Time parameter updating method and computing device
By using statistical analysis and outlier handling based on target operating data, the time parameters of semiconductor process equipment are automatically updated, solving the problems of low update efficiency and insufficient accuracy in existing technologies. This achieves efficient and accurate updating of time parameters, thereby improving production efficiency.
Patent Information
- Application Number
- CN202411046004.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2026-02-03
AI Technical Summary
In existing technologies, the time parameter update efficiency of semiconductor process equipment is low and the accuracy is insufficient, resulting in the inability to guarantee production efficiency.
Based on the target operating data of semiconductor process equipment, the first statistical data of each time parameter is determined. Through data category distribution information and outlier handling, the target values of the time parameters are automatically updated for use by the scheduling algorithm.
This improves the efficiency and accuracy of time parameter updates, ensures the effectiveness of scheduling algorithm results, and enhances the production efficiency of semiconductor process equipment.
Smart Images
Figure CN121457761A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of semiconductor manufacturing, and in particular, to a time parameter updating method and a computing device. BACKGROUND
[0002] In the integrated circuit industry, the processing of semiconductor process equipment can be described as the transmission and process of materials in the machine. The scheduling algorithm determines the parameter value of the time parameter (such as the action time of the manipulator, the process time of the chamber, the door opening and closing time, etc.) of each module of the semiconductor process equipment according to the corresponding action, finely schedules the production process, and reasonably allocates each module of the semiconductor process equipment, so as to shorten the processing time of a fixed batch of materials and achieve the goal of maximizing the production capacity. Therefore, the accuracy of the parameter value of the time parameter of each module of the semiconductor process equipment performing the corresponding action is an important factor affecting the scheduling result of the scheduling algorithm.
[0003] At present, the target value of the time parameter of each module performing the corresponding action is usually determined by manually analyzing the relevant running data of the semiconductor process equipment when the semiconductor process equipment is not running, and the target value is written into a configuration file to update the time parameter, so as to be used for scheduling by the scheduling algorithm. It is time-consuming and laborious, thereby resulting in low updating efficiency of the time parameter and being unable to guarantee the accuracy of the updating result of the time parameter, so as to be unable to guarantee the production efficiency of the semiconductor process equipment. SUMMARY
[0004] To solve the above technical problems, the present application provides a time parameter updating method and a computing device to solve the problems of low updating efficiency and low accuracy of the updating result of the time parameter in the prior art.
[0005] To achieve the above technical purposes, the embodiments of the present application provide the following technical solutions:
[0006] In a first aspect, the embodiments of the present application provide a time parameter updating method, comprising:
[0007] determining first statistical data of each time parameter of a semiconductor process equipment based on target running data of the semiconductor process equipment; the target running data comprises detection values of each time parameter at a plurality of historical time points, and the first statistical data is used to represent the change state of the time parameter;
[0008] when the change state of the time parameter is determined to be a stable state based on the first statistical data of the time parameter, determining data category distribution information of the time parameter based on the detection values of the time parameter at a plurality of historical time points;
[0009] When the data category distribution information represents a number of data categories contained in the time parameter is less than or equal to a predetermined value, a target value of the time parameter is determined based on detection values of the time parameter at a plurality of historical time points, and the target value of the time parameter is used for device scheduling by a scheduling algorithm of the semiconductor process equipment.
[0010] In an embodiment, the data category distribution information of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points, including:
[0011] Second statistical data of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points, and the second statistical data is used to represent a distribution state of parameter values of the time parameter;
[0012] An abnormal value in the detection values of the time parameter at the plurality of historical time points is determined based on the second statistical data, and the abnormal value is eliminated to obtain preferred data of the time parameter;
[0013] The data category distribution information of the time parameter is determined based on the preferred data of the time parameter.
[0014] In an embodiment, the second statistical data of the time parameter includes a first mean value and a standard deviation of the time parameter;
[0015] The abnormal value in the detection values of the time parameter at the plurality of historical time points is determined based on the second statistical data, including:
[0016] A target value range of the time parameter is determined based on the first mean value and the standard deviation of the time parameter;
[0017] The abnormal value in the detection values of the time parameter at the plurality of historical time points is determined based on a comparison result of the detection values of the time parameter at the plurality of historical time points and the target value range.
[0018] In an embodiment, the data category distribution information of the time parameter is determined based on the preferred data of the time parameter, including:
[0019] Distance information between each data in the preferred data of the time parameter is obtained to obtain a plurality of distance information;
[0020] Based on the plurality of distance information, a data category contained in the time parameter is iteratively updated, and the data category distribution information of the time parameter is determined based on a result of the iterative update.
[0021] In an embodiment, the data category contained in the time parameter is iteratively updated based on the plurality of distance information, including:
[0022] In the current iteration, based on the plurality of distance information, an inter-class distance between each data class contained in the time parameter is determined.
[0023] When the inter-class distance is less than or equal to a preset distance threshold, data in a target data class contained in the time parameter is merged to obtain an update result of the data class contained in the time parameter in the current iteration, and the target data class includes data classes whose inter-class distances are all less than or equal to the preset distance threshold.
[0024] In an embodiment, the determination of the target value of the time parameter based on the detection values of the time parameter at the plurality of historical time points includes:
[0025] Based on the preferred data of the time parameter, a second mean value of the time parameter is determined.
[0026] The deviation between the second mean value and a current configuration value of the time parameter is obtained, and when the deviation is greater than a preset deviation value, the second mean value is taken as the target value of the time parameter.
[0027] In an embodiment, the first statistical data of the time parameter includes a variance of the time parameter and / or a range of the time parameter.
[0028] Based on the first statistical data of the time parameter, it is determined whether the change state of the time parameter is a stable state, including:
[0029] Based on a comparison result of the variance of the time parameter and a preset variance value, and / or based on a comparison result of the range of the time parameter and a preset range value, it is determined whether the change state of the time parameter is a stable state.
[0030] In an embodiment, the method further includes:
[0031] When the change state of the time parameter is a fluctuation state, or the data class distribution information represents that the number of data classes contained in the time parameter is greater than the predetermined value, a prompt information is generated.
[0032] In a second aspect, the embodiments of the present specification provide a computing device, including at least one processor and at least one memory, the memory storing a computer program, and the computer program is executed by the processor to implement the time parameter update method according to any one of the above embodiments.
