Parameter budget range determination method and apparatus, storage medium, and computer device

Abstract

The application discloses a parameter budget range determination method and device, a storage medium and computer equipment, relates to the high-resolution imaging system budget decomposition technical field, and is applied to a photoetching system. The method comprises the following steps: obtaining a test photoetching parameter, wherein the test photoetching parameter comprises a line width parameter and a period parameter; inputting the line width parameter and the period parameter into a pre-trained classification model; determining a target negative result farthest from a classification boundary of the classification model in negative results output by the classification model, and determining a line width parameter and the period parameter corresponding to the target negative result; and determining a parameter budget range of the photoetching system according to the line width parameter, the period parameter corresponding to the target negative result and the classification model. The above method can significantly improve the determination precision of the parameter budget range of the photoetching system.

Description

Technical Field

[0001] This invention relates to the field of high-resolution imaging system budget decomposition technology, and in particular to a method, apparatus, storage medium, and computer device for determining the parameter budget range. Background Technology

[0002] With the continuous development of photolithography technology, photolithography systems have become crucial production systems in semiconductor manufacturing. Among them, photolithography machines are widely used in the manufacturing of large-scale integrated circuits. The photolithography process uses optical exposure to copy the pattern on a photomask into a photoresist coated on a silicon wafer. Subsequently, processes such as development and etching further transfer the pattern onto the silicon wafer. Therefore, the technological level of the photolithography system directly determines the feature size in integrated circuit devices, playing a key role in the manufacturing of large-scale integrated circuits.

[0003] Currently, as integrated circuit process nodes continue to shrink, the requirements for lithography imaging quality are also constantly increasing. For extreme ultraviolet (EUV) lithography systems, various lithography parameters, such as light source polarization and energy stability, pupil uniformity, polar symmetry, and ellipticity of the illumination system, and numerical aperture and stray light of the projection system, have a crucial impact on the exposure results. However, these lithography parameters are not fixed; their values ​​may fluctuate within a certain range due to limitations such as production capacity. Furthermore, to meet the requirements for lithography exposure results, the variation range of these lithography parameters needs to be limited to meet the process parameters. This requires budget decomposition of the lithography parameters to obtain the budget range of lithography parameters that meets the process requirements. However, existing methods for determining the budget range of lithography system parameters often involve conducting extensive experiments to establish the correspondence between each lithography parameter and process conditions, thereby determining the budget range of the lithography parameters. However, this method is susceptible to the mutual influence between parameters, resulting in poor accuracy in calculating the budget range of the lithography system parameters. Summary of the Invention

[0004] In view of this, this application provides a method, apparatus, storage medium and computer equipment for determining the parameter budget range, the main purpose of which is to solve the technical problem of poor accuracy in calculating the budget range of lithography parameters of a lithography system.

[0005] According to a first aspect of the present invention, a method for determining a parameter budget range is provided, applied to a photolithography system, the method comprising:

[0006] Obtain test lithography parameters, wherein the test lithography parameters include linewidth parameters and period parameters;

[0007] The linewidth parameter and the period parameter are input into the pre-trained classification model to determine the target negative result that is furthest from the classification boundary of the classification model among the negative results output by the classification model, and the linewidth parameter and the period parameter corresponding to the target negative result are determined.

[0008] Based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result, the parameter budget range of the lithography system is determined.

[0009] According to a second aspect of the present invention, an apparatus for determining a parameter budget range is provided, the apparatus comprising:

[0010] The parameter acquisition module is used to acquire test lithography parameters, wherein the test lithography parameters include linewidth parameters and period parameters;

[0011] The parameter determination module is used to input the line width parameter and the period parameter into the pre-trained classification model, determine the target negative result that is farthest from the classification boundary of the classification model among the negative results output by the classification model, and determine the line width parameter and the period parameter corresponding to the target negative result;

[0012] The range determination module is used to determine the parameter budget range of the lithography system based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result.

[0013] According to a third aspect of the present invention, a storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for determining the above-described parameter budget range.

[0014] According to a fourth aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for determining the above-mentioned parameter budget range.

