A screening method and device for semiconductor diaphragm material proportioning

Abstract

The application provides a screening method and device for a semiconductor diaphragm material ratio, wherein, in the application, the performance values of the stability terms of each candidate component formula are predicted according to the obtained prediction model of each stability term, then the prediction values and target performance values of each stability term are compared respectively, the performance achievement degrees of each stability term are determined, then the minimum performance that can be achieved under the constraint of the respective variables of each stability term is determined, so as to determine the anti-performance achievement degrees of each stability term by using the minimum performance, finally, the expected difference value of the current component ratio and the expected component ratio corresponding to the target performance value is determined according to the weight of each stability term, then a plurality of component ratios with the minimum expected difference value are selected as the basis for subsequent performance tests, through the above method, the relatively suitable component configuration can be quickly found, and the screening efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically, to a method and apparatus for screening the proportions of semiconductor separator materials. Background Technology

[0002] Diaphragm valves in semiconductor manufacturing processes are critical components for controlling the transport of ultrapure gases and chemical liquids. The performance of their seat or diaphragm materials directly determines the valve's reliability and lifespan. These materials are typically composites of various known polymer matrices (such as polytetrafluoroethylene or soluble polytetrafluoroethylene) and functional fillers (such as carbon fibers, glass microspheres, lubricants, etc.). The materials must simultaneously meet requirements for extremely low leakage rates (airtightness), long-term stable pressure holding capability (positive pressure holding), and dimensional and performance stability over a wide temperature range (e.g., -20°C to 150°C) (temperature resistance).

[0003] With the development of technology, people can use simulation to determine a certain property of a material. However, how to integrate multiple conflicting performance objectives (such as airtightness, positive pressure holding, and temperature resistance) and find a relatively suitable composition ratio remains a technical challenge. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method and apparatus for screening the formulation of semiconductor separator materials, so as to accurately screen out the formulation with relatively good overall performance.

[0005] In a first aspect, embodiments of this application provide a method for screening the formulation of semiconductor separator materials, the method comprising:

[0006] After obtaining the prediction model for each stability term, the target performance value set for each stability term is obtained. The independent variables of the prediction model for each stability term are the same, and each prediction model is a performance expectation model. The types of performance expectation models include: small expectation model and large expectation model.

[0007] For each candidate component ratio, the candidate component ratio is used as an independent variable and input into the prediction model of each stability term to obtain the predicted value output by the prediction model of each stability term.

[0008] For each stability term, calculate the first ratio between the predicted value corresponding to the stability term and the target performance value corresponding to the stability term;

[0009] Determine the interval in which the first ratio falls within the type configuration of the prediction model for the stability term, and determine the performance achievement of the stability term based on the interval;

[0010] Using the range of values ​​of the independent variables of the stability term as constraints, and based on the type of performance expectation model of the stability term, calculate the worst value corresponding to the lowest performance of the stability term;

[0011] Calculate the second ratio of the target performance value and the worst value corresponding to the stability term, and use the second ratio as the inverse performance achievement of the stability term;

[0012] Based on the performance achievement and anti-performance achievement of each stability item, determine the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value;

[0013] After obtaining the expected difference values ​​corresponding to all candidate component proportions, determine the target expected difference value of a preset number in order of the expected difference values ​​from smallest to largest.

[0014] Output the target component ratio corresponding to the target expected difference, so as to perform performance testing on the material corresponding to the target component ratio.

[0015] Secondly, embodiments of this application provide a screening device for the formulation of semiconductor separator materials, the device comprising:

[0016] The acquisition unit is used to acquire the target performance value set for each stability term after obtaining the prediction model of each stability term. The independent variables of the prediction model of each stability term are the same, and each prediction model is a performance expectation model. The types of the performance expectation model include: small expectation model and large expectation model.

[0017] The first determining unit is used to input each candidate component ratio as an independent variable into the prediction model of each stability term to obtain the predicted value output by the prediction model of each stability term.

[0018] The first calculation unit is used to calculate a first ratio between the predicted value corresponding to the stability term and the target performance value corresponding to the stability term for each stability term.

[0019] The second determining unit is used to determine the interval in which the first ratio is located for the type configuration of the prediction model for the stability term, so as to determine the performance achievement of the stability term based on the interval;

[0020] The second calculation unit is used to calculate the worst value of the stability term at the lowest performance, based on the range of values ​​of the independent variables of the stability term and the type of the performance expectation model of the stability term.