[0033] In a third aspect, the embodiments of the present specification provide a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program is run by a processor, the time parameter updating method according to any one of the preceding aspects is implemented.
[0034] In a fourth aspect, the embodiments of the present specification provide a computer program product or a computer program. The computer program product includes a computer program stored in a computer readable storage medium. The processor of the computer device reads the computer program from the computer readable storage medium, and the processor implements the time parameter updating method according to any one of the preceding aspects when executing the computer program.
[0035] It can be seen from the above technical solutions that the embodiments of the present application provide a time parameter updating method and a computing device. The time parameter updating method determines first statistical data of each time parameter of a semiconductor process equipment based on target running data of the semiconductor process equipment. The target running data includes detection values of each time parameter at a plurality of historical time points. The first statistical data is used to represent the change state of the time parameter. When the change state of the time parameter is determined to be a stable state based on the first statistical data of the time parameter, the data category distribution information of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points. When the number of data categories contained in the time parameter is less than or equal to a predetermined value, the target value of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points. The target value of the time parameter is used for device scheduling by a scheduling algorithm of the semiconductor process equipment. Therefore, the automatic updating of the parameter values of each time parameter of the semiconductor process equipment can be realized, and the updating efficiency of the time parameter is improved. At the same time, by comprehensively considering the change state of the time parameter, the number of data categories contained in the time parameter, and the size of the detection value, the accuracy of the target value of the time parameter can be effectively ensured, and the effective updating of the time parameter is realized, thereby ensuring the effectiveness of the scheduling result of the scheduling algorithm. BRIEF DESCRIPTION OF DRAWINGS
[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creative labor.
[0037] Figure 1 A structural schematic diagram of a semiconductor process equipment is provided for an embodiment of the present specification.
[0038] Figure 2A flowchart of a time parameter updating method provided for an embodiment of the present specification;
[0039] Figure 3 A distribution diagram of a detection value corresponding to a vacuum conversion duration of a vacuum lock provided for an embodiment of the present specification;
[0040] Figure 4 A distribution diagram of a detection value corresponding to a door closing duration of a process chamber provided for an embodiment of the present specification;
[0041] Figure 5 A flowchart of another time parameter updating method provided for an embodiment of the present specification. DETAILED DESCRIPTION
[0042] Unless otherwise defined, technical or scientific terms used in the embodiments of the present specification shall have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments of the present specification belong. Unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present specification are not limited to the commonly understood meanings of the terms. The terms “first”, “second”, and similar terms are not intended to denote any order, quantity, or importance, but are used to avoid confusion among the components.
[0043] Unless otherwise required by context, “plurality” shall mean “at least two” throughout the specification. “Include” is to be interpreted as open, inclusive, meaning “including, but not limited to.” In the description of the specification, the terms “one embodiment”, “some embodiments”, “exemplary embodiments”, “example”, “specific example” or “some examples” are intended to indicate a feature, structure, material, or characteristic that is included in at least one embodiment of the specification. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example.
[0044] The technical solutions in the embodiments of the present specification will be described clearly and completely below in conjunction with the accompanying drawings of the embodiments of the present specification. Obviously, the described embodiments are only part of the embodiments of the present specification, not all. Based on the embodiments in the present specification, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the present specification.
[0045] SUMMARY
[0046] As described in the background, in the integrated circuit industry, the processing of semiconductor process equipment can be described as the transmission and process of materials in the machine. The scheduling algorithm schedules the production process according to the parameter values of the time parameters of the respective actions of the modules of the semiconductor process equipment (such as the action time of the robot, the process time of the chamber, the opening and closing time of the door, etc.), and reasonably allocates the modules of the semiconductor process equipment to shorten the processing time of the fixed batch of materials and achieve the goal of maximizing the production capacity. Therefore, the accuracy of the parameter values of the time parameters of the respective actions of the modules of the semiconductor process equipment is an important factor affecting the scheduling results of the scheduling algorithm.
[0047] For example, the structure of the semiconductor process equipment can be as shown in Figure 1 The process processing path can include: cassette chamber 106→air robot 103→aligner 105→load lock 104→vacuum robot 102→process chamber 101→vacuum robot 102→load lock 104→air robot 103→cassette chamber 106. The time parameters involved in the scheduling algorithm corresponding to the process processing path can include: the wafer picking time t1 of the air robot 103, the wafer placing time t2 of the air robot 103, the wafer picking time t3 of the vacuum robot 102, the wafer placing time t4 of the vacuum robot 102, the alignment time t5 of the aligner 105, the processing time t6 of the process chamber 101, the opening and closing valve time t7, and the vacuum conversion time t8 of the load lock 104.
[0048] However, when the semiconductor process equipment is running for a long time or there is mechanical wear, the time length of the respective actions of the modules in the actual running process will change, that is, the parameter values of the corresponding time parameters will change, thereby greatly affecting the scheduling effect of the scheduling algorithm, and thus the production efficiency of the semiconductor process equipment cannot be guaranteed.
[0049] At present, the target values of the time parameters of the respective actions of the modules are usually determined by manually analyzing the relevant running data of the semiconductor process equipment when the semiconductor process equipment is not running, and the target values are written into the configuration file to update the time parameters for the scheduling algorithm to schedule, which is time-consuming and laborious, thereby resulting in low efficiency of updating the time parameters and being unable to guarantee the accuracy of the updating results of the time parameters, thereby being unable to guarantee the production efficiency of the semiconductor process equipment.
[0050] To solve the problems of low updating efficiency and low accuracy of updating results of the time parameters in the conventional method, in the technical solution of the present application, first statistical data of each time parameter of a semiconductor process equipment is determined based on target operation data of the semiconductor process equipment, the target operation data including detection values of each time parameter at a plurality of historical time points, the first statistical data being used to represent a change state of the time parameter, and when the change state of the time parameter is determined to be a stable state based on the first statistical data of the time parameter, data category distribution information of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points, and when the data category distribution information of the time parameter represents that the number of data categories contained by the time parameter is less than or equal to a predetermined value, a target value of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points, the target value of the time parameter being used for device scheduling by a scheduling algorithm of the semiconductor process equipment, so that automatic updating of the parameter values of each time parameter of the semiconductor process equipment can be realized, and the updating efficiency of the time parameters is improved. Meanwhile, by comprehensively considering the change state of the time parameter, the number of data categories contained by the time parameter, and the size of the detection values, the accuracy of the target value of the time parameter can be effectively ensured, and effective updating of the time parameter is realized, and the effectiveness of the scheduling result of the scheduling algorithm is ensured.
