Methods, systems, equipment, and storage media for determining the influence of auxiliary material parameters

Abstract

This application discloses a method, system, device, and storage medium for determining the influence of auxiliary material parameters, relating to the field of semiconductor analysis technology. The method includes: acquiring target auxiliary material parameters and corresponding cutting results, whereby the target auxiliary material parameters characterize historical parameter values ​​of the target auxiliary material; dividing the value range of the target auxiliary material parameters into multiple first intervals; merging adjacent first intervals based on the distribution of cutting results within each first interval to obtain multiple second intervals; and determining the influence curve of the target auxiliary material parameters on the cutting results based on the center value of the target auxiliary material parameters within each second interval and the median value of the cutting results, whereby the influence curve is used to determine the influence of the target auxiliary material parameters on the cutting results. This application, through dynamic binning and noise suppression of the data, can accurately depict the stage-wise influence curve of the auxiliary material parameters on the cutting results, improving the accuracy of determining the stage-wise influence of the auxiliary material parameters on the cutting results.

Description

Technical Field

[0001] This application relates to the field of semiconductor analysis technology, specifically to a method, system, device, and storage medium for determining the influence of auxiliary material parameters. Background Technology

[0002] In silicon wafer dicing, parameters of auxiliary materials such as dicing fluid have a significant impact on the final dicing result. This impact typically exhibits a convex or concave function trend and displays a "stage-based" characteristic; that is, parameter changes have a relatively small effect on the result within a given range, but significant changes may occur when crossing that range. Therefore, it is necessary to study and extract the stage-based impact curves of these parameters on the dicing result. However, due to environmental noise interference and the large errors of existing data acquisition devices, this stage-based pattern may be masked, making it difficult to accurately define the stage-based impact of auxiliary material parameters on the dicing result. Summary of the Invention

[0003] The embodiments of this application provide a method, system, device, and storage medium for determining the influence of auxiliary material parameters, so as to improve the accuracy of determining the staged influence of auxiliary material parameters on cutting results.

[0004] To address the aforementioned technical problems, embodiments of this application disclose the following technical solutions:

[0005] In a first aspect, a method for determining the influence of excipient parameters is provided, including: obtaining target excipient parameters and cutting results corresponding to the target excipient parameters, wherein the target excipient parameters are used to characterize the historical parameter values ​​of the target excipient;

[0006] The range of values ​​for the target excipient parameters is divided into multiple first intervals;

[0007] Based on the distribution of the cutting results in each first interval, adjacent first intervals are merged to obtain multiple second intervals;

[0008] Based on the center value of the target auxiliary material parameter located in each second interval and the median value of the cutting result, an influence curve of the target auxiliary material parameter on the cutting result is determined, and the influence curve is used to determine the influence of the target auxiliary material parameter on the cutting result.

[0009] In some embodiments, the method for obtaining a plurality of second intervals includes:

[0010] Based on the distribution of the cutting results located in each of the first intervals, the similarity between adjacent first intervals is determined;

[0011] Starting from the first first interval, adjacent first intervals with similarity less than the similarity threshold are merged one by one to obtain multiple second intervals.

[0012] In some embodiments, merging adjacent first intervals with similarity less than a similarity threshold one by one to obtain a plurality of second intervals includes:

[0013] If the similarity between the first interval and the adjacent first interval is less than the similarity threshold, then the first interval and the adjacent first interval are merged into a new first interval, and the operation of comparing the similarity between the new first interval and the adjacent first interval with the similarity threshold is repeated until the similarity between the new first interval and the adjacent first interval is greater than or equal to the similarity threshold. Then the new first interval is determined as the second interval, and the comparison result of the similarity between the adjacent first interval and the next adjacent first interval with the similarity threshold is determined.

[0014] In some embodiments, a method for determining the distribution of the cutting results located in each of the first intervals includes:

[0015] Determine the normal and abnormal proportions of the cutting results located in each of the first intervals;

[0016] Based on the normal proportion and the abnormal proportion, the distribution of the cutting results located in the corresponding first interval is determined.

[0017] In some embodiments, the method for determining the normal percentage and abnormal percentage of the cutting results located in each of the first intervals includes:

[0018] The cutting results corresponding to the target auxiliary material parameters are subjected to binary classification processing based on the preset target yield threshold to obtain the binary classification processing result.