[0021] The third calculation unit is used to calculate the second ratio of the target performance value and the worst value corresponding to the stability term, so as to use the second ratio as the inverse performance achievement degree of the stability term;

[0022] The third determining unit is used to determine the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement degree and anti-performance achievement degree of each stability item.

[0023] The fourth determining unit is used to determine a preset number of target expected differences in order of the expected differences from smallest to largest after obtaining the expected differences corresponding to all candidate component ratios.

[0024] The output unit is used to output the target component ratio corresponding to the target expected difference, so as to perform performance testing on the material corresponding to the target component ratio.

[0025] The technical solution provided in this application includes, but is not limited to, the following beneficial effects:

[0026] In this application, the performance values ​​(i.e., predicted values) of each stability term of each candidate ingredient formulation are predicted based on the prediction model of each stability term. Then, the predicted values ​​and target performance values ​​of each stability term are compared to determine the performance achievement degree of each stability term. Next, the minimum performance that can be achieved under the constraints of the independent variables of each stability term is determined so as to determine the inverse performance achievement degree of each stability term using the minimum performance. Then, the performance achievement degree of the corresponding stability term is normalized by using the inverse performance achievement degree of each stability term as a normalization condition. Finally, the expected difference between the current component ratio and the expected component ratio corresponding to the target performance value is determined according to the weight of each stability term. Then, the component ratios with the smallest expected difference are selected as the basis for subsequent performance testing. The above method can quickly find relatively suitable component configurations, which is beneficial to improving screening efficiency.

[0027] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A schematic flowchart illustrating a method for screening the formulation of semiconductor separator materials provided in this application embodiment;

[0030] Figure 2 A schematic flowchart illustrating another method for screening the formulation of semiconductor separator materials provided in this application embodiment;

[0031] Figure 3 This is a schematic diagram of a screening device for the proportion of semiconductor membrane materials provided in an embodiment of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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 some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0033] Figure 1 A flowchart illustrating a method for screening the formulation of semiconductor separator materials provided in this application is shown below. Figure 1 As shown, the method includes the following steps:

[0034] Step 101: After obtaining the prediction model for each stability term, obtain the target performance value set for each stability term. The independent variables of the prediction model for each stability term are the same, and each prediction model is a performance expectation model. The types of performance expectation models include: small-scale expectation model and large-scale expectation model.

[0035] Step 102: For each candidate component ratio, input the candidate component ratio as an independent variable into the prediction model of each stability term to obtain the predicted value output by the prediction model of each stability term.

[0036] Step 103: For each stability term, calculate the first ratio between the predicted value corresponding to the stability term and the target performance value corresponding to the stability term.

[0037] Step 104: Determine the interval in which the first ratio is located for the type configuration of the prediction model for the stability term, so as to determine the performance achievement of the stability term based on the interval.

[0038] Step 105: Using the range of values ​​of the independent variables of the stability term as constraints, and based on the type of performance expectation model of the stability term, calculate the worst value corresponding to the stability term at the lowest performance.

[0039] Step 106: Calculate the second ratio of the target performance value and the worst value corresponding to the stability term, and use the second ratio as the inverse performance achievement degree of the stability term.

[0040] Step 107: Based on the performance achievement degree and anti-performance achievement degree of each stability item, determine the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value.

[0041] Step 108: After obtaining the expected difference values ​​corresponding to all candidate component ratios, determine the preset number of target expected differences in order of the expected differences from smallest to largest.

[0042] Step 109: Output the target component ratio corresponding to the target expected difference, so as to perform performance testing on the material corresponding to the target component ratio.

[0043] Specifically, after obtaining the prediction models for each stability item (including leakage rate, pressure decay rate, and temperature resistance, where pressure decay rate is equivalent to positive pressure holding), a component ratio (i.e., mass percentage, such as: components include: component 1, component 2, and component 3, where the mass percentage of component 1 is 20%, the mass percentage of component 2 is 50%, and the mass percentage of component 3 is 30%) can be simultaneously input into the prediction models corresponding to each stability item. Each prediction model predicts the predicted value of one stability item for that component ratio. When the stability item includes leakage rate, pressure decay rate, and temperature resistance, the output predicted value includes the predicted value of leakage rate, the predicted value of pressure decay rate, and the predicted value of temperature resistance. Of course, it can also be input into the prediction model corresponding to a certain stability item to obtain only the predicted value corresponding to that stability item.