[0051] Meanwhile, to further improve the accuracy of the determination result of the target value of the time parameter, the technical solution provided by the present application further limits the specific ways of determining the data category distribution information of the time parameter, determining the target value of the time parameter, and determining the change state of the time parameter.
[0052] Based on the above inventive concept, the time parameter updating method provided by the present application is exemplarily described below.
[0053] Exemplary method
[0054] The present application provides a time parameter updating method, as shown in Figure 2 The present application provides a time parameter updating method, as shown in
[0055] S201, first statistical data of each time parameter of a semiconductor process equipment is determined based on target operation data of the semiconductor process equipment, the target operation data including detection values of each time parameter at a plurality of historical time points, the first statistical data being used to represent a change state of the time parameter.
[0056] Specifically, the semiconductor process equipment can be a cluster equipment including a plurality of modules. The target operation data of the semiconductor process equipment can include detection values of each time parameter at a plurality of historical time points, and the detection value is the time length of the corresponding module performing the corresponding action in the actual operation process.
[0057] In implementations, the data list corresponding to each time parameter can be obtained by parsing the scheduling algorithm action time data recorded in log files, databases or memories, and for any time parameter, the data list corresponding to the time parameter can be used to record the parameter value sequence of the time parameter, which can include configuration values and detection values of the time parameter at each historical time. For example, the format of the data list corresponding to the time parameter can be "Dictionary<string, List <double>>, where Dictionary is a dictionary, which can be represented as a list of data, string represents the parameter name of the time parameter, and List <double>a parameter value sequence of the time parameter, a first element in the parameter value sequence can be a configuration value of the time parameter, other elements in the parameter value sequence can be detection values of the time parameter at historical time points, and a data format of each parameter value in the parameter value sequence can be a double-precision floating point number (double). As an optional implementation, a data list of the time parameter can be "<parameter name, [configuration value, detection value 1, detection value 2, …, detection value N]>", N is the number of detection values of the time parameter, for example, taking the case of the atmospheric mechanical hand 103 picking a wafer from the vacuum lock 104 as an example, the parameter list of the corresponding time parameter can be <ATR Pick LL, [2.0, 1.8, 2.1, 2.0, 2.2, 1.7, …, 2.1]>; for example, taking the case of the vacuum mechanical hand 102 picking a wafer from the process chamber 101 as an example, the parameter list of the corresponding time parameter can be <VTR Pick PM, [5.0, 5.4, 5.0, 5.1, 5.0, 4.9, …, 5.2]>.
[0058] For any time parameter, a first statistical data of the time parameter can be determined based on the detection values of the time parameter at historical time points, the first statistical data can represent a variation state of the time parameter, the variation state of the time parameter can be a stable state or a fluctuation state, the stable state represents that a maximum variation amplitude of the detection values of the time parameter at the historical time points is less than or equal to a preset amplitude, and the fluctuation state represents that the maximum variation amplitude of the detection values of the time parameter at the historical time points is greater than the preset amplitude. For example, the first statistical data can include a variance of the time parameter and / or a range of the time parameter.
[0059] S202, when the variation state of the time parameter is determined to be a stable state based on the first statistical data of the time parameter, determining data class distribution information of the time parameter based on the detection values of the time parameter at a plurality of historical time points.
[0060] Specifically, for any time parameter, the variation state of the time parameter can be determined based on the first statistical data of the time parameter, for example, the variation state of the time parameter can be determined based on a comparison result of the first statistical data of the time parameter and a corresponding preset threshold, and when the variation state of the time parameter is a stable state, the data class distribution information of the time parameter can be determined based on the detection values of the time parameter at a plurality of historical time points, thereby effectively avoiding the influence of a large fluctuation amplitude of the time parameter on the accuracy of the recognition result of the data class distribution information of the time parameter when the hardware and / or software of the semiconductor process equipment fails.
[0061] The data categories included in the time parameter can include one or more modal values of the action execution duration corresponding to the time parameter, and the modal values are concentrated distribution values. The distance between the modal values represented by different data categories can be greater than a preset distance threshold. In implementation, the data categories can be divided according to the action execution duration corresponding to the time parameter. The data category distribution information of the time parameter can represent the distribution region of the detection values of the time parameter at each historical time. The data category distribution information can include the number of data categories, and can also include the statistical result of the detection values corresponding to each data category. The statistical result can include the mean value, the central value, etc. For example, if the time parameter is the door closing duration of the process chamber 101, and the detection values of the door closing duration of the process chamber 101 at multiple historical times are concentrated at two different modal values of 2 seconds and 5 seconds, respectively, the door closing duration of the process chamber 101 can include two data categories of 2 seconds and 5 seconds. In implementation, the door closing duration of the process chamber 101 can be subdivided into two time parameters of the door closing duration after the wafer is placed in the process chamber 101 by the vacuum manipulator 102 and the door closing duration after the wafer is taken out of the process chamber 101 by the vacuum manipulator 102, and the target values of the two time parameters are 2 seconds and 5 seconds, respectively.
[0062] In implementation, the distance information between the detection values corresponding to the time parameter can be obtained, and the data category distribution information of the time parameter can be determined based on the distance information between the detection values and the preset distance threshold between different data categories. For example, the data categories included in the time parameter can be initialized, and the data categories included in the time parameter can be iteratively updated based on the distance information between the detection values and the preset distance threshold between different data categories, to obtain the data category distribution information of the time parameter. The maximum distance in the distance information can also be obtained. When the maximum distance is less than or equal to the preset distance threshold, it is determined that the number of data categories included in the time parameter is 1. When the maximum distance is greater than the preset distance threshold, it is determined that the number of data categories included in the time parameter is greater than 1, and the numerical interval corresponding to the time parameter is divided based on the preset distance threshold to obtain multiple subintervals. The length of each subinterval is less than or equal to the preset distance threshold. The data category distribution information of the time parameter can be determined according to the subinterval in which each detection value corresponding to the time parameter is located. The numerical interval corresponding to the time parameter can be determined based on the maximum value and the minimum value in the multiple detection values corresponding to the time parameter.