[0015] This invention provides a method, apparatus, storage medium, and computer device for determining the parameter budget range. First, test lithography parameters are acquired. Multiple sets of test lithography parameters are available, each set including linewidth and period parameters. Then, the test lithography parameters are input into a pre-trained classification model to identify negative results that do not meet the process parameters. Further, among the negative results, the target negative result furthest from the classification boundary of the classification model is identified, yielding the linewidth and period parameters corresponding to the worst exposure pattern. Finally, the linewidth and period parameters corresponding to the worst exposure pattern are input into the classification model. Based on these parameters, the parameter budget range that the lithography system's lithography parameters must fall within in order for the classification model to output a positive result is determined. The technical solution of this application can determine the worst exposure pattern generated under the worst parameter conditions according to the classification model, and determine the expression of the classification model when the critical parameters of the process conditions are met under the exposure pattern, so as to obtain the budget range of each parameter, thereby significantly improving the accuracy of the parameter budget range determination of the lithography system under multiple patterns.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0018] Figure 1 A flowchart illustrating a method for determining a parameter budget range according to an embodiment of the present invention is shown;

[0019] Figure 2 A schematic diagram of a parameter budget range determination device provided by an embodiment of the present invention is shown. Detailed Implementation

[0020] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0021] Currently, as integrated circuit process nodes continue to shrink, the requirements for lithography imaging quality are also constantly increasing. For extreme ultraviolet (EUV) lithography systems, various lithography parameters, such as light source polarization and energy stability, pupil uniformity, polar symmetry, and ellipticity of the illumination system, and numerical aperture and stray light of the projection system, have a crucial impact on the exposure results. However, these lithography parameters are not fixed; their values ​​may fluctuate within a certain range due to limitations such as production capacity. Furthermore, to meet the requirements for lithography exposure results, the variation range of these lithography parameters needs to be limited to meet the process parameters. This requires budget decomposition of the lithography parameters to obtain the budget range of lithography parameters that meets the process requirements. However, existing methods for determining the budget range of lithography system parameters often involve conducting extensive experiments to establish the correspondence between each lithography parameter and process conditions, thereby determining the budget range of the lithography parameters. However, this method is susceptible to the mutual influence between parameters, resulting in poor accuracy in calculating the budget range of the lithography system parameters.

[0022] To address the above problems, in one embodiment, such as Figure 1 As shown, a method for determining the parameter budget range is provided. Taking the application of this method to a computer device in a lithography system as an example, the method includes the following steps:

[0023] 101. Obtain test lithography parameters.

[0024] The test lithography parameters include linewidth parameters and period parameters. Furthermore, the test lithography parameters may also include parameters that affect the exposure results, such as light source polarization and energy stability, pupil uniformity, polar symmetry, and ellipticity of the illumination system, and numerical aperture and stray light of the projection system. The specific parameter types can be determined according to the actual situation.

[0025] Specifically, multiple sets of test lithography parameters can be obtained for subsequent parameter budget range determination.

[0026] 102. Input the line width parameter and the period parameter into the pre-trained classification model, determine the target negative result that is furthest from the classification boundary of the classification model among the negative results output by the classification model, and determine the line width parameter and the period parameter corresponding to the target negative result.

[0027] The classification model can be a binary classification model based on the support vector machine algorithm. This model receives the test lithography parameters and distinguishes whether the test lithography parameters can yield an exposure result that meets the process parameters required by the lithography system. If the test lithography parameters cannot yield an exposure result that meets the process parameters, the classification model will classify it in the negative direction of the classification boundary, resulting in a negative result. Conversely, if the test lithography parameters can yield an exposure result that meets the process parameters, the classification model will classify it in the positive direction of the classification boundary, resulting in a positive result.

[0028] Furthermore, the boundary formula for the classification model can be shown in Equation 1:

[0029]

[0030] Where Y represents the output of the classification model, a positive result is output when Y is greater than zero, and a negative result is output when Y is less than zero. i For support vector coefficients, and This is an n-dimensional vector obtained based on the test lithography parameters, where n can be the number of test lithography parameters, including linewidth and period parameters. b is a constant. This is the kernel function of the classification model. When the tested lithography parameters cannot produce exposure results that meet the process parameters required by the lithography system, the classification model will output -1; when the tested lithography parameters can produce exposure results that meet the process parameters required by the lithography system, the classification model will output 1.