[0044] To determine the difference between the overall performance and the target overall performance of the above-mentioned component ratios, it is necessary to determine the difference between the performance of each stability item and the target performance of each stability item. For example, the difference between the predicted leakage rate of the above-mentioned component ratios and the set target performance value for leakage rate; the difference between the predicted pressure decay rate of the above-mentioned component ratios and the set target performance value for pressure decay rate; and the difference between the predicted temperature resistance of the above-mentioned component ratios and the set target performance value for temperature resistance. That is, the difference between the two is represented by a first ratio. The interval in which the first ratio corresponding to each stability item falls within the type of prediction model configured for that stability item is determined. Then, the achievement degree configured for that interval is taken as the performance achievement degree of that stability item. The performance achievement degree represents completion. Furthermore, for each stability item, multiple intervals are defined using the percentage of the target performance value for that stability item. This determines which interval the predicted value of that stability item falls into, and a different achievement level is set for each interval. After determining the interval in which the predicted value of that stability item falls, the achievement level corresponding to that interval is taken as the performance achievement level of that stability item. For example, for a small-scale model, when the predicted value is less than 0.8 × the target performance value, it indicates an overachievement of more than 20% in performance. For a large-scale model, when the predicted value is greater than 1.2 × the target performance value, it indicates an overachievement of more than 20% in performance. That is, the smaller the predicted value of a small-scale model, the higher the performance of the corresponding stability item; the larger the predicted value of a large-scale model, the higher the performance of the corresponding stability item.

[0045] After determining the performance achievement level of each stability item, it is also necessary to determine the worst value of each performance within the feasible range for subsequent normalization processing. For each performance index, its worst value is solved under constraints. For example, leakage rate for component 1, pressure decay rate for component 2, and temperature resistance for component 3. The maximum and minimum mass percentages of each component are set, and then the worst value of the corresponding performance is determined within the corresponding mass percentage range. For example, the highest value of leakage rate that can be achieved is determined within the mass percentage range of component 1, the highest value of pressure decay rate that can be achieved is determined within the mass percentage range of component 2, and the minimum value of temperature resistance that can be achieved is determined within the mass percentage range of component 3. That is, the worst value corresponding to the lowest performance of each stability item is obtained. Then, the second ratio of the target performance value and the worst value corresponding to the stability item is calculated, and the second ratio is used as the inverse performance achievement level of the stability item.

[0046] After obtaining the performance achievement and anti-performance achievement of each stability term, the multi-objective problem can be transformed into a single-objective optimization problem. By using the performance achievement and anti-performance achievement, the gap between the current formulation and the target formulation can be found. Then, the preset number of candidate component ratios with the smallest gap is output. Based on the output results, the testers conduct actual performance tests on the materials corresponding to the candidate component ratios. Through the above method, the material with the closest performance to the target can be found without actual performance experiments, which is beneficial to improving material selection efficiency and reducing costs.

[0047] In one feasible implementation, the stability term includes:

[0048] Leakage rate, pressure decay rate, and temperature resistance;

[0049] The prediction models for leakage rate and attenuation are small-scale models, while the prediction model for temperature resistance is large-scale models.

[0050] In one feasible implementation, the step of determining the interval in which the first ratio falls within a type configuration of the prediction model for the stability term, and determining the performance achievement of the stability term based on the interval, includes:

[0051] For small-scale models, the performance achievement of this stability term is calculated using the following formula:

[0052] Formula 1;

[0053] in, Let j be the predicted value of the prediction model for the j-th stability term. The target performance value of the j-th stability term;

[0054] The performance achievement of this stability term is calculated using the following formula:

[0055] Formula 2;

[0056] in, These are the predicted values ​​from the prediction model corresponding to the temperature resistance. This represents the target performance value corresponding to the temperature resistance.

[0057] In one feasible implementation, the worst-case values ​​corresponding to the leakage rate and the pressure decay rate are minimum values, and the worst-case value corresponding to the temperature resistance is maximum value.

[0058] In one feasible implementation, the step of determining the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement and anti-performance achievement of each stability term includes:

[0059] The expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value is determined using the following formula:

[0060] D( =√[∑W i ×((1-d i ( )) / (1-d i - Formula 3

[0061] Where √ represents the square root, W i Let d be the weight corresponding to the i-th stability term. i ( ) represents the performance achievement degree corresponding to the i-th stability term, d i - Let represent the anti-performance achievement degree corresponding to the i-th stability term, where i is the number of stability terms.