[0063] S203, when the data category distribution information represents that the number of data categories contained by the time parameter is less than or equal to a predetermined value, determining a target value of the time parameter based on the detection values of the time parameter at the plurality of historical time points, the target value of the time parameter being used for device scheduling by a scheduling algorithm of the semiconductor process equipment.
[0064] Specifically, the predetermined value can be set according to actual needs, for example, when the data category distribution information represents that the number of data categories contained by the time parameter is 1, the target value of the time parameter can be determined based on the detection values of the time parameter at the plurality of historical time points, thereby effectively avoiding the influence of unreasonable data interface setting on the accuracy of the determination result of the target value of the time parameter.
[0065] The target statistical value of the time parameter can include the mean or central value of the time parameter, and the deviation between the target statistical value of the time parameter and the current configuration value of the time parameter is obtained to determine the target value of the time parameter according to the deviation, thereby realizing automatic updating of the parameter values of the time parameters of the semiconductor process equipment and improving the updating efficiency of the time parameters. In addition, the embodiment method comprehensively considers the change state of the time parameter, the number of data categories contained and the size of the detection value, which can effectively ensure the accuracy of the target value of the time parameter, thereby realizing effective updating of the time parameter and ensuring the effectiveness of the scheduling result of the scheduling algorithm.
[0066] The target statistical value of the time parameter can include the mean or central value of the time parameter, and the deviation between the target statistical value of the time parameter and the current configuration value of the time parameter is obtained to determine the target value of the time parameter according to the deviation, thereby realizing automatic updating of the parameter values of the time parameters of the semiconductor process equipment and improving the updating efficiency of the time parameters. In addition, the embodiment method comprehensively considers the change state of the time parameter, the number of data categories contained and the size of the detection value, which can effectively ensure the accuracy of the target value of the time parameter, thereby realizing effective updating of the time parameter and ensuring the effectiveness of the scheduling result of the scheduling algorithm.
[0067] In the implementation, when the target value of the time parameter represents that the current configuration value of the time parameter is unreasonable, the configuration value of the time parameter can be updated based on the target value for device scheduling by the scheduling algorithm, for example, if the semiconductor process equipment is in a non-running state at the current time, the target value of the time parameter can be written into a configuration file to configure the time parameter through the configuration file; if the semiconductor process equipment is in a running state at the current time, the target value of the time parameter can be directly transmitted to the scheduling algorithm, thereby ensuring the effectiveness of the scheduling result.
[0068] It can be understood that the configuration value of the time parameter can remain unchanged when the target value of the time parameter represents the current configuration value of the time parameter.
[0069] To further improve the accuracy of the determination result of the data category distribution information of the time parameter, in an embodiment of the present specification, the determination of the data category distribution information of the time parameter based on the detection values of the time parameter at the plurality of historical time points comprises:
[0070] determining second statistical data of the time parameter based on the detection values of the time parameter at the plurality of historical time points, the second statistical data being used to represent the distribution state of the parameter value of the time parameter;
[0071] determining abnormal values in the detection values of the time parameter at the plurality of historical time points based on the second statistical data, and eliminating the abnormal values to obtain preferred data of the time parameter;
[0072] determining the data category distribution information of the time parameter based on the preferred data of the time parameter.
[0073] Specifically, for any time parameter, the second statistical data of the time parameter can be used to represent the distribution state of the parameter value of the time parameter, for example, the second statistical data can include one or more of the mean value, the central value and the standard deviation of the time parameter.
[0074] In implementation, the abnormal values in the detection values of the time parameter at the plurality of historical time points can be determined based on the second statistical data of the time parameter, for example, the target value interval of the time parameter can be determined based on the second statistical data, and the abnormal values in the detection values of the time parameter at the plurality of historical time points can be determined based on the comparison result of the detection values of the time parameter at the plurality of historical time points and the target value interval.
[0075] The target value interval of the time parameter can be determined based on the target deviation value, the mean value or the central value of the time parameter and the theoretical value interval of the time parameter, the target deviation value can represent the allowable fluctuation range of the parameter value of the time parameter, the target deviation value can be a preset value, and the target deviation value can also be determined based on the standard deviation of the time parameter.
[0076] After determining the abnormal values in the detection values of the time parameter at the plurality of historical time points, the abnormal values can be eliminated to obtain the preferred data of the time parameter, and the preferred data includes the non-abnormal values in the detection values of the time parameter at the plurality of historical time points. Thus, the data category distribution information of the time parameter can be determined based on the preferred data of the time parameter, so that the influence of abnormal data on the determination result of the data category distribution information can be effectively avoided, and the accuracy of the determination result of the data category distribution information of the time parameter is improved.
[0077] In view of the large difference in the time length of different actions performed by the semiconductor process equipment, in order to quickly and accurately determine the data class distribution information of each time parameter, the preferred data of each time parameter can be normalized before determining the data class distribution information of the time parameter based on the preferred data of the time parameter, that is, the preferred data of each time parameter is mapped to the range of 0-1 to eliminate the dimensional difference between different time parameters, thereby effectively improving the recognition efficiency and accuracy of the data class distribution information of each time parameter. Among them, for any data in the preferred data of the time parameter, the normalization result of the data can include the ratio of the value of the data to the maximum value in the preferred data of the time parameter. For example, the preferred data of the wafer taking time length of the atmospheric mechanical hand 103 in the wafer box cavity 106 can include [1, 1, 1, 2, 1, 0.9, 2, 1.5, 1.7, 2.1], and the preferred data of the processing time length of the process chamber 101 can include [90, 91, 92, 91.5, 90.1, 89.7, 94.3, 88, 91, 89.5]. After data normalization, the preferred data of the wafer taking time length of the atmospheric mechanical hand 103 in the wafer box cavity 106 is updated to [0.48, 0.48, 0.48, 0.95, 0.48, 0.43, 0.95, 0.71, 0.81, 1.0], and the preferred data of the processing time length of the process chamber 101 is updated to [0.95, 0.97, 0.98, 0.97, 0.96, 0.95, 1.0, 0.93, 0.97, 0.95]. Both kinds of data are mapped to the interval [0, 1], eliminating the dimensional difference between different time parameters, facilitating subsequent uniform processing and data class distribution information recognition operation of the preferred data of each time parameter.