[0019] Based on the binary classification results and the total number of the cutting results within the first interval, the normal and abnormal proportions of the cutting results located in the first interval are determined.

[0020] In some embodiments, dividing the value range of the target excipient parameter into multiple first intervals includes:

[0021] The range of values ​​for the target auxiliary material parameters is divided into multiple first intervals according to the equal-width rule.

[0022] In some embodiments, the median value of the cutting results includes the median or average value of the cutting results.

[0023] Secondly, a system for determining the influence of excipient parameters is provided, including:

[0024] The acquisition module is used to acquire the target auxiliary material parameters and the cutting results corresponding to the target auxiliary material parameters, wherein the target auxiliary material parameters are used to characterize the historical parameter values ​​of the target auxiliary material;

[0025] The interval division module is used to divide the value range of the target excipient parameter into multiple first intervals;

[0026] The interval merging module is used to merge adjacent first intervals according to the distribution of the cutting results located in each first interval, to obtain multiple second intervals;

[0027] The influence curve determination module is used to determine the influence curve of the target auxiliary material parameter on the cutting result based on the center value of the target auxiliary material parameter located in each second interval and the median value of the cutting result. The influence curve is used to determine the influence of the target auxiliary material parameter on the cutting result.

[0028] Thirdly, an electronic device is provided, the electronic device including a processor and a memory; the memory stores a computer program, and the processor is used to execute the computer program stored in the memory to implement the method for determining the influence of auxiliary material parameters as described in any of the first aspects.

[0029] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the method for determining the influence of auxiliary material parameters as described in any of the first aspects.

[0030] One of the above technical solutions has the following advantages or beneficial effects:

[0031] Compared with the prior art, the present application provides a method for determining the influence of excipient parameters, comprising: obtaining target excipient parameters and corresponding cutting results, wherein the target excipient parameters are used to characterize the historical parameter values ​​of the target excipient; dividing the value range of the target excipient parameters into multiple first intervals; merging adjacent first intervals according to the distribution of cutting results in each first interval to obtain multiple second intervals; and determining the influence curve of the target excipient parameters on the cutting results based on the center value of the target excipient parameters in each second interval and the median value of the cutting results, wherein the influence curve is used to determine the influence of the target excipient parameters on the cutting results. The method for determining the influence of auxiliary material parameters provided in this application divides the range of target auxiliary material parameters into intervals and merges the intervals based on the distribution of the cutting results. This achieves dynamic binning and noise suppression of the data. Based on the center value of the target auxiliary material parameter in the merged interval and the median value of the cutting results, the method accurately depicts the stage-by-stage influence curve of the auxiliary material parameter on the cutting results. This improves the accuracy of determining the stage-by-stage influence of auxiliary material parameters on the cutting results, which in turn facilitates the precise optimization of auxiliary material ratios and the control of cutting yield, reduces the cost of auxiliary material usage, and improves the stability of the cutting process.

[0032] This application discloses a system for determining the influence of auxiliary material parameters. By dividing the range of target auxiliary material parameters into intervals and merging the intervals based on the distribution of the cutting results, the system achieves dynamic binning and noise suppression of the data. Based on the center value of the target auxiliary material parameter in the merged interval and the median value of the cutting results, the system accurately depicts the stage-by-stage influence curve of the auxiliary material parameters on the cutting results. This improves the accuracy of determining the stage-by-stage influence of auxiliary material parameters on the cutting results, thereby facilitating precise optimization of auxiliary material ratios and cutting yield control, reducing auxiliary material usage costs, and improving the stability of the cutting process. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart of a method for determining the influence of excipient parameters provided in the embodiments of this application;

[0035] Figure 2 This is a schematic diagram of the fitting results of the effect of cutting fluid turbidity on jumper wires and break seams provided in the embodiments of this application;

[0036] Figure 3 This is a schematic diagram of the structure of the excipient parameter influence determination system according to an embodiment of this application;

[0037] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application;

[0038] Figure label:

[0039] 100 - Influence determination system for auxiliary material parameters; 101 - Acquisition module; 102 - Interval division module; 103 - Interval merging module; 104 - Influence curve determination module; 501 - Memory; 502 - Processor. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0041] In the description of this application, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, and "at least one" can mean one, two, or more, unless otherwise explicitly specified.