[0062] d i ( ) represents the gap between the i-th performance and the target, 1-d i - The normalization factor makes the differences between different performance characteristics comparable. The square root of the sum of squares represents the Euclidean distance, considering the overall difference in performance. An example is provided:

[0063] Performance achievement: d( = [0.87, 0.92, 0.97];

[0064] Anti-performance achievement: d - =[0.25,0.30,0.73];

[0065] Weights: w = [0.4, 0.3, 0.3];

[0066] Then: D( )=

[0067] √[0.4×((1-0.87) / (1-0.25))²+0.3×((1-0.92) / (1-0.30))²+0.3×((1-0.97) / (1-0.73))²]=0.155.

[0068] The expected difference is not only an optimization objective, but also a comprehensive indicator for evaluating the quality of the solution. By using the expected difference, we can find a recipe that is relatively close to the ideal point (i.e., the overall performance of the objective) under a given weight, thus ensuring a smooth transition from mathematical optimality to engineering usability.

[0069] Figure 2 A schematic flowchart illustrating another method for screening the formulation of semiconductor separator materials provided in this application embodiment is shown below. Figure 2 As shown, the prediction models for each stability term can be obtained through the following steps:

[0070] Step 201: Obtain the stability parameters of each stability item and the mass percentage of each component for each experimental material. Some experimental materials contain the same types of components.

[0071] Step 202: Based on the stability parameters of each stability item and the mass ratio, construct a sample set corresponding to each stability item. Each sample set corresponding to each stability item includes multiple sample data, and each sample data consists of the stability parameters corresponding to that stability item and the mass ratio of each experimental material.

[0072] Step 203: Input each sample data corresponding to the stability term into Formula 4 to obtain multiple models corresponding to the stability term:

[0073] Formula 4

[0074] in, This refers to the stability parameter in a sample data set under this stability item. For constant terms, Let be the coefficient of the linear term of the i-th component in the sample data. Let be the coefficient of the quadratic term of the i-th component in the sample data. The coefficient of the interaction term between the i-th and j-th components in this sample data. The mass percentage of the i-th component in this sample data The mass percentage of the j-th component in this sample data The number of component types included in each experimental material. This is the random error term.

[0075] Step 204: Use the least squares method in multiple linear regression to fit multiple models corresponding to the stability term to obtain the prediction model corresponding to the stability term.

[0076] Specifically, for various experimental materials, each material is tested and experimented on to obtain the stability parameters of each stability item and the mass percentage of each component for each material. A sample set is constructed based on the data obtained above. For example, the stability items include: stability item 1, stability item 2, and stability item 3; the mass percentage of the components includes: mass percentage 1, mass percentage 2, and mass percentage 3. The sample data obtained from a certain experimental material includes: [stability item 1, mass percentage 1, mass percentage 2, mass percentage 3], [stability item 2, mass percentage 1, mass percentage 2, mass percentage 3], and [stability item 3, mass percentage 1, mass percentage 2, mass percentage 3]. That is, three sample data can be obtained for each sample.

[0077] When the experimental materials include experimental material 1 and experimental material 2, the sample data of experimental material 1 are [stability item 1, mass percentage 1, mass percentage 2, mass percentage 3], [stability item 2, mass percentage 1, mass percentage 2, mass percentage 3], and [stability item 3, mass percentage 1, mass percentage 2, mass percentage 3], and the sample data of experimental material 2 are [stability item 1, mass percentage 4, mass percentage 5, mass percentage 6], [stability item 2, mass percentage 4, mass percentage 5, mass percentage 6], and [stability item 3, mass percentage 4, mass percentage 5, mass percentage 6], which yields 3. There are three sample sets. Sample set 1 corresponding to stability item 1 includes: [stability item 1, quality percentage 1, quality percentage 2, quality percentage 3] and [stability item 1, quality percentage 4, quality percentage 5, quality percentage 6]. Sample set 2 corresponding to stability item 2 includes: [stability item 2, quality percentage 1, quality percentage 2, quality percentage 3] and [stability item 2, quality percentage 4, quality percentage 5, quality percentage 6]. Sample set 3 corresponding to stability item 3 includes: [stability item 3, quality percentage 1, quality percentage 2, quality percentage 3] and [stability item 3, quality percentage 4, quality percentage 5, quality percentage 6].