[0078] In one possible implementation, the second statistical data of the time parameter includes a first mean value and a standard deviation of the time parameter.
[0079] The determining of the abnormal value in the detection values of the time parameter at the plurality of historical time points based on the second statistical data includes:
[0080] Determining a target value range of the time parameter based on the first mean value and the standard deviation of the time parameter.
[0081] Determining the abnormal value in the detection values of the time parameter at the plurality of historical time points based on a comparison result of the detection values of the time parameter at the plurality of historical time points and the target value range.
[0082] Specifically, for any time parameter, the second statistical data can include a first mean value and a standard deviation of the time parameter, the first mean value can include a mean value of the detected values of the time parameter at the plurality of historical time points in the target running data, and the standard deviation can include a standard deviation of the detected values of the time parameter at the plurality of historical time points in the target running data.
[0083] In an implementation, a target value range of the time parameter can be determined based on the first mean value and the standard deviation of the time parameter. For example, a target deviation value of the time parameter can be determined based on the standard deviation of the time parameter, and a target value range of the time parameter can be determined based on the first mean value and the target deviation value of the time parameter. For example, the sum of the first mean value and the target deviation value can be taken as an upper limit value of the target value range, and the difference between the first mean value and the target deviation value can be taken as a lower limit value of the target value range. The standard deviation of the time parameter can be directly taken as the target deviation value, or the standard deviation of the time parameter can be corrected based on a preset correction coefficient, and the result of the correction is taken as the target deviation value. In addition, the target value range of the time parameter can be corrected based on a theoretical value range of the time parameter. For example, the intersection of the theoretical value range of the time parameter and the target value range of the time parameter can be taken as the result of the correction of the target value range of the time parameter. The theoretical value range of the time parameter can be a preset value range.
[0084] After obtaining the target value range of the time parameter, the detected values of the time parameter at the plurality of historical time points can be compared with the target value range respectively, and the abnormal values in the detected values of the time parameter at the plurality of historical time points can be determined based on the comparison result. For example, if the detected value is in the target value range, the detected value is taken as a non-abnormal value, and if the detected value is not in the target value range, the detected value is taken as an abnormal value. Thus, the effectiveness of the detection result of the abnormal value can be effectively ensured, and the effectiveness of the preferred data of each time parameter is improved, thereby providing data support for improving the accuracy of the determination result of the data category distribution information of the time parameter.
[0085] In one possible implementation, the determination of the data category distribution information of the time parameter based on the preferred data of the time parameter includes:
[0086] Obtaining distance information between each data in the preferred data of the time parameter to obtain a plurality of distance information;
[0087] Iteratively updating the data categories contained in the time parameter based on the plurality of distance information, and determining the data category distribution information of the time parameter based on the result of the iterative update.
[0088] Specifically, for any time parameter, distance information between each two data in the preferred data of the time parameter can be obtained to obtain a plurality of distance information. The distance information between the two data can be an absolute value of a difference between the two data.
[0089] In an implementation, the data categories contained in the time parameter can be iteratively updated based on the plurality of distance information. For example, the data categories contained in the time parameter can be initialized, and the data categories contained in the time parameter can be iteratively updated based on the plurality of distance information and a preset distance threshold between different data categories to obtain the data category distribution information of the time parameter. For example, in each iteration, the inter-class distance between each data category contained in the time parameter can be determined based on each distance information, and the data categories can be merged based on a comparison result of the inter-class distance and the preset distance threshold to obtain an update result of the data categories contained in the time parameter in the current iteration.
[0090] It can be understood that when the inter-class distance between each data category is greater than the preset distance threshold, or the number of the data categories contained in the time parameter is 1, it can be determined that the iteration termination condition is met, and the data categories contained in the time parameter in the current iteration are taken as the data category distribution information of the time parameter, thereby effectively ensuring the accuracy of the determination result of the data category distribution information of each time parameter.
[0091] In an implementation, the data categories contained in the time parameter can be iteratively updated based on the plurality of distance information. For example, the data categories contained in the time parameter can be initialized, and the data categories contained in the time parameter can be iteratively updated based on the plurality of distance information and a preset distance threshold between different data categories to obtain the data category distribution information of the time parameter. For example, in each iteration, the inter-class distance between each data category contained in the time parameter can be determined based on each distance information, and the data categories can be merged based on a comparison result of the inter-class distance and the preset distance threshold to obtain an update result of the data categories contained in the time parameter in the current iteration.
[0092] In the current iteration, the inter-class distance between each data category contained in the time parameter can be determined based on the plurality of distance information.
[0093] When there is an inter-class distance less than or equal to a preset distance threshold, the data in a target data category contained in the time parameter can be merged to obtain an update result of the data categories contained in the time parameter in the current iteration, and the target data category includes data categories with an inter-class distance less than or equal to the preset distance threshold.
[0094] Specifically, the current iteration can be any iteration in the process of iteratively updating the data categories contained in the time parameter. In the current iteration, the inter-class distance between the data categories contained in the time parameter can be determined based on the plurality of distance information. Wherein, for any two data categories, the two data categories respectively contain m first data and n second data, when m = n = 1, that is, the two data categories each contain one data, the distance information between the first data and the second data can be taken as the inter-class distance between the two data categories; when m > 1 and / or n > 1, that is, at least one of the two data categories contains multiple data, the distance information between each first data and each second data can be obtained respectively as target distance information, and the mean of each target distance information can be taken as the inter-class distance between the two data categories.
[0095] In implementation, the inter-class distance between the data categories can be compared with a preset distance threshold. When the inter-class distance between the data categories is greater than the preset distance threshold, it indicates that the iteration termination condition is met, and the data categories contained in the time parameter in the current iteration can be taken as the data category distribution information of the time parameter.