[0042] Figure 1 This is a flowchart illustrating a method for determining the influence of auxiliary material parameters, as provided in an embodiment of this application. This method is applicable to semiconductor silicon wafer analysis systems, aiming to improve the accuracy of determining the stage-specific influence of auxiliary material parameters on the cutting results. This method can be executed by an auxiliary material parameter influence determination system, which can be implemented in software and / or hardware and can be configured within the processor of the semiconductor analysis system. Please refer to... Figure 1 The method includes the following steps:

[0043] Step 110: Obtain the target auxiliary material parameters and the corresponding cutting results. The target auxiliary material parameters are used to characterize the historical parameter values ​​of the target auxiliary material.

[0044] Among them, the target auxiliary material parameters refer to the parameters of various cutting auxiliary materials in the silicon wafer cutting process, such as the conductivity, pH value, concentration, turbidity, temperature, chemical oxygen demand (COD), surface tension, and single-blade liquid supply of cutting fluid and other auxiliary materials. The specific parameters can be set according to the actual situation, and no specific limitations are made here.

[0045] Since the performance of cutting materials directly affects cutting quality, the setting of material parameters directly impacts the quality of the cutting result, such as whether the cutting is normal or abnormal. Therefore, the cutting result corresponding to the target material parameters refers to the cutting result obtained by cutting according to the cutting process set according to the target material parameters. The cutting result corresponding to the target material parameters directly reflects the influence of the target material parameter settings on the cutting quality. For example, assuming the target material parameter is the concentration of the cutting fluid, and the concentration of the cutting fluid is set to M, then cutting according to the cutting fluid concentration M will yield the corresponding cutting result.

[0046] The target auxiliary material parameters are used to characterize the historical parameter values ​​of the target auxiliary material. Correspondingly, the cutting results corresponding to the target auxiliary material parameters are used to characterize the cutting results corresponding to each historical parameter value of the target auxiliary material. The historical parameter values ​​of the target auxiliary material refer to the target auxiliary material parameter values ​​used for cutting within a historical time period (e.g., the past month, a quarter, etc., which can be set according to actual conditions). For example, taking the cutting fluid turbidity as an auxiliary material parameter, the historical parameter value of the cutting fluid turbidity refers to the cutting fluid turbidity corresponding to each of the set cutting fluid turbidities within a historical time period (e.g., the past quarter). For example, if the cutting fluid turbidity is set to X1, X2, X3, ... Xn within the past quarter, and cutting processes are performed accordingly, the corresponding cutting results for each cutting fluid turbidity X1, X2, X3, ... Xn will be obtained. Here, X1, X2, X3, ... Xn are the historical parameter values ​​of the cutting fluid turbidity within the past quarter.

[0047] Step 120: Divide the range of values ​​for the target auxiliary material parameters into multiple first intervals.

[0048] The value range of the target excipient parameters is generally set according to the actual business needs. For example, taking a certain project as an example, the value range of each excipient parameter is as follows: raw material addition ratio: (0, 1); replacement ratio: (0.02, 0.12); ionic liquid addition ratio: (0.0, 0.15); citric acid: (0.0, 0.2); small component GB: (0.0, 0.2); small component HT: (0.0, 0.2); small component AC: (0.0, 0.2); small component small component P: (0.0, 0.2); small component small component D: (0.0, 0.2); small component P4: (0.0, 0.2); small component small component T: (0.0, 0.2); small component triethanolamine: (0.0, 0.2); single-blade liquid supply volume: (1.0, 10.0).

[0049] Specifically, the value range of the target auxiliary material parameters is divided into multiple first intervals to achieve adaptive binning of the target auxiliary material parameters, reduce or eliminate noise interference, and facilitate subsequent accurate characterization of the phased impact of the target auxiliary material parameters on the cutting results. For example, the turbidity value range of the cutting fluid over the past quarter, X1, X2, X3, ... Xn, is divided into multiple first intervals. The specific number of first intervals can be set according to the actual situation and is not specifically limited here.