[0078] After obtaining multiple sample sets, for each sample set, each sample data in that set is input into Formula 4, thus obtaining multiple models for the stability term to which that sample set belongs. Taking stability term 1 as an example, each sample data in the sample set of stability term 1 is substituted into Formula 4, thus obtaining multiple models for stability term 1. Then, the models corresponding to stability term 1 are fitted to obtain the model corresponding to stability term 1. , , , and Then get , , , and Substituting back into Equation 4, we obtain Prediction Model 1 for Stability 1, and so on, we can obtain Prediction Model 2 for Stability 2 and Prediction Model 3 for Stability 3.

[0079] For example, three experimental materials are known, all of which consist only of polytetrafluoroethylene and carbon fiber. The stability parameters of these three materials were determined experimentally. The leakage rate will be used as an example to illustrate the stability parameters. Table 1 shows the obtained data:

[0080]

[0081] Table 1

[0082] Substitute the data from Table 1 into Formula 5 (i.e., the expansion of Formula 4):

[0083] Formula 5

[0084] The resulting models for leakage rate include:

[0085]

[0086] The leakage rate is fitted using these three equations (more data points will be used in practice). , , , and Then get , , , and Substituting back into Equation 4, we obtain the prediction model for the leakage rate. By analogy, we can obtain the prediction models for other stability terms.

[0087] It should be noted that the constant term The coefficient of the linear term represents the baseline performance level after all effects have been balanced. The quadratic coefficient is used to reveal the independent main effect of each component on performance, i.e., whether it promotes or inhibits it. To reveal the excess effect of a single component, i.e., whether there is an optimal amount to add or whether too much is as bad as too little, interaction term coefficients. To reveal the synergistic or antagonistic effect between the two components, i.e., whether 1+1>2 or 1+1<2, the random error term. This represents variations that the model could not explain, such as measurement errors and minor process variations. Different results can be obtained from multiple models corresponding to a certain stability term. The mean is determined.

[0088] It should be noted that when the composition includes three components, such as polytetrafluoroethylene (A), carbon fiber (B), and graphite (C), the component variables are x1 = the mass percentage of A, x2 = the mass percentage of B, and x3 = the mass percentage of C. Substituting these variables into Formula 4, the expanded formula is:

[0089] + + ;

[0090] There are a total of 10 β parameters (1 constant + 3 linear + 3 quadratic + 3 interactive).

[0091] When there are 4 components, there are 6 possible pairwise combinations, resulting in a total of 15 β parameters. These include 1 constant term, 4 linear terms (β1, β2, β3, β4), and 4 quadratic terms (β...). 11 ,β 22 ,β 33 ,β 44 There are 6 interactive items.

[0092] It should be noted that, typically, the number of sample data for each stability term is greater than the number of β parameters.

[0093] In one feasible implementation, the stability parameters include: the leakage rate (in sccm) measured using a helium mass spectrometer leak detector, the decay rate (in Pa / s) corresponding to the pressure decay curve detected at a specific positive pressure (e.g., 6 Bar), and the heat distortion temperature (in °C) determined by a thermomechanical analyzer.

[0094] It should be noted that the calculations involved in this application are performed using dimensionless exponents, so there is no need to consider whether the dimensions on both sides of the equation are the same.

[0095] Figure 3 This application provides a schematic diagram of a screening device for the proportioning of semiconductor separator materials, as shown in the embodiments of the present application. Figure 3 As shown, the device includes:

[0096] The acquisition unit 31 is used to acquire the target performance value set for each stability item after obtaining the prediction model of each stability item. The independent variables of the prediction model of each stability item are the same, and each prediction model is a performance expectation model. The types of the performance expectation model include: small expectation model and large expectation model.

[0097] The first determining unit 32 is used to input each candidate component ratio as an independent variable into the prediction model of each stability term to obtain the predicted value output by the prediction model of each stability term.

[0098] The first calculation unit 33 is used to calculate a first ratio between the predicted value corresponding to the stability term and the target performance value corresponding to the stability term for each stability term.