[0096] When there is an inter-class distance less than or equal to the preset distance threshold in the inter-class distance between the data categories, the data in the target data categories contained in the time parameter can be merged to obtain a new data category. Wherein, the target data categories can include each data category whose inter-class distance with another data category is less than or equal to the preset distance threshold, so that the data categories contained in the time parameter can be iteratively updated quickly and accurately.
[0097] As a preferred implementation, the target data categories can include the two data categories with the smallest inter-class distance, so that the accuracy of the update result of the data categories contained in the time parameter in the current iteration can be effectively ensured.
[0098] It can be understood that after the data in the target data categories contained in the time parameter is merged, it can be determined whether the number of the data categories contained in the time parameter is greater than 1. If it is greater than 1, the next iteration is performed, and if it is equal to 1, it indicates that the iteration termination condition is met, and the update result of the data categories contained in the time parameter in the current iteration can be taken as the data category distribution information of the time parameter.
[0099] In order to improve the accuracy of the determination result of the target value of the time parameter, in an embodiment of the present specification, the determination of the target value of the time parameter based on the detection values of the time parameter at the plurality of historical time points comprises:
[0100] determining a second mean value of the time parameter based on the preferred data of the time parameter;
[0101] obtaining a deviation between the second mean value and a current configuration value of the time parameter, and taking the second mean value as a target value of the time parameter when the deviation is greater than a preset deviation value.
[0102] Specifically, for any time parameter, a second mean value of the time parameter can be determined based on preferred data of the time parameter, the second mean value of the time parameter being a mean value of each data in the preferred data of the time parameter. Due to factors such as data acquisition equipment, environmental interference, etc., there may be abnormal values in the detected values of the time parameter in the target running data. For example, as shown in FIG. 1, the vacuum conversion time length of the vacuum lock 104 is stable around 15 seconds, and there are some abnormal values (around 25 seconds) deviating from the overall level. If the abnormal values are not removed, the target value calculated is about 17 seconds, which does not conform to the actual running state of the semiconductor process equipment. After removing the abnormal values, the target value calculated is about 15 seconds, which conforms to the actual running state of the semiconductor process equipment. Thus, in the process of determining the target value of the time parameter based on the second mean value of the time parameter, the influence of abnormal values on the determination result of the target value can be effectively avoided, so that the determination result of the target value is more in line with the actual running state of the semiconductor process equipment, thereby ensuring the accuracy of the determination result of the target value. Figure 3
[0103] In implementation, a deviation between the second mean value of the time parameter and a current configuration value of the time parameter can be obtained, and the deviation is compared with a preset deviation value to determine the target value of the time parameter according to the comparison result. The deviation can be an absolute value of a difference between the second mean value of the time parameter and the current configuration value of the time parameter.
[0104] When the deviation is greater than the preset deviation value, it indicates that the time length of the module corresponding to the time parameter to perform the corresponding action in the actual running process has changed significantly, and the scheduling algorithm cannot guarantee the effectiveness of the scheduling result when performing equipment scheduling based on the current configuration value of the time parameter. At this time, the second mean value of the time parameter can be taken as the target value of the time parameter, i.e., the target value of the time parameter indicates that the current configuration value of the time parameter is unreasonable, and the configuration value of the time parameter can be updated based on the target value of the time parameter for the scheduling algorithm to perform equipment scheduling, thereby ensuring the effectiveness of the scheduling result of the scheduling algorithm, and further improving the production efficiency of the semiconductor process equipment.
[0105] It can be understood that when the deviation is less than or equal to the preset deviation value, it indicates that the time length of the module corresponding to the time parameter in the actual running process to perform the corresponding action has not changed significantly, at this time, the current configuration value of the time parameter can be taken as the target value of the time parameter, that is, the target value of the time parameter represents that the current configuration value of the time parameter is reasonable, and the configuration value of the time parameter does not need to be updated.
[0106] In order to ensure the reliability of the determination result of the change state of each time parameter, in an embodiment of the present specification, the first statistical data of the time parameter includes the variance of the time parameter and / or the range of the time parameter.
[0107] Based on the first statistical data of the time parameter, it is determined whether the change state of the time parameter is a stable state, including:
[0108] Based on the comparison result of the variance of the time parameter and the preset variance value, and / or, based on the comparison result of the range of the time parameter and the preset range value, it is determined whether the change state of the time parameter is a stable state.
[0109] Specifically, for any time parameter, the variance of the time parameter includes the variance of the detection value of the time parameter at a plurality of historical time points in the target running data. The range of the time parameter can include the difference between the maximum value and the minimum value of the detection value of the time parameter at a plurality of historical time points.
[0110] The first statistical data of the time parameter can include the variance of the time parameter and / or the range of the time parameter, and when determining the change state of the time parameter based on the first statistical data of the time parameter, the change state of the time parameter can be determined based on the comparison result of the variance of the time parameter and the preset variance value, and / or, the comparison result of the range of the time parameter and the preset range value.
[0111] For example, when the variance of the time parameter is greater than the preset variance value, and / or, the range of the time parameter is greater than the preset range value, it is determined that the change state of the time parameter is a fluctuation state; when the variance of the time parameter is less than or equal to the preset variance value, and the range of the time parameter is less than or equal to the preset range value, it is determined that the change state of the time parameter is a stable state, thereby effectively ensuring the reliability of the determination result of the change state of the time parameter.
[0112] In a feasible embodiment, it further includes:
[0113] When the change state of the time parameter is a fluctuation state, or the data category distribution information represents that the number of data categories contained in the time parameter is greater than the predetermined value, a prompt information is generated.
[0114] Specifically, for any time parameter, when the change state of the time parameter is the fluctuation state, it indicates that the stability of the detection values of the time parameter at multiple historical time points is poor, that is, there is a risk of hardware failure (such as wear and tear) or a risk of software operation anomaly of the module corresponding to the time parameter in the semiconductor process equipment, at this time, the first prompt information can be generated, the first prompt information can include the detection values of the time parameter at multiple historical time points, and can also include the first statistical data of the time parameter, to prompt the relevant personnel to troubleshoot, to ensure the stability of the time length of the semiconductor process equipment in the running process to perform the corresponding action, thereby effectively reducing the influence of the instability of the time parameter on the accuracy of the update result of the target value of the time parameter, and further ensuring the production efficiency of the semiconductor process equipment.