[0050] In some embodiments, dividing the range of values ​​for the target excipient parameter into multiple first intervals includes: dividing the range of values ​​for the target excipient parameter into multiple first intervals according to an equal-width rule.

[0051] Specifically, the value range of the target auxiliary material parameters is divided into multiple first intervals according to the equal-width rule to achieve adaptive binning of the target auxiliary material parameters, reduce or eliminate noise interference, and facilitate subsequent accurate characterization of the stage-by-stage impact of the target auxiliary material parameters on the cutting results. The number of intervals and the width of the intervals divided according to the equal-width rule can be set according to actual conditions and are not specifically limited here.

[0052] For example, the specific process of dividing the value range of the target excipient parameter is as follows: Taking the target excipient parameter X as an example, the value range of the target excipient parameter X is divided into N first intervals according to the equal width rule:

[0053] ;

[0054] in, and For the first The endpoints of each interval; N is a positive integer. For example, N is 100. The value of N can be set according to the actual situation, and no specific limitation is made here.

[0055] The interval division width is:

[0056] ;

[0057] in, Define the width of the interval; The maximum value within the range of the target excipient parameter X; The minimum value within the range of values ​​for the target auxiliary material parameter X.

[0058] For example, if the target excipient parameter X has a value range of [0, 1] and N is 100, then the first interval is [0, 0.01].

[0059] Step 130: Based on the distribution of the cutting results in each first interval, merge adjacent first intervals to obtain multiple second intervals.

[0060] Since the target excipient parameters are used to characterize the historical parameter values ​​of the target excipient, the corresponding cutting results are used to characterize the cutting results corresponding to each historical parameter value of the target excipient. Therefore, each value of the target excipient parameter corresponds to a corresponding cutting result. Similarly, each value within each first interval obtained by dividing the value range of the target excipient parameters also corresponds to a corresponding cutting result. Thus, after dividing into multiple first intervals, merging adjacent first intervals according to the distribution of cutting results within each first interval helps to reduce or eliminate the impact of data acquisition errors and noise interference on the determination of the excipient parameters, thereby improving the accuracy of determining the staged impact of the excipient parameters on the cutting results.

[0061] In some embodiments, the method for obtaining a plurality of second intervals includes the following steps:

[0062] Step 1: Determine the similarity between adjacent first intervals based on the distribution of the segmentation results in each first interval.

[0063] Specifically, after dividing the data into multiple first intervals, the similarity between adjacent first intervals is determined based on the distribution of the cutting results within each first interval. This facilitates the dynamic merging of adjacent first intervals, thereby reducing or eliminating the impact of data acquisition errors and noise interference on the determination of auxiliary material parameters, and ultimately improving the accuracy of determining the stage-specific impact of auxiliary material parameters on the cutting results.

[0064] In some embodiments, the method for determining the distribution of cutting results in each first interval includes: determining the normal proportion and the abnormal proportion of cutting results in each first interval; and determining the distribution of cutting results in the corresponding first interval based on the normal proportion and the abnormal proportion.

[0065] The cutting results include normal and abnormal cuts. Correspondingly, the distribution of cutting results is related to the proportion of normal and abnormal cuts. Therefore, determining the proportion of normal and abnormal cuts in each first interval can further determine the distribution of cutting results in each first interval.

[0066] In some embodiments, the method for determining the normal and abnormal proportions of cutting results located in each first interval includes: performing binary classification processing on the cutting results corresponding to the target auxiliary material parameters according to a preset target yield threshold to obtain a binary classification processing result; and determining the normal and abnormal proportions of cutting results located in the first interval based on the binary classification processing result and the total number of cutting results in the first interval.

[0067] Specifically, the cutting results corresponding to the target excipient parameters are transformed into a binary classification problem. That is, the cutting results corresponding to the target excipient parameters are subjected to binary classification according to a preset target yield threshold to obtain the binary classification result. For example, the abnormality rate of the cutting results... Transform it into a binary classification problem:

[0068] ;

[0069] in, This is a preset target yield threshold. For example, It is 0.0034, in addition Other values ​​are also possible, and the specific values ​​can be set according to the actual situation. No specific restrictions are made here.