[0099] The second determining unit 34 is used to determine the interval in which the first ratio is located for the type configuration of the prediction model of the stability term, so as to determine the performance achievement of the stability term based on the interval;

[0100] The second calculation unit 35 is used to calculate the worst value of the stability term at the lowest performance, based on the range of values ​​of the independent variables of the stability term and the type of the performance expectation model of the stability term.

[0101] The third calculation unit 36 ​​is used to calculate the second ratio of the target performance value and the worst value corresponding to the stability item, so as to use the second ratio as the inverse performance achievement degree of the stability item;

[0102] The third determining unit 37 is used to determine the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement degree and anti-performance achievement degree of each stability item.

[0103] The fourth determining unit 38 is used to determine a preset number of target expected differences in order of the expected differences from smallest to largest after obtaining the expected differences corresponding to all candidate component ratios.

[0104] Output unit 39 is used to output the target component ratio corresponding to the target expected difference, so as to perform performance testing on the material corresponding to the target component ratio.

[0105] In one feasible implementation, the stability term includes:

[0106] Leakage rate, pressure decay rate, and temperature resistance;

[0107] The prediction models for leakage rate and attenuation are small-scale models, while the prediction model for temperature resistance is large-scale models.

[0108] In one feasible implementation, the second determining unit is used to determine the interval in which the first ratio falls within a type configuration of the prediction model for the stability term, and to determine the performance achievement of the stability term based on the interval, including:

[0109] For small-scale models, the performance achievement of this stability term is calculated using the following formula:

[0110] Formula 1;

[0111] in, Let j be the predicted value of the prediction model for the j-th stability term. The target performance value of the j-th stability term;

[0112] The performance achievement of this stability term is calculated using the following formula:

[0113] Formula 2;

[0114] in, These are the predicted values ​​from the prediction model corresponding to the temperature resistance. This represents the target performance value corresponding to the temperature resistance.

[0115] In one feasible implementation, the worst-case values ​​corresponding to the leakage rate and the pressure decay rate are minimum values, and the worst-case value corresponding to the temperature resistance is maximum value.

[0116] In one feasible implementation, when the third determining unit determines the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement degree and anti-performance achievement degree of each stability item, it includes:

[0117] The expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value is determined using the following formula:

[0118] D( =√[∑W i ×((1-d i ( )) / (1-d i - Formula 3

[0119] Where √ represents the square root, W i Let d be the weight corresponding to the i-th stability term. i ( ) represents the performance achievement degree corresponding to the i-th stability term, d i - Let represent the anti-performance achievement degree corresponding to the i-th stability term, where i is the number of stability terms.

[0120] about Figure 3 For explanations of the principles behind the content shown, please refer to [link / reference]. Figure 1 The detailed explanations of the relevant content shown will not be repeated here.

[0121] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0122] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0123] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0124] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0125] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0126] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for screening the formulation of semiconductor separator materials, characterized in that, The method includes: After obtaining the prediction model for each stability term, the target performance value set for each stability term is obtained. The independent variables of the prediction model for each stability term are the same, and each prediction model is a performance expectation model. The types of performance expectation models include: small expectation model and large expectation model. For each candidate component ratio, the candidate component ratio is used as an independent variable and input into the prediction model of each stability term to obtain the predicted value output by the prediction model of each stability term. For each stability term, calculate the first ratio between the predicted value corresponding to the stability term and the target performance value corresponding to the stability term; Determine the interval in which the first ratio falls within the type configuration of the prediction model for the stability term, and determine the performance achievement of the stability term based on the interval; Using the range of values ​​of the independent variables of the stability term as constraints, and based on the type of performance expectation model of the stability term, calculate the worst value corresponding to the lowest performance of the stability term; Calculate the second ratio of the target performance value and the worst value corresponding to the stability term, and use the second ratio as the inverse performance achievement of the stability term; Based on the performance achievement and anti-performance achievement of each stability item, determine the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value; After obtaining the expected difference values ​​corresponding to all candidate component proportions, determine the target expected difference value of a preset number in order of the expected difference values ​​from smallest to largest. Output the target component ratio corresponding to the target expected difference, so as to perform performance testing on the material corresponding to the target component ratio.

2. The screening method as described in claim 1, characterized in that, The stability terms include: Leakage rate, pressure decay rate, and temperature resistance; The prediction models for leakage rate and attenuation are small-scale models, while the prediction model for temperature resistance is large-scale models.