[0031] Specifically, the linewidth and period parameters of each test lithography set can be input into the kernel function of the classification model to obtain the classification result. Then, based on a genetic algorithm, the target negative result furthest from the classification boundary of the classification model is determined. Here, the target negative result can refer to the worst exposure pattern obtained based on the test lithography parameters, thereby determining the linewidth and period parameters corresponding to the target negative result. Furthermore, the linewidth and period parameters of each test lithography set can also be input into Formula 2 to determine the minimum value of Formula 2 obtained based on the linewidth and period parameters. Specifically, Formula 2 can be the value in the boundary formula, as shown below:

[0032]

[0033] Among them, w i For support vector coefficients, and This is an n-dimensional vector obtained based on the test lithography parameters, where n can be the number of test lithography parameters, including linewidth and period parameters. b is a constant. This is the kernel function for the classification model.

[0034] Furthermore, the linewidth and period parameters that obtain the minimum value are determined as the linewidth and period parameters of the target negative result.

[0035] 103. Determine the parameter budget range of the lithography system based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result.

[0036] Specifically, the linewidth and period parameters of the target negative result can be substituted into... Among the values, The value can be shown in Formula 3:

[0037]

[0038] Where, x1 to x n-2 To test the other parameters in the lithography parameters besides the linewidth and period parameters, width is the linewidth parameter of the target negative result, and pitch is the period parameter of the target negative result.

[0039] Furthermore, the values ​​assigned to Formula 3 will be... Substituting these values ​​into the boundary formula of the classification model, the boundary formula is determined based on... of The parameter budget boundary when the value of x is zero is calculated, and the range from x1 to x is calculated when the classification model outputs a positive result. n-2 The required range of lithography parameters is determined, and the parameter budget range of the lithography system is determined based on the range of each lithography parameter. This parameter budget range can be the overall budget range between each lithography parameter.

[0040] The method for determining the parameter budget range provided in this embodiment first obtains test lithography parameters. The number of test lithography parameters can be multiple sets, each set including linewidth parameters and period parameters. Then, the test lithography parameters are input into a pre-trained classification model to identify negative results that do not meet the process parameters. Further, among the negative results, the target negative result furthest from the classification boundary of the classification model is identified, obtaining the linewidth and period parameters corresponding to the worst exposure pattern in the exposure results. Finally, the linewidth and period parameters corresponding to the worst exposure pattern are input into the classification model. Based on the linewidth and period parameters corresponding to the worst exposure pattern, the parameter budget range required for the lithography system's lithography parameters to achieve a positive classification result is determined. The technical solution of this application can determine the worst exposure pattern generated under the worst parameter conditions according to the classification model, and determine the expression of the classification model when the critical parameters of the process conditions are met under this exposure pattern, thus obtaining the budget range of each parameter. This significantly improves the accuracy of determining the parameter budget range of the lithography system under multiple patterns.

[0041] In one embodiment, step 103 can be implemented as follows: First, the linewidth parameter corresponding to the target negative result is determined as the target linewidth parameter, and the period parameter corresponding to the target negative result is determined as the target period parameter; then, based on the target linewidth parameter, the target period parameter, and preset lithography parameter variables, a first parameter combination is constructed; wherein, the lithography parameter variables can be for x1 to x n-2 For each lithography parameter setting, the unknowns are combined with the variable values ​​of each lithography parameter, the target linewidth parameter, and the target period parameter to form a first parameter combination. The form of the first parameter combination can be as shown in Formula 3. Further, the numerical range of the variable values ​​of the lithography parameters is determined when the classification result obtained by the classification model based on the first parameter combination is positive. Specifically, the first parameter combination is substituted into Formula 1 above to calculate the range from x1 to x when Y is greater than zero. n-2 The required numerical range is determined. Finally, the numerical range of the variable lithography parameters is defined as the parameter budget range. The embodiments provided in this application can quickly locate the preset range of lithography parameters, improving the efficiency of determining the parameter budget range of the lithography system.