[0070] It should be noted that the dicing yield of this application can be expressed by various different indicators. For example, Total Thickness Variation (TTV) failure rate, line mark failure rate, etc. Taking TTV failure rate as an example, TTV represents the difference between the maximum and minimum thickness of the silicon wafer and is an important indicator for measuring the uniformity of silicon wafer thickness. In semiconductor manufacturing, the thickness of the silicon wafer must be highly uniform across the entire surface. A single cut can typically produce four to five thousand silicon wafers. Inspection will detect wafers with TTV failures. The number of TTV-qualified wafers divided by the total number of wafers can be defined as the TTV pass rate. If the TTV failure rate for that cut is greater than 0.34%, it is marked as 1; otherwise, it is marked as 0.

[0071] Specifically, the method for determining the normal and abnormal proportions of the segmentation results within the first interval based on the binary classification results and the total number of segmentation results within the first interval includes: for each first interval Calculate the normal percentage (i.e., the percentage of category 0) within each first interval. And the percentage of abnormal cases (i.e., the percentage of category 1). The details are as follows:

[0072] ;

[0073] ;

[0074] in, y is an exponential function representing the abnormality rate y of the target variable segmentation result.

[0075] For example, taking the turbidity of the cutting fluid as an example, let the range of the turbidity of the cutting fluid be [0, 1]. Suppose that 100 cuts were made using the cutting fluid with this turbidity range, of which 30 cuts had a TTV pass rate higher than 0.34% and 70 cuts had a TTV pass rate lower than 0.34%. Then the abnormal percentage of the cutting results is 30%, i.e., Pi(1) = 30%, and the normal percentage is 70%, i.e., Pi(0) = 70%.

[0076] Step 2: Starting from the first first interval, merge adjacent first intervals with similarity less than the similarity threshold one by one to obtain multiple second intervals.

[0077] Specifically, assuming the range of values ​​for the target excipient parameter is divided into n first intervals, the similarity of these n first intervals is calculated. Starting from the first first interval, the similarity between the first and second first intervals is compared. If the similarity between the first and second first intervals is less than the similarity threshold, the first and second first intervals are merged, and then compared with the next first interval. If they are similar, the merging continues; if they are not similar, a new round of similarity comparison and merging begins from the dissimilar first interval. In this way, adjacent first intervals are merged one by one to obtain multiple second intervals.

[0078] In some embodiments, adjacent first intervals with similarity less than a similarity threshold are merged one by one to obtain multiple second intervals, including: if the similarity between a first interval and an adjacent first interval is less than a similarity threshold, the first interval and the adjacent first interval are merged into a new first interval, and the operation of comparing the similarity between the new first interval and the adjacent first interval with the similarity threshold is repeatedly performed until the similarity between the new first interval and the adjacent first interval is greater than or equal to the similarity threshold, then the new first interval is determined as a second interval, and the comparison result of the similarity between the adjacent first interval and the next adjacent first interval with the similarity threshold is further determined.

[0079] The rule for merging adjacent first intervals one by one is as follows: For adjacent first intervals and If their 0 and 1 distributions are similar, they are merged into a new interval. The similarity measure... for:

[0080] ;

[0081] like Then merge the intervals and recalculate the 1-distribution of the merged intervals. :

[0082] ;

[0083] in, This is the similarity threshold. The specific value can be set according to the actual situation, and no specific limit is set here.

[0084] in, For the first interval The number of data points.

[0085] If the similarity between adjacent first intervals is less than the similarity threshold, it indicates that the adjacent first intervals are relatively similar. If the similarity between adjacent first intervals is greater than or equal to the similarity threshold, it indicates that the adjacent first intervals are not similar. Therefore, merging adjacent first intervals with high similarity, that is, merging first intervals with similar anomaly proportions (proportion of category 1), is beneficial for subsequent stage analysis of the impact of target auxiliary material parameters on the cutting results, and improves the accuracy of determining the stage impact of auxiliary material parameters on the cutting results.

[0086] For example, suppose the range of values ​​for the target excipient parameter is divided into n first intervals. The similarity of these n first intervals is calculated. Starting with the first first interval, the similarity between the first and second first intervals is compared. If the similarity between the first and second first intervals is less than a similarity threshold, the first and second first intervals are merged to obtain a new first interval. The similarity between this new first interval and the next first interval (i.e., the third first interval) is then compared with the similarity threshold. If they are similar, the merging continues to obtain another new first interval, which is then compared with the next first interval (i.e., the fourth first interval). If the new first interval obtained by merging the first and second first intervals is not similar to the third first interval, this new first interval is taken as the second interval, and a new round of similarity comparison and merging begins from the third first interval. This process is repeated multiple times until the nth first interval is compared and merged, resulting in multiple second intervals.

