiple first maximum slope absolute differences .

[0052] Step S4022: For each data sampling point of the second slope sequence, calculate the absolute difference between the slope value of the data sampling point and the slope value of the data sampling point at the same position in the first slope sequence.

[0053] Step S4023: Based on the minimum value and maximum value of the slope absolute difference, the absolute difference, and a preset resolution coefficient, determine the correlation coefficient between each data sampling point at the same position between the first slope sequence and the second slope sequence.

[0054] Specifically, the slope value of the kth sampling point in the second slope sequence and the slope value of the kth sampling point at the same position in the first slope sequence The absolute difference is .

[0055] The correlation coefficient is the correlation degree value between the second slope sequence and the first slope sequence at each data sampling point. The correlation coefficient of the kth sampling point and is as shown in formula (1):

[0056] In the formula, is the correlation coefficient, is the second minimum slope absolute difference, is the second maximum slope absolute difference, is the absolute difference of the slope values of the data sampling points at the same positions between sequences, is the resolution coefficient, which is used to improve the significance of the differences between sequences, , and usually takes 0.5.

[0057] Through the correlation coefficient, the slope differences between different comparison sequences and the reference sequence at the same time point can be quantified to help identify the sequence closest to the reference sequence, providing a basis for subsequent data analysis or decision-making.

[0058] Step S5: Based on the correlation coefficients of each data sampling point, obtain the correlation degree used to quantitatively represent the magnitude of the correlation between the reference sequence and each comparison sequence.

[0059] Specifically, since the correlation coefficient is a quantitative representation of the correlation degree values between the comparison sequence and the reference sequence at each sampling point, there is more than one of its data, and the information described is too scattered. In order to facilitate the overall comparison of sequences, it is necessary to concentrate the correlation coefficients of all sampling points into one value and perform information concentration processing using the correlation degree. The measure of the magnitude of the correlation that changes over time or different objects between two systems or two factors of a system is called the correlation degree. The correlation degree is used to quantitatively represent the overall magnitude of the correlation between the reference sequence and each comparison sequence.

[0060] The correlation degree is represented by the average value of the correlation coefficients of all data sampling points. The comparison sequence x i The correlation degree with the reference sequence x0 is:

[0061] In the formula, is the correlation degree, is the correlation coefficient, n is the length of the sequence, and k = 1, 2,..., n.

[0062] As a quantitative representation of the degree of correlation between the comparison sequence and the reference sequence, the correlation degree quantitatively describes the relative change situation between different factors in the wafer semiconductor processing and manufacturing process, that is, the relativity of the magnitude, direction, and speed of the change, etc. If the relative changes of the two are basically the same, it is considered that the correlation degree between the two is large; otherwise, the correlation degree between the two is small. Through the correlation degree, it can be intuitively understood which comparison sequence is more similar or has a higher degree of correlation with the reference sequence.

[0063] Exemplarily, assuming a scenario with GAS01 as the target parameter for illustration, the selected reference sequence is: [2.2, 4.3, 1.5, 3.7, 5.8, 7.6, 8, 3.3, 4.5, 3] Randomly select a comparison sequence as: [1.8, 3, 1.2, 2.9, 4.6, 5.5, 7.5, 3, 4, 2.5] Initialize the sequence. Divide all numbers in the sequence by the first number in the sequence to obtain: Reference sequence: [1.0, 1.95, 0.68, 1.68, 2.64, 3.45, 3.64, 1.5, 2.05, 1.36] Comparison sequence: [1.0, 1.67, 0.67, 1.61, 2.56, 3.06, 4.17, 1.67, 2.22, 1.39] Convert the reference sequence and the comparison sequence into a first slope sequence and a second slope sequence respectively to obtain: First slope sequence: [0, 0.95, -1.27, 1.0, 0.96, 0.81, 0.19, -2.14, 0.55, -0.69] Second slope sequence: [0, 0.67, -1.0, 0.94, 0.95, 0.5, 1.11, -2.5, 0.55, -0.83] Calculate the correlation coefficient: a) Calculate the minimum and maximum values of the absolute differences of the slopes at the same positions of the data sampling points in the first slope sequence and the second slope sequence respectively as: .

[0064] b) Take , and use Equation (1) to calculate the correlation coefficient to obtain: Correlation coefficient

[0065] c) Use Equation (2) to calculate the correlation degree, and obtain r = 3.6, which is the similarity between the comparison sequence and the target sequence.

[0066] For the business scenario of wafer processing quality analysis, this application also provides a client interface for automatically analyzing the temporal similarity of sensor data. Various query conditions can be input, such as Fab (factory), EQP Type (equipment type), Wafer (wafer number), Parameter (target parameter), etc. Then input the time range of the search comparison sequence and the'similarity threshold', and click the search button. The system starts to calculate and return the data sequences whose similarity exceeds the'similarity threshold'. To search for data sequences that are highly similar to the reference sequence corresponding to the target parameter within a certain time range. Therefore, a corresponding grey relational analysis module is added to the FDC system logic familiar to business personnel. Through automatic calculation and search, the efficient analysis of a large amount of sensor time-series data during wafer processing is realized, and the analysis efficiency of wafer processing quality is improved. In addition, the system does not require coding, reducing the usage threshold, enabling business personnel to easily get started without relying on the experience of specific technical personnel.