[0042] Furthermore, the lithography parameters to be exposed can be obtained and input into the classification model. The model's output is then used to determine if it is positive. If the model's output is positive, it means that the lithography parameters meet the process parameters of the lithography system and exposure can be performed based on them. If the model's output is negative, it means that the lithography parameters do not meet the process parameters of the lithography system and exposure cannot be performed based on them.

[0043] In one embodiment, step 103 can also be implemented as follows: First, the linewidth parameter corresponding to the target negative result is determined as the target linewidth parameter, and the period parameter corresponding to the target negative result is determined as the target period parameter; then, based on the target linewidth parameter, the target period parameter, and a preset budget parameter range, a second parameter combination is constructed; wherein, the preset budget parameter range can be a range pre-defined for x1 to x n-2 The numerical range of each lithography parameter setting is defined. Further, the target parameter range corresponding to the budget parameter range is determined when the classification result obtained by the classification model based on the second parameter combination is positive. Specifically, the second parameter combination can be substituted into Formula 1 above to determine at what value within the budget parameter range Y is greater than zero. The value within the budget parameter range of the lithography parameter that makes the output value Y of the classification model greater than zero is determined as the target parameter range of that lithography parameter. Finally, the target parameter range is determined as the parameter budget range. The embodiments provided in this application can quickly determine the range of lithography parameters that enable the classification model to output a positive result and determine the parameter budget range of the lithography system.

[0044] In one embodiment, the training method for the classification model includes: first, obtaining exposure parameters. The exposure parameters include positive exposure parameters and negative exposure parameters. The positive exposure parameters include lithography parameters that meet the process requirements of the lithography system, and the negative exposure parameters include lithography parameters that do not meet the process requirements of the lithography system.

[0045] The positive exposure parameters include positive exposure input parameters and positive exposure result parameters. The positive exposure input parameters are lithography parameters that meet the process requirements of the lithography system. Exposure of the lithography system based on the positive exposure input parameters yields an exposure result that meets the process requirements. The positive exposure result parameters are the exposure results obtained by exposing the lithography system based on the positive exposure input parameters, including the pattern critical dimension (CD) result and the pattern offset (PS) result. The negative exposure parameters include negative exposure input parameters and negative exposure result parameters. The negative exposure parameters are lithography parameters that do not meet the process requirements of the lithography system. Exposure of the lithography system based on the negative exposure input parameters yields an exposure result that does not meet the process requirements. The negative exposure result parameters are the exposure results obtained by exposing the lithography system based on the negative exposure input parameters.

[0046] Furthermore, a classification model is established based on the positive and negative exposure parameters using the Support Vector Machine (SVM) algorithm. The positive and negative exposure parameters respectively include the linewidth parameter and the period parameter. Specifically, based on the applicability of SVM to binary classification modeling under multiple influences, the boundaries of the two types of exposure parameters (positive and negative) are determined. The classification model is trained, and the penalty function in the SVM and the gamma in the Gaussian kernel function are optimized using the Particle Swarm Optimization (PSO) algorithm to quickly obtain the optimal values ​​of the penalty function and the gamma in the Gaussian kernel function, thus finally obtaining the final classification model. The boundary formula of the classification model can be shown in Equation 1, and the Gaussian kernel function of the classification model can be shown in Equation 4.

[0047]

[0048] in, and This is an n-dimensional vector obtained based on the test lithography parameters, where n is the number of test lithography parameters, including linewidth and period parameters. 2 within It can be the distance between n-dimensional vectors.

[0049] The classification model obtained using the above method can receive lithography parameters and quickly determine which parameters meet the process requirements of the lithography system and which do not. It outputs positive results for parameters that meet the requirements and negative results for parameters that do not. The embodiments provided in this application can utilize positive exposure parameters, negative exposure parameters, support vector machine algorithms, and particle swarm optimization algorithms to establish and optimize a classification model, resulting in a high-precision classification model for subsequent budget determination.