[0087] Step 140: Based on the center value of the target auxiliary material parameter in each second interval and the median value of the cutting result, determine the influence curve of the target auxiliary material parameter on the cutting result. The influence curve is used to determine the influence of the target auxiliary material parameter on the cutting result.

[0088] Since each first interval is formed by merging adjacent first intervals with similar abnormal proportions, the stage-specific impact of the target auxiliary material parameters on the cutting results can be accurately characterized based on the target auxiliary material parameters of each second interval and the corresponding cutting results of each second interval, thereby improving the accuracy of determining the stage-specific impact of auxiliary material parameters on the cutting results.

[0089] The center value of the target excipient parameter in each second interval is the center point of the range of values ​​for the target excipient parameter in that second interval. For example, the range of values ​​for a certain second interval is: Then the center point of the range of values ​​in the second interval. for:

[0090] ;

[0091] The median value of the segmentation results for each second interval is either the median or the average value of the segmentation results.

[0092] Specifically, a dataset is generated based on the center value of the target auxiliary material parameter and the median value of the cutting result within each second interval. The generated dataset is then fitted to obtain the influence curve of the target auxiliary material parameter on the cutting result. The fitting method can be polynomial fitting or spline interpolation fitting, etc.

[0093] For example, the median value of the segmentation results for each second interval is used as the median of the segmentation results. For example, the center point of the target auxiliary material parameter value range for each second interval and the median of the cutting results are... The generated dataset is:

[0094] ;

[0095] The dataset was fitted to obtain the influence curve of the target auxiliary material parameters on the cutting results. for:

[0096] ;

[0097] Therefore, by dividing the target auxiliary material parameter's value range into intervals through the above steps, adaptive binning is achieved. The similarity of the first interval is calculated based on the distribution of cutting results within that interval. Intervals with high similarity (i.e., intervals with similar anomaly rates) are dynamically merged to reduce or eliminate the impact of data acquisition errors and noise interference on the determination of auxiliary material parameters. This facilitates a more accurate characterization of the stage-specific impact of target auxiliary material parameters on the cutting results. Furthermore, an impact curve of the target auxiliary material parameters on the cutting results is obtained by fitting the median value of the target auxiliary material parameters and the median value of the cutting results within the merged second interval. This impact curve allows for precise analysis of the stage-specific impact of target auxiliary material parameters on the cutting results, thereby improving the accuracy of determining the stage-specific impact of auxiliary material parameters on the cutting results. Ultimately, this improves silicon wafer cutting yield and process stability, while reducing auxiliary material usage costs.

[0098] Figure 2 This is a schematic diagram showing the fitting results of the effect of cutting fluid turbidity on jumpers and breakage seams provided in the embodiments of this application. For example, see [link to relevant documentation]. Figure 2 L1 is the fitted curve of the effect of cutting fluid turbidity on jumpers, L2 is the fitted curve of the effect of cutting fluid turbidity on fractures, and L3 is the fitted curve of the effect of cutting fluid turbidity on both jumpers and fractures. For example, taking the effect of cutting fluid turbidity on jumpers as an example, referring to curve L1, it can be seen that when the cutting fluid turbidity is set in the range of 30-40, the cutting abnormality rate is relatively low, while when the cutting fluid turbidity is set in the range of 70-80, the cutting abnormality rate is relatively high. Therefore, the influence curve of the target auxiliary material parameter on the cutting result can be obtained through the above-mentioned auxiliary material parameter influence determination method of this application. Furthermore, the influence curve can be used to analyze the stage-by-stage influence of the target auxiliary material parameter on the cutting result, thereby improving the accuracy of determining the stage-by-stage influence of auxiliary material parameters on the cutting result. This, in turn, facilitates precise optimization of auxiliary material ratios and cutting yield control, reduces auxiliary material usage costs, and improves the stability of the silicon wafer cutting process.