[0067] In summary, a method for analyzing the temporal similarity of sensor data during wafer processing provided by this application includes: obtaining the time-series data collected by sensors corresponding to target parameters during wafer processing; based on the time-series data, respectively determining a reference sequence and at least one comparison sequence; performing data preprocessing of dimensionless and normalization on the reference sequence and at least one comparison sequence respectively; for each comparison sequence, based on the grey relational analysis algorithm, calculating the correlation coefficients of each data sampling point at the same position between the reference sequence and the comparison sequence; based on the correlation coefficients of each data sampling point, obtaining the correlation degree used to quantitatively represent the magnitude of the correlation between the reference sequence and each comparison sequence. This application improves the grey relational analysis algorithm and combines it with the FDC system to realize automatic temporal analysis of a large amount of time-series data. It helps engineers in wafer processing factories analyze and summarize the data law of the change trend of sensor data of target parameters over time. This analysis method reduces the usage threshold of the temporal analysis algorithm, can complete specific analysis tasks with high quality without relying on rich analysis experience, reduces the manual analysis cost, improves the analysis efficiency and quality of the wafer processing process, and brings significant economic benefits to the production and management of enterprises.

[0068] Those skilled in the art can understand that in the above method of the specific implementation manner, the writing order of each step does not mean a strict execution order that constitutes any limitation to the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0069] In addition, some embodiments of the present application also provide an electronic device. The electronic device may be various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and so on. The electronic device may also be various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices.

[0070] The electronic device includes: one or more processors; and a memory storing computer program instructions that, when executed, cause the processors to perform the wafer processing process sensor data temporal similarity analysis method provided in any one or more of the above embodiments. Figure 2 An exemplary structural diagram of the electronic device is disclosed. As Figure 2 shown, the electronic device includes: one or more processors 1101, a memory 1102, and interfaces for connecting the components, including a high-speed interface and a low-speed interface. Each component is interconnected using different buses and may be mounted on a common motherboard or otherwise mounted as required. The processor may process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display graphical information of a GUI on an external input / output device (such as, a display device coupled to the interface). In some other embodiments, if necessary, multiple processors and / or multiple buses may be used together with multiple memories and multiple memories. Similarly, multiple electronic devices may be connected, each device providing some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Among them, the components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0071] The electronic device may further include: an input device 1103 and an output device 1104. The processor 1101, the memory 1102, the input device 1103, and the output device 1104 may be connected by a bus or other means, Figure 2 taking connection by bus as an example.

[0072] The input device 1103 can receive input digital or character information and generate key signal inputs related to the user settings and function controls of the electronic device, such as input devices like touchscreens, keypads, mice, trackpads, touchpads, pointing sticks, one or more mouse buttons, trackballs, joysticks, etc. The output device 1104 can include display devices, auxiliary lighting devices (e.g., LEDs), and haptic feedback devices (e.g., vibration motors), etc. The display device can include, but is not limited to, liquid crystal displays (LCDs), light-emitting diode (LED) displays, and plasma displays. In some embodiments, the display device can be a touchscreen.

[0073] To provide interaction with the user, the electronic device can be a computer. The computer has: a display device for displaying information to the user (e.g., a cathode ray tube (CRT) or an LCD monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball), through which the user can provide inputs to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or haptic feedback); and inputs from the user can be received in any form (including acoustic input, voice input, or haptic input).

[0074] In the embodiments of this application, a computer program / instructions is stored on a computer-readable medium. When the computer program / instructions are executed by a processor, the wafer processing process sensor data temporal similarity analysis method provided in any one or more of the above embodiments is implemented. The computer-readable medium can be included in the electronic device described in the above embodiments; or it can exist separately without being assembled into the device. The above computer-readable medium carries one or more computer-readable instructions.

[0075] The memory 1102 can be used as a non-transitory computer-readable storage medium for storing non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 1102, so as to implement the program instructions / modules corresponding to the method provided in any one or more of the above embodiments of this application.

[0076] The memory 1102 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the electronic device and the like. In addition, the memory 1102 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory 1102 may optionally include memories remotely provided relative to the processor 1101, and these remote memories may be connected to the electronic device through a network. Examples of the above networks include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0077] It should be noted that more specific examples of computer-readable storage media may include but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

[0078] A computer-readable storage medium includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0079] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0080] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. For example, an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device can be used. In some embodiments, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) can be stored in a computer-readable recording medium, such as a RAM memory, a magnetic or optical drive, or a floppy disk and similar devices. Additionally, some steps or functions of the present application can be implemented by hardware, for example, as a circuit that cooperates with the processor to execute each step or function.

[0081] The computer program product provided by the embodiments of the present application includes one or more computer programs / instructions. When the computer program / instructions are executed by a processor, they wholly or partly generate the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive, SSD).