[0115] When the data category distribution information of the time parameter indicates that the number of data categories contained by the time parameter is greater than a predetermined value, it indicates that there is a risk of unreasonable configuration of the data interface of the time parameter. For example, as shown in Figure 4 As shown, if the door closing time length of the process chamber 101 is concentrated at two different modal values of 2 seconds and 5 seconds at multiple historical time points, the target value of the time parameter can be obtained by taking the average of the detection values of the door closing time length of the process chamber 101 at multiple historical time points, which is about 3.5 seconds, which does not match the actual running state of the semiconductor process equipment, thereby unable to ensure the effectiveness of the scheduling result of the scheduling algorithm. It can be understood that the method of taking the average of the detection values of the door closing time length of the process chamber 101 at multiple historical time points can adopt the existing average calculation method, which is not limited here, for example, the average of each detection value can be directly taken, the average can also be taken after removing the abnormal values in each detection value, and the average can also be taken after removing the extreme values in each detection value. At this time, the second prompt information can be generated, the second prompt information can include the number of data categories contained by the time parameter, and can also include the data contained by each data category in the preferred data of the time parameter and the average of the data contained by each data category, to prompt the relevant personnel to optimize the configuration of the data interface, for example, the door closing time length of the process chamber 101 can be subdivided into two time parameters, such as the door closing time length (2 seconds) of the vacuum mechanical hand 102 after placing the wafer in the process chamber 101 and the door closing time length (5 seconds) of the vacuum mechanical hand 102 after taking the wafer from the process chamber 101, thereby effectively avoiding the influence of unreasonable configuration of the data interface of the time parameter on the accuracy of the update result of the target value of the time parameter, and further ensuring the production efficiency of the semiconductor process equipment.
[0116] In addition, after the target values of the time parameters are determined, a data analysis report can be automatically generated and outputted, and the data analysis report can include an original data list, a fluctuation data list, a multi-class distribution data list, and a configuration error data list, so as to monitor the stability, time distribution characteristics, and configuration correctness of the time parameters. The original data list can include the time parameters, the fluctuation data list can include the time parameters in the fluctuation state, the multi-class distribution data list can include the time parameters whose number of data classes is greater than a predetermined value, and the configuration error data list can include the time parameters whose second mean value deviates from the current configuration value by more than a preset deviation value.
[0117] The specific implementation process of the time parameter updating method is described in detail below according to an optional embodiment. As shown in FIG. 5, Figure 5
[0118] S501, obtaining a data list corresponding to each time parameter by analyzing scheduling algorithm action time data recorded in a log file, a database, or a memory; wherein the data list includes detection values of the corresponding time parameter at a plurality of historical time points;
[0119] S502, traversing the data list corresponding to each time parameter;
[0120] S503, determining first statistical data of the current time parameter based on the data list corresponding to the current time parameter, and determining a change state of the current time parameter based on the first statistical data of the current time parameter; if the change state of the current time parameter is a stable state, executing step S505, and if the change state of the current time parameter is a fluctuation state, executing step S504;
[0121] S504, generating a first prompt information to prompt that the current time parameter has a large fluctuation, and the hardware and / or software problems need to be investigated, and executing step S510;
[0122] S505, determining second statistical data of the current time parameter based on the data list corresponding to the current time parameter, and identifying and removing abnormal values in the data list corresponding to the current time parameter based on the second statistical data to obtain preferred data of the current time parameter;
[0123] S506, determining data class distribution information of the current time parameter based on the preferred data of the current time parameter; if the data class distribution information of the current time parameter indicates that the number of data classes contained by the current time parameter is greater than 1, executing step S507, and if the data class distribution information of the current time parameter indicates that the number of data classes contained by the current time parameter is equal to 1, executing step S508;
[0124] S507, generate second prompt information to prompt that the current time parameter contains multiple data categories, and transmit the time parameter conforming to the actual distribution to the scheduling algorithm to perform step S510;
[0125] S508, determine the second mean value of the current time parameter based on the preferred data of the current time parameter, and obtain the deviation of the second mean value of the current time parameter from the current configuration value of the current time parameter, if the deviation is greater than a preset deviation value, perform step S509, if the deviation is less than or equal to the preset deviation value, perform step S510;
[0126] S509, take the second mean value as the target value of the current time parameter and transmit it to the scheduling algorithm;
[0127] S510, determine whether the time parameter has ended, if yes, the program ends, if not, perform step S502.
[0128] For example, taking the preferred data of the current time parameter as [2.0, 2.1, 2.15, 5.0, 5.21], the method for determining the data category distribution information of the current time parameter in step S506 is illustrated. The normalized processing result of the preferred data of the current time parameter is X = [0.38, 0.4, 0.41, 0.96, 1], and the preset distance threshold is set to 0.1. The method for determining the data category distribution information of the current time parameter can include:
[0129] (a) initialize each data in X as a data category, respectively G i = {x i}, i = 1, 2,.., 5.
[0130] (b) calculate the distance information between each two data in X to obtain the distance matrix D, which can be shown as formula (1):
[0131]
[0132] In the formula, d ij is the distance information between the i-th data and the j-th data in X, j = 1, 2,.., 5.
[0133] (c) calculate the inter-class distance between each two of the five data categories according to the distance matrix D. Since each data in X is a data category, the inter-class distance between the data categories is the distance information between the corresponding data, that is, the smallest inter-class distance between each data category is D 23 = 0.01, which is less than the preset distance threshold 0.1, wherein D 23 is the inter-class distance between G2 and G3, therefore, G2 and G3 are merged into a new data category G6, denoted as G6 = {x2, x3}.
[0134] (d) Calculate the inter-class distance between G6 and G1, G4 and G5 respectively, D 61 = 0.025, D 64 = 0.555, D 65 = 0.595, in addition, the inter-class distance D 14 = 0.58 between G1 and G4, the inter-class distance D 15 = 0.62 between G1 and G5, and the inter-class distance D 45 = 0.04 between G4 and G5, wherein D 61 is the smallest inter-class distance, and D 61 is less than the preset distance threshold 0.1, therefore, G6 and G1 are merged into a new data class G7, denoted as G7 = {x1, x2, x3}.