[0050] In one embodiment, the implementation of establishing the classification model based on the positive and negative exposure parameters using the support vector machine algorithm can be as follows: First, the first-dimensional space containing the positive and negative exposure parameters is transformed into a second-dimensional space based on a kernel function algorithm, wherein the dimension of the second-dimensional space is higher than that of the first-dimensional space; specifically, the first-dimensional space containing the positive and negative exposure parameters can be transformed into a second-dimensional space based on a kernel function; as an example, if the space containing the positive and negative exposure parameters is a two-dimensional space, it can be transformed into a three-dimensional space based on a kernel function.

[0051] Then, a hyperplane for classifying the positive and negative exposure parameters is determined in the second-dimensional space. Specifically, the positive and negative exposure parameters are divided to obtain a hyperplane that is simultaneously furthest from both the positive and negative exposure parameters. Further, the second-dimensional space is transformed into a third-dimensional space, and the classification boundaries of the positive and negative exposure parameters are obtained based on the hyperplane. The second-dimensional space has a higher dimension than the third-dimensional space. For example, this three-dimensional space can be converted to a two-dimensional space to obtain the classification boundaries generated by the hyperplane dimensionality reduction operation, where the positive direction of the classification boundary represents the positive exposure parameters, and the negative direction represents the negative exposure parameters. Finally, a classification model including the classification boundaries is established. The embodiments provided in this application can accurately obtain the hyperplane of the classification model and the boundaries for classifying the parameters, improving the classification accuracy of the classification model.

[0052] In one embodiment, the acquisition of the positive and negative exposure parameters includes: first, inputting preset exposure parameters into a pre-trained exposure prediction model to obtain exposure results corresponding to the exposure parameters. The exposure results include positive exposure result parameters that meet the process requirements of the lithography system, and negative exposure result parameters that do not meet the process requirements of the lithography system. Specifically, the exposure prediction model is a pre-trained artificial neural network model that outputs the exposure results corresponding to the received exposure parameters. The exposure parameters may include parameters that affect the exposure results, such as light source polarization, energy stability, pupil uniformity, polar symmetry, and ellipticity of the illumination system, numerical aperture and stray light of the projection system, as well as linewidth and periodic parameters. The exposure results include graphic critical dimension (CD) results and graphic offset (PS) results.

[0053] Furthermore, exposure results where the key dimension of the image is within ±10% and the offset is within ±1 nanometer can be defined as positive exposure result parameters. Exposure results where the key dimension of the image is not within ±10% or the offset is not within ±1 nanometer can be defined as negative exposure result parameters. The specific criteria for judging positive and negative exposure result parameters can be determined according to the actual situation.

[0054] Furthermore, the exposure parameters used to generate the positive exposure result parameters based on the exposure prediction model are determined as positive exposure input parameters, and both the positive exposure result parameters and the positive exposure input parameters are determined as the positive exposure parameters. Conversely, the exposure parameters used to generate the negative exposure result parameters based on the exposure prediction model are determined as negative exposure input parameters, and both the negative exposure result parameters and the negative exposure input parameters are determined as the negative exposure parameters. The embodiments provided in this application can output lithography results, including those that meet or do not meet the lithography process requirements, based on a pre-trained neural network model and lithography parameters, avoiding the process of obtaining lithography results based on actual lithography operations, and improving the efficiency of determining positive and negative exposure parameters.

[0055] In one embodiment, the training method for the exposure prediction model includes: first, acquiring training samples. The training samples include exposure execution parameters and corresponding exposure output parameters; further, the exposure execution parameters include parameters that significantly influence the exposure results, such as light source polarization and energy stability, pupil uniformity, polar symmetry, and ellipticity of the illumination system, and numerical aperture and stray light of the projection system, as well as linewidth and periodic parameters; the exposure output parameters include key dimension results and offset results.

[0056] Furthermore, the exposure execution parameters include linewidth parameters such as 14nm, 16nm, 18nm, 20nm, 22nm, and 30nm. The ratio of the difference between the period parameter and the linewidth parameter to the linewidth parameter can be 1:1, 1:2, 1:3, 1:4, 1:5, 1:10, and some random values. In order to make the network applicable to all linewidths and periods, experiments with some random linewidths and random periods can be introduced into the training set.