[0082] The flowcharts or block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of devices, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions denoted in the blocks may occur in a different order than that denoted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0083] As described above, the above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily mention changes or substitutions, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A method for analyzing the temporal similarity of sensor data in a wafer processing process, characterized in that, Including: Obtaining the time-series data collected by a sensor corresponding to a target parameter during the wafer processing; Based on the time-series data, respectively determining a reference sequence and at least one comparison sequence; Performing data preprocessing of dimensionless and normalization on the reference sequence and the at least one comparison sequence respectively; For each of the comparison sequences, based on the grey relational analysis algorithm, calculating the correlation coefficients of each data sampling point at the same position between the reference sequence and the comparison sequence; Based on the correlation coefficients of each data sampling point, obtaining a correlation degree for quantitatively representing the magnitude of the correlation between the reference sequence and each comparison sequence; The step of respectively determining a reference sequence and at least one comparison sequence based on the time-series data includes: Selecting a representative data sequence from the time-series data as the reference sequence; Regarding the remaining time-series data after removing the reference sequence as a data sequence for intercepting a comparison sequence matching the length of the reference sequence; According to the set window size and step size, gradually moving the sliding window on the data sequence. After each movement, the data segment within the sliding window will be updated to form a new comparison sequence.

2. The method for analyzing the temporal similarity of sensor data in a wafer processing process according to claim 1, wherein The step of, for each of the comparison sequences, calculating the correlation coefficients of each data sampling point at the same position between the reference sequence and the comparison sequence based on the grey relational analysis algorithm includes: Respectively converting the reference sequence and the at least one comparison sequence after data preprocessing into a first slope sequence and at least one second slope sequence; For each of the second slope sequences, respectively calculating the correlation coefficients of each data sampling point at the same position between the second slope sequence and the first slope sequence.

3. The method for analyzing the temporal similarity of sensor data in a wafer processing process according to claim 2, wherein The step of respectively converting the reference sequence and the at least one comparison sequence after data preprocessing into a first slope sequence and at least one second slope sequence includes: Setting the slope value of the first data sampling point of the first slope sequence and the at least one second slope sequence to 0; For the second and subsequent data sampling points in the first slope sequence and the at least one second slope sequence, subtracting the value of the previous data sampling point from the value of the data sampling point as the slope value of the data sampling point.

4. The method for analyzing the temporal similarity of wafer processing process sensor data according to claim 3, characterized in that The step of, for each of the second slope sequences, respectively calculating the correlation coefficients of each data sampling point at the same position between the second slope sequence and the first slope sequence includes: Based on the first slope sequence and the at least one second slope sequence, obtaining the minimum value and the maximum value of the absolute differences of the slopes between all data sampling points of different sequences; For each data sampling point of the second slope sequence, calculating the absolute difference of the slope value of the data sampling point at the same position in the first slope sequence; Based on the minimum value and the maximum value of the absolute differences of the slopes, the absolute difference, and a preset resolution coefficient, determining the correlation coefficients of each data sampling point at the same position between the first slope sequence and the second slope sequence.

5. The method for analyzing the temporal similarity of sensor data in the wafer processing process according to claim 4, wherein The step of obtaining the minimum value and the maximum value of the absolute differences of slopes between all data sampling points among different sequences based on the first slope sequence and the at least one second slope sequence includes: For each of the second slope sequences, calculate the first absolute slope difference between the slope values of the respective data sampling points of the second slope sequence and the data sampling points at the corresponding positions in the first slope sequence, and determine a first minimum absolute slope difference and a first maximum absolute slope difference from among the multiple first absolute slope differences; When there are multiple second slope sequences, obtain multiple first minimum absolute slope differences and multiple first maximum absolute slope differences; Determine a second minimum absolute slope difference from among the multiple first minimum absolute slope differences, and determine a second maximum absolute slope difference from among the multiple first maximum absolute slope differences, to respectively serve as the minimum value and the maximum value of the absolute differences of slopes.

6. The method for analyzing the temporal similarity of sensor data in a wafer processing process according to claim 1, wherein The step of obtaining the degree of association for quantitatively representing the degree of association between the reference sequence and each of the comparison sequences based on the correlation coefficients of the respective data sampling points includes: For each of the comparison sequences, take the average of the correlation coefficients of all data sampling points to obtain a quantitative representation of the degree of association with the reference sequence, i.e., the degree of association.

7. An electronic device, characterized in that, The electronic device includes: One or more processors; and a memory storing computer program instructions, the computer program instructions, when executed, causing the processor to execute the method for analyzing the temporal similarity of sensor data in a wafer processing process according to any one of claims 1-6.

8. A computer-readable storage medium having computer programs / instructions stored thereon, characterized in that, The computer program / instructions, when executed by the processor, implement the method for analyzing the temporal similarity of sensor data in a wafer processing process according to any one of claims 1-6.

9. A computer program product, comprising a computer program / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the method for analyzing the temporal similarity of sensor data in a wafer processing process according to any one of claims 1-6.

Images