[0099] Accordingly, please refer to Figure 3 , Figure 3This is a schematic diagram of the structure of the excipient parameter influence determination system according to an embodiment of this application. The excipient parameter influence determination system 100 provided in this embodiment includes: an acquisition module 101, used to acquire target excipient parameters and corresponding cutting results, wherein the target excipient parameters are used to characterize the historical parameter values ​​of the target excipient; an interval division module 102, used to divide the value range of the target excipient parameters into multiple first intervals; an interval merging module 103, used to merge adjacent first intervals according to the distribution of cutting results in each first interval to obtain multiple second intervals; and an influence curve determination module 104, used to determine the influence curve of the target excipient parameters on the cutting results based on the center value of the target excipient parameters in each second interval and the median value of the cutting results, wherein the influence curve is used to determine the influence of the target excipient parameters on the cutting results.

[0100] It is understood that the auxiliary material parameter influence determination system in this application divides the target auxiliary material parameter value range into intervals and merges the intervals based on the distribution of the cutting results of the intervals. This achieves dynamic binning and noise suppression of the data. Based on the center value of the target auxiliary material parameter in the merged interval and the median value of the cutting results, it accurately depicts the stage influence curve of the auxiliary material parameter on the cutting results. This improves the accuracy of determining the stage influence of the auxiliary material parameter on the cutting results, which is conducive to the precise optimization of auxiliary material ratio and cutting yield control, reduces the cost of auxiliary material use, and improves the stability of the cutting process.

[0101] In some embodiments, the interval merging module 103 is further configured to: determine the similarity between adjacent first intervals based on the distribution of the cutting results located in each first interval;

[0102] Starting from the first first interval, adjacent first intervals with similarity less than the similarity threshold are merged one by one to obtain multiple second intervals.

[0103] In some embodiments, the interval merging module 103 is further configured to: if the similarity between the first interval and the adjacent first interval is less than the similarity threshold, merge the first interval and the adjacent first interval into a new first interval, and repeatedly perform the operation of comparing the similarity between the new first interval and the adjacent first interval with the similarity threshold until the similarity between the new first interval and the adjacent first interval is greater than or equal to the similarity threshold, then determine the new first interval as the second interval, and continue to determine the comparison result of the similarity between the adjacent first interval and the next adjacent first interval with the similarity threshold.

[0104] In some embodiments, the interval merging module 103 is further configured to: determine the normal percentage and abnormal percentage of the cutting results located in each of the first intervals;

[0105] Based on the normal proportion and the abnormal proportion, the distribution of the cutting results located in the corresponding first interval is determined.

[0106] In some embodiments, the interval merging module 103 is further configured to: perform binary classification processing on the cutting results corresponding to the target auxiliary material parameters according to a preset target yield threshold, and obtain a binary classification processing result;

[0107] Based on the binary classification results and the total number of the cutting results within the first interval, the normal and abnormal proportions of the cutting results located in the first interval are determined.

[0108] In some embodiments, the interval division module 102 is further configured to: divide the range of values ​​of the target auxiliary material parameter into multiple first intervals according to the equal width rule.

[0109] In some embodiments, the median value of the cutting results includes the median or average value of the cutting results.

[0110] Accordingly, please refer to Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. An electronic device provided in this application includes a memory 501 and a processor 502. The memory 501 is used to store a computer program. The processor 502 is used to execute the computer program stored in the memory 501. When the computer program stored in the memory 501 is executed, the processor 502 executes the method for determining the influence of auxiliary material parameters according to the foregoing embodiments of this application.

[0111] Accordingly, embodiments of this application also provide a computer-readable storage medium storing computer instructions for causing a processor to execute the method for determining the influence of auxiliary material parameters as described in the foregoing embodiments of this application.