[0057] In actual exposure operations, the lithography system can be exposed using exposure execution parameters to obtain exposure output parameters. Furthermore, exposure results from historical exposure operations can be used as exposure output parameters, while the lithography parameters used to perform the current exposure can be used as exposure execution parameters.

[0058] Furthermore, the exposure prediction model is obtained by training a preset artificial neural network model based on the exposure execution parameters and the exposure output parameters. Specifically, the BP neural network model can be trained using the aforementioned exposure execution parameters as training samples and the exposure output parameters as training labels to obtain the exposure prediction model. The embodiments provided in this application can train the neural network model based on historical lithography parameters and exposure results to obtain an exposure prediction model for predicting exposure results, reducing the number of exposure experiments required in subsequent budget determination work and improving the efficiency of budget determination work.

[0059] The parameter budget range determination method provided in this embodiment can obtain a large number of exposure results corresponding to a large number of lithography parameters by replacing a large amount of exposure experimentation with an exposure prediction model, and train a classification model based on the data generated by the exposure prediction model. Furthermore, the worst exposure pattern generated under the worst parameter conditions is determined according to the classification model, and the expression of the classification model is determined when the critical parameters satisfy the process conditions under this exposure pattern, thus obtaining the budget range of each parameter, thereby significantly improving the accuracy of determining the parameter budget range of the lithography system.

[0060] Furthermore, as Figure 1 The specific implementation of the method shown in this embodiment provides a device for determining the parameter budget range, such as... Figure 2 As shown, the device includes: a parameter acquisition module 21, a parameter determination module 22, and a range determination module 23.

[0061] The parameter acquisition module 21 can be used to acquire test lithography parameters, wherein the test lithography parameters include linewidth parameters and period parameters.

[0062] The parameter determination module 22 can be used to input the line width parameter and the period parameter into the pre-trained classification model, determine the target negative result that is farthest from the classification boundary of the classification model among the negative results output by the classification model, and determine the line width parameter and the period parameter corresponding to the target negative result.

[0063] The range determination module 23 can be used to determine the parameter budget range of the lithography system based on the linewidth parameter, the period parameter and the classification model corresponding to the target negative result.

[0064] In a specific application scenario, the range determination module 23 can be used to determine the linewidth parameter corresponding to the target negative result as the target linewidth parameter, and the period parameter corresponding to the target negative result as the target period parameter; construct a first parameter combination based on the target linewidth parameter, the target period parameter, and preset lithography parameter variables; determine the numerical range of the lithography parameter variables when the classification result obtained by the classification model based on the first parameter combination is a positive result; and determine the numerical range of the lithography parameter variables as the parameter budget range.

[0065] In a specific application scenario, the range determination module 23 can be used to determine the linewidth parameter corresponding to the target negative result as the target linewidth parameter, and the period parameter corresponding to the target negative result as the target period parameter; construct a second parameter combination based on the target linewidth parameter, the target period parameter, and a preset budget parameter range; determine the target parameter range corresponding to the budget parameter range when the classification result obtained by the classification model based on the second parameter combination is a positive result; and determine the target parameter range as the parameter budget range.

[0066] In a specific application scenario, the parameter acquisition module 21 can be used to acquire exposure parameters, wherein the exposure parameters include positive exposure parameters and negative exposure parameters. The positive exposure parameters include lithography parameters that meet the process requirements of the lithography system, and the negative exposure parameters include lithography parameters that do not meet the process requirements of the lithography system. Using a support vector machine algorithm, a classification model is established based on the positive exposure parameters and the negative exposure parameters, wherein the positive exposure parameters and the negative exposure parameters respectively include the linewidth parameter and the period parameter.

[0067] In a specific application scenario, the parameter acquisition module 21 can be used to transform a first-dimensional space containing the positive exposure parameters and the negative exposure parameters into a second-dimensional space based on a kernel function algorithm, wherein the dimension of the second-dimensional space is higher than the dimension of the first-dimensional space; determine a hyperplane in the second-dimensional space to classify the positive exposure parameters and the negative exposure parameters; transform the second-dimensional space into a third-dimensional space, obtain the classification boundary of the positive exposure parameters and the negative exposure parameters based on the hyperplane, wherein the dimension of the second-dimensional space is higher than the dimension of the third-dimensional space; and establish the classification model including the classification boundary.