[0135] (e) Calculate the inter-class distance between G7 and G4 and G5 respectively, D 74 = 0.56, D 75 = 0.6, and the inter-class distance D 45 = 0.04 between G4 and G5, wherein D 45 is the smallest inter-class distance, and D 45 is less than the preset distance threshold 0.1, therefore, G4 and G5 are merged into a new data class G8, denoted as G8 = {x4, x5}.
[0136] (f) Calculate the inter-class distance D 78 = 0.58 between G7 and G8, D 78 is greater than the preset distance threshold 0.1, which does not satisfy the merging condition, and the iteration ends, thereby obtaining two data classes G7 = {x1, x2, x3} and G8 = {x4, x5}, and the number of data classes contained in the current time parameter can be output as 2.
[0137] Exemplary apparatus
[0138] The embodiments of the present specification also provide a computing device comprising at least one processor and at least one memory, the memory having stored therein a computer program which, when executed by the processor, implements the time parameter updating method according to any one of the above embodiments.
[0139] Exemplary computer program product and storage medium
[0140] In addition to the methods and apparatus described above, the time parameter updating method provided by the embodiments of the present disclosure can also be a computer program product, which includes computer program instructions that, when executed by a processor, cause the processor to perform the steps in the time parameter updating method according to various embodiments of the present disclosure described in the above "Exemplary Method" section of the present disclosure.
[0141] The computer program product can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, etc., and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server.
[0142] In addition, the embodiments of the present disclosure also provide a computer readable storage medium, which stores a computer program. The computer program causes a processor to perform the steps in the time parameter updating method according to various embodiments of the present disclosure described in the above "Exemplary Method" section of the present disclosure.
[0143] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium and can include the processes of the above-mentioned embodiments when executed. Any reference to memory, storage, databases, or other media in the embodiments provided by the present disclosure can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0144] Any combination of the technical features in the above embodiments can be made, and for the sake of brevity, not all possible combinations are described, however, it is to be understood that the scope of protection includes all possible combinations of the technical features.
[0145] The above embodiments only express several implementation manners of the present specification, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the solutions provided by the embodiments of the present specification. It should be pointed out that for ordinary skilled persons in the art, without departing from the concept of the present specification, a number of modifications and improvements can be made, which are within the protection scope of the present specification. Therefore, the protection scope of the present specification patent should be subject to the appended claims.< / double> < / double>
Claims
1. A time parameter updating method, characterized by, The method comprises: determining first statistical data of each time parameter of a semiconductor process equipment based on target operation data of the semiconductor process equipment; the target operation data comprises detected values of each time parameter at multiple historical time points, and the first statistical data is used to represent a variation state of the time parameter; when the variation state of the time parameter is determined to be a stable state based on the first statistical data of the time parameter, data category distribution information of the time parameter is determined based on the detected values of the time parameter at the multiple historical time points; when the data category distribution information represents that the number of data categories contained in the time parameter is less than or equal to a predetermined value, a target value of the time parameter is determined based on the detected values of the time parameter at the multiple historical time points, and the target value of the time parameter is used for equipment scheduling by a scheduling algorithm of the semiconductor process equipment.
2. The method of claim 1, wherein, The method of determining the data category distribution information of the time parameter based on the detected values of the time parameter at the multiple historical time points comprises: determining second statistical data of the time parameter based on the detected values of the time parameter at the multiple historical time points, wherein the second statistical data is used to represent a distribution state of parameter values of the time parameter; determining an abnormal value in the detected values of the time parameter at the multiple historical time points based on the second statistical data, and eliminating the abnormal value to obtain preferred data of the time parameter; determining the data category distribution information of the time parameter based on the preferred data of the time parameter.
3. The method of claim 2, wherein, The second statistical data of the time parameter comprises a first mean value and a standard deviation of the time parameter. The method of determining the abnormal value in the detected values of the time parameter at the multiple historical time points based on the second statistical data comprises: determining a target value interval of the time parameter based on the first mean value and the standard deviation of the time parameter; determining the abnormal value in the detected values of the time parameter at the multiple historical time points based on a comparison result of the detected values of the time parameter at the multiple historical time points and the target value interval.
4. The method of claim 2, wherein, The method of determining the data category distribution information of the time parameter based on the preferred data of the time parameter comprises: obtaining distance information between each data in the preferred data of the time parameter to obtain multiple distance information; iteratively updating data categories contained in the time parameter based on the multiple distance information, and determining the data category distribution information of the time parameter based on a result of the iterative update.
5. The method of claim 4, wherein, The method of iteratively updating the data categories contained in the time parameter based on the multiple distance information comprises: in a current iteration, determining an inter-class distance between each data category contained in the time parameter based on the multiple distance information; when there is an inter-class distance less than or equal to a preset distance threshold, performing merging processing on data in a target data category contained in the time parameter to obtain an update result of the data category contained in the time parameter in the current iteration, and the target data category comprises data categories with an inter-class distance less than or equal to the preset distance threshold.
6. The method of claim 2, wherein, The target value of the time parameter is determined based on the detection values of the time parameter at the plurality of historical time points, comprising: determining a second mean value of the time parameter based on the preferred data of the time parameter; obtaining a deviation between the second mean value and a current configuration value of the time parameter, and taking the second mean value as the target value of the time parameter when the deviation is greater than a preset deviation value.
7. The method of claim 1, wherein, The first statistical data of the time parameter comprises a variance of the time parameter and / or a range of the time parameter; determining whether the change state of the time parameter is a stable state based on the first statistical data of the time parameter, comprising: determining whether the change state of the time parameter is a stable state based on a comparison result of the variance of the time parameter and a preset variance value, and / or based on a comparison result of the range of the time parameter and a preset range value.
8. The method according to any one of claims 1 to 7, characterized in that, Further comprising: generating a prompt information when the change state of the time parameter is a fluctuation state, or when the data category distribution information represents that the number of data categories contained in the time parameter is greater than the predetermined value.
9. A computing device, comprising: The computer program is stored in the computer readable storage medium and is executed by the processor to implement the time parameter updating method according to any one of claims 1 to 8.
10. A computer-readable storage medium, characterized in that, The computer program is stored in the computer readable storage medium and is executed by the processor to implement the time parameter updating method according to any one of claims 1 to 8.