[0112] The above provides a detailed description of the method, system, device, and storage medium for determining the influence of auxiliary material parameters provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the technical solutions and core ideas of this application. Those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for determining the influence of excipient parameters, characterized in that, This method, applied in silicon wafer dicing, determines the stage-specific impact of dicing auxiliary material parameters on the dicing results. Obtain the target auxiliary material parameters and the corresponding cutting results, wherein the target auxiliary material parameters are used to characterize the historical parameter values ​​of the target auxiliary material; the target auxiliary material parameters are the cutting auxiliary material parameters in the silicon wafer cutting process; The range of values ​​for the target excipient parameters is divided into multiple first intervals; Based on the distribution of the cutting results in each first interval, adjacent first intervals are merged to obtain multiple second intervals; the cutting results include normal cutting and abnormal cutting; the distribution of the cutting results in each first interval is related to the normal proportion and abnormal proportion of the cutting results in each corresponding first interval; Based on the center value of the target auxiliary material parameter located in each second interval and the median value of the cutting result, a stage influence curve of the target auxiliary material parameter on the cutting result is determined. The stage influence curve is used to determine the stage influence of the target auxiliary material parameter on the cutting result.

2. The method for determining the influence of excipient parameters according to claim 1, characterized in that, Methods for obtaining multiple second intervals include: Based on the distribution of the cutting results located in each of the first intervals, the similarity between adjacent first intervals is determined; Starting from the first first interval, adjacent first intervals with similarity less than the similarity threshold are merged one by one to obtain multiple second intervals.

3. The method for determining the influence of excipient parameters according to claim 2, characterized in that, The step of merging adjacent first intervals with similarity less than a similarity threshold one by one to obtain multiple second intervals includes: If the similarity between the first interval and the adjacent first interval is less than the similarity threshold, then the first interval and the adjacent first interval are merged into a new first interval, and the operation of comparing the similarity between the new first interval and the adjacent first interval with the similarity threshold is repeated until the similarity between the new first interval and the adjacent first interval is greater than or equal to the similarity threshold. Then the new first interval is determined as the second interval, and the comparison result of the similarity between the adjacent first interval and the next adjacent first interval with the similarity threshold is determined.

4. The method for determining the influence of excipient parameters according to claim 2, characterized in that, The method for determining the distribution of the cutting results located in each of the first intervals includes: Determine the normal and abnormal proportions of the cutting results located in each of the first intervals; Based on the normal proportion and the abnormal proportion, the distribution of the cutting results located in the corresponding first interval is determined.

5. The method for determining the influence of excipient parameters according to claim 4, characterized in that, The method for determining the normal and abnormal proportions of the cutting results located in each of the first intervals includes: The cutting results corresponding to the target auxiliary material parameters are subjected to binary classification processing based on the preset target yield threshold to obtain the binary classification processing result. Based on the binary classification results and the total number of the cutting results within the first interval, the normal and abnormal proportions of the cutting results located in the first interval are determined.

6. The method for determining the influence of excipient parameters according to claim 1, characterized in that, The process of dividing the value range of the target excipient parameter into multiple first intervals includes: The range of values ​​for the target auxiliary material parameters is divided into multiple first intervals according to the equal-width rule.

7. The method for determining the influence of excipient parameters according to claim 1, characterized in that, The median value of the cutting results includes the median or average value of the cutting results.

8. A system for determining the influence of excipient parameters, characterized in that, This system, applied in silicon wafer dicing, determines the stage-specific impact of dicing auxiliary material parameters on the dicing results. The system includes: The acquisition module is used to acquire target auxiliary material parameters and the cutting results corresponding to the target auxiliary material parameters. The target auxiliary material parameters are used to characterize the historical parameter values ​​of the target auxiliary material. The target auxiliary material parameters are the cutting auxiliary material parameters in the silicon wafer cutting process. The interval division module is used to divide the value range of the target excipient parameter into multiple first intervals; The interval merging module is used to merge adjacent first intervals according to the distribution of the cutting results in each first interval to obtain multiple second intervals; the cutting results include normal cutting and abnormal cutting; the distribution of the cutting results in each first interval is related to the normal proportion and abnormal proportion of the cutting results in each corresponding first interval; The influence curve determination module is used to determine the stage influence curve of the target auxiliary material parameter on the cutting result based on the center value of the target auxiliary material parameter located in each second interval and the median value of the cutting result. The stage influence curve is used to determine the stage influence of the target auxiliary material parameter on the cutting result.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory; the memory stores a computer program, and the processor executes the computer program stored in the memory to implement the method for determining the influence of auxiliary material parameters as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the method for determining the influence of auxiliary material parameters as described in any one of claims 1 to 7.

Images