[0068] In a specific application scenario, the parameter acquisition module 21 can be used to input preset exposure parameters into a pre-trained exposure prediction model to obtain exposure results corresponding to the exposure parameters. The exposure results include positive exposure result parameters that meet the process requirements of the lithography system, and negative exposure result parameters that do not meet the process requirements of the lithography system. The exposure parameters generated based on the exposure prediction model for the positive exposure result parameters are determined as positive exposure input parameters, and both the positive exposure result parameters and the positive exposure input parameters are determined as the positive exposure parameters. Similarly, the exposure parameters generated based on the exposure prediction model for the negative exposure result parameters are determined as negative exposure input parameters, and both the negative exposure result parameters and the negative exposure input parameters are determined as the negative exposure parameters.

[0069] In a specific application scenario, the parameter acquisition module 21 can be used to acquire training samples, wherein the training samples include exposure execution parameters and exposure output parameters corresponding to the exposure execution parameters; and the preset artificial neural network model is trained based on the exposure execution parameters and the exposure output parameters to obtain the exposure prediction model.

[0070] It should be noted that other corresponding descriptions of the functional units involved in the parameter budget range determination device provided in this embodiment can be found in [reference needed]. Figure 1 The corresponding descriptions in [the document] will not be repeated here.

[0071] Based on the above, Figure 1 Accordingly, this embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the above-described method. Figure 1 The method for determining the parameter budget range is shown.

[0072] Based on this understanding, the technical solution of this application can be embodied in the form of a software product. The software product to be identified can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or portable hard drive), including several instructions to cause a computer device (such as a personal computer, server, or network device) to execute the methods described in the various implementation scenarios of this application.

[0073] Based on the above, Figure 1 The method shown, and Figure 2The illustrated embodiment of the parameter budget range determination device, in order to achieve the above objective, also provides a physical device for determining the parameter budget range. Specifically, this physical device can be a personal computer, server, smartphone, tablet computer, smartwatch, or other network device, etc. The physical device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-described... Figure 1 The method shown.

[0074] Optionally, the physical device may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0075] Those skilled in the art will understand that the physical device structure with a defined parameter budget range provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0076] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs to be identified. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0077] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented in hardware. By applying the technical solution of this application, firstly, test lithography parameters are obtained, wherein the test lithography parameters include linewidth parameters and period parameters; then, the linewidth parameters and the period parameters are input into a pre-trained classification model to determine the target negative result that is furthest from the classification boundary of the classification model among the negative results output by the classification model, and the linewidth parameters and period parameters corresponding to the target negative result are determined; finally, based on the linewidth parameters, the period parameters, and the classification model corresponding to the target negative result, the parameter budget range of the lithography system is determined. Compared with the prior art, this significantly improves the accuracy of determining the parameter budget range of the lithography system.

[0078] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be modified to be located in one or more apparatuses different from this embodiment. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.

[0079] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A method for determining the parameter budget range, applied to a photolithography system, characterized in that, The method includes: Obtain test lithography parameters, wherein the test lithography parameters include linewidth parameters and period parameters; The linewidth parameter and the period parameter are input into a pre-trained classification model to determine the target negative result that is furthest from the classification boundary of the classification model among the negative results output by the classification model, and to determine the linewidth parameter and the period parameter corresponding to the target negative result. The training method of the classification model includes: obtaining exposure parameters, wherein the exposure parameters include positive exposure parameters and negative exposure parameters, the positive exposure parameters including lithography parameters that meet the process requirements of the lithography system, and the negative exposure parameters including lithography parameters that do not meet the process requirements of the lithography system; and applying a kernel function algorithm to process the linewidth parameter and the period parameter. The first-dimensional space of the light parameters is transformed into a second-dimensional space, wherein the number of dimensions of the second-dimensional space is higher than that of the first-dimensional space; a hyperplane is determined in the second-dimensional space to classify the positive exposure parameters and the negative exposure parameters; the second-dimensional space is transformed into a third-dimensional space, and the classification boundary of the positive exposure parameters and the negative exposure parameters is obtained based on the hyperplane, wherein the number of dimensions of the second-dimensional space is higher than that of the third-dimensional space; a classification model including the classification boundary is established, wherein the positive exposure parameters and the negative exposure parameters respectively include the linewidth parameter and the period parameter; Based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result, the parameter budget range of the lithography system is determined.