3. The screening method as described in claim 2, characterized in that, The step of determining the interval in which the first ratio falls within a type configuration of the prediction model for the stability term, and determining the performance achievement of the stability term based on the interval, includes: For small-scale models, the performance achievement of this stability term is calculated using the following formula: Formula 1; in, Let j be the predicted value of the prediction model for the j-th stability term. The target performance value of the j-th stability term; The performance achievement of this stability term is calculated using the following formula: Formula 2; in, These are the predicted values ​​from the prediction model corresponding to the temperature resistance. This represents the target performance value corresponding to the temperature resistance.

4. The screening method as described in claim 2, characterized in that, The worst-case values ​​corresponding to the leakage rate and the pressure decay rate are minimum values, and the worst-case value corresponding to the temperature resistance is maximum value.

5. The screening method as described in claim 1, characterized in that, The step of determining the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement degree and anti-performance achievement degree of each stability item includes: The expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value is determined using the following formula: D( =√[∑W i ×((1-d i ( )) / (1-d i - Formula 3 Where √ represents the square root, W i Let d be the weight corresponding to the i-th stability term. i ( ) represents the performance achievement degree corresponding to the i-th stability term, d i - Let represent the anti-performance achievement degree corresponding to the i-th stability term, where i is the number of stability terms.

6. A screening device for the proportioning of semiconductor separator materials, characterized in that, The device includes: The acquisition unit is used to acquire the target performance value set for each stability term after obtaining the prediction model of each stability term. The independent variables of the prediction model of each stability term are the same, and each prediction model is a performance expectation model. The types of the performance expectation model include: small expectation model and large expectation model. The first determining unit is used to input each candidate component ratio as an independent variable into the prediction model of each stability term to obtain the predicted value output by the prediction model of each stability term. The first calculation unit is used to calculate a first ratio between the predicted value corresponding to each stability term and the target performance value corresponding to each stability term. The second determining unit is used to determine the interval in which the first ratio is located for the type configuration of the prediction model for the stability term, so as to determine the performance achievement of the stability term based on the interval. The second calculation unit is used to calculate the worst value of the stability term at the lowest performance, based on the range of values ​​of the independent variables of the stability term and the type of the performance expectation model of the stability term. The third calculation unit is used to calculate the second ratio of the target performance value and the worst value corresponding to the stability item, so as to use the second ratio as the inverse performance achievement degree of the stability item; The third determining unit is used to determine the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement degree and anti-performance achievement degree of each stability item. The fourth determining unit is used to determine a preset number of target expected differences in order of the expected differences from smallest to largest after obtaining the expected differences corresponding to all candidate component ratios. The output unit is used to output the target component ratio corresponding to the target expected difference, so as to perform performance testing on the material corresponding to the target component ratio.

7. The screening device as described in claim 6, characterized in that, The stability terms include: Leakage rate, pressure decay rate, and temperature resistance; The prediction models for leakage rate and attenuation are small-scale models, while the prediction model for temperature resistance is large-scale models.

8. The screening device as described in claim 7, characterized in that, The second determining unit is used to determine the interval in which the first ratio falls within the type configuration of the prediction model for the stability term, and to determine the performance achievement of the stability term based on the interval, including: For small-scale models, the performance achievement of this stability term is calculated using the following formula: Formula 1; in, Let j be the predicted value of the prediction model for the j-th stability term. The target performance value of the j-th stability term; The performance achievement of this stability term is calculated using the following formula: Formula 2; in, These are the predicted values ​​from the prediction model corresponding to the temperature resistance. This represents the target performance value corresponding to the temperature resistance.

9. The screening device as described in claim 7, characterized in that, The worst-case values ​​corresponding to the leakage rate and the pressure decay rate are minimum values, and the worst-case value corresponding to the temperature resistance is maximum value.

10. The screening device as described in claim 6, characterized in that, When the third determining unit determines the expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value based on the performance achievement degree and anti-performance achievement degree of each stability item, it includes: The expected difference between the candidate component ratio and the expected component ratio corresponding to the target performance value is determined using the following formula: D( =√[∑W i ×((1-d i ( )) / (1-d i - Formula 3 Where √ represents the square root, W i Let d be the weight corresponding to the i-th stability term. i ( ) represents the performance achievement degree corresponding to the i-th stability term, d i - Let represent the anti-performance achievement degree corresponding to the i-th stability term, where i is the number of stability terms.

Images