Adaptive spectral adjustment method, device and equipment for wafer detection UV light source

Abstract

The application provides a wafer detection UV light source adaptive spectrum adjustment method, device and equipment, the method comprises the following steps: obtaining the first spectrum data of the preliminary spectrum detection of the target wafer; the first spectrum data is processed by variational mode decomposition, and the spectrum characteristic parameter of the first spectrum data is determined; the spectrum characteristic parameter is processed by fine composite multi-scale weighted multi-dimensional feature entropy, and the spectrum feature vector of the spectrum characteristic parameter is determined; the mapping relationship between the spectrum feature and the wafer material characteristics is determined by analyzing the spectrum feature vector; the spectrum of the UV light source is adaptively adjusted according to the mapping relationship. Through the implementation of the application scheme, the spectrum of the UV light source is adaptively adjusted according to the mapping relationship between the spectrum feature and the wafer material characteristics, which can ensure that the spectrum characteristics of the light source match the wafer detection requirements, avoid unnecessary spectrum redundancy or loss, improve spectrum utilization, shorten the detection time, and significantly improve the detection efficiency.

Description

Technical Field

[0001] This application relates to the field of light source control technology, and in particular to an adaptive spectral adjustment method, apparatus and equipment for wafer inspection UV light source. Background Technology

[0002] In the semiconductor manufacturing industry, wafer inspection is a critical step in ensuring product quality. UV light sources, as an important tool in wafer inspection, significantly impact the accuracy and reliability of inspection results due to their spectral characteristics. In existing technologies, the spectra of UV light sources are typically fixed, making it difficult to adapt to variations in different wafer materials and inspection requirements. While some technologies attempt to adjust the light source, these methods often lack precise control and adaptability regarding spectral characteristics. Summary of the Invention

[0003] This application provides an adaptive spectral adjustment method, apparatus, and device for wafer inspection UV light source, which solves the problem in related technologies that the UV light source spectrum cannot be dynamically adjusted according to the characteristics of wafer materials.

[0004] The first aspect of this application provides an adaptive spectral adjustment method for a UV light source for wafer inspection, the adaptive spectral adjustment method for the UV light source for wafer inspection comprising:

[0005] Acquire the first spectral data for preliminary spectral detection of the target wafer;

[0006] The first spectral data is subjected to variational mode decomposition to determine the spectral characteristic parameters of the first spectral data.

[0007] The spectral characteristic parameters are determined by processing the spectral characteristic parameters with refined composite multi-scale weighted multi-dimensional feature entropy.

[0008] The mapping relationship between spectral features and wafer material properties is determined by analyzing the spectral feature vectors.

[0009] The spectrum of the