2. The method according to claim 1, characterized in that, The step of determining the parameter budget range of the lithography system based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result includes: The linewidth parameter corresponding to the target negative result is determined as the target linewidth parameter, and the period parameter corresponding to the target negative result is determined as the target period parameter; Based on the target linewidth parameter, the target period parameter, and the preset lithography parameter variables, a first parameter combination is constructed; The range of values ​​for the lithography parameters when the classification result obtained by the classification model based on the first parameter combination is determined to be a positive result; The numerical range of the variable photolithography parameters is defined as the parameter budget range.

3. The method according to claim 1, characterized in that, The method for determining the parameter budget range of the lithography system based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result further includes: The linewidth parameter corresponding to the target negative result is determined as the target linewidth parameter, and the period parameter corresponding to the target negative result is determined as the target period parameter; Based on the target linewidth parameter, the target period parameter, and the preset budget parameter range, a second parameter combination is constructed; When the classification result obtained by the classification model based on the second parameter combination is determined to be a positive result, the target parameter range corresponding to the budget parameter range; The target parameter range is defined as the parameter budget range.

4. The method according to claim 1, characterized in that, The methods for obtaining the positive exposure parameters and the negative exposure parameters include: The preset exposure parameters are input into the pre-trained exposure prediction model to obtain the exposure results corresponding to the exposure parameters. The exposure results include positive exposure result parameters that meet the process requirements of the lithography system and negative exposure result parameters that do not meet the process requirements of the lithography system. The exposure parameter that generates the positive exposure result parameter based on the exposure prediction model is determined as the positive exposure input parameter, and the positive exposure result parameter and the positive exposure input parameter are determined as the positive exposure parameter; The exposure parameter that generates the negative exposure result parameter based on the exposure prediction model is determined as the negative exposure input parameter, and the negative exposure result parameter and the negative exposure input parameter are determined as the negative exposure parameter.

5. The method according to claim 4, characterized in that, The training method for the exposure prediction model includes: Obtain training samples, wherein the training samples include exposure execution parameters and exposure output parameters corresponding to the exposure execution parameters; The exposure prediction model is obtained by training a preset artificial neural network model based on the exposure execution parameters and the exposure output parameters.

6. A device for determining the parameter budget range, applied to a photolithography system, characterized in that, The device includes: The parameter acquisition module is used to acquire test lithography parameters, wherein the test lithography parameters include linewidth parameters and period parameters; A parameter determination module is used to input the linewidth parameter and the period parameter into a pre-trained classification model, determine the target negative result that is furthest from the classification boundary of the classification model among the negative results output by the classification model, and determine the linewidth parameter and the period parameter corresponding to the target negative result. The training method of the classification model includes: obtaining exposure parameters, wherein the exposure parameters include positive exposure parameters and negative exposure parameters, the positive exposure parameters include lithography parameters that meet the process requirements of the lithography system, and the negative exposure parameters include lithography parameters that do not meet the process requirements of the lithography system; and using a kernel function algorithm to process the linewidth parameter and the period parameter. The first-dimensional space of the negative exposure parameters is transformed into a second-dimensional space, wherein the number of dimensions of the second-dimensional space is higher than that of the first-dimensional space; a hyperplane is determined in the second-dimensional space to classify the positive and negative exposure parameters; the second-dimensional space is transformed into a third-dimensional space, and the classification boundary of the positive and negative exposure parameters is obtained based on the hyperplane, wherein the number of dimensions of the second-dimensional space is higher than that of the third-dimensional space; a classification model including the classification boundary is established, wherein the positive and negative exposure parameters respectively include the linewidth parameter and the period parameter; The range determination module is used to determine the parameter budget range of the lithography system based on the linewidth parameter, the period parameter, and the classification model corresponding to the target negative result.

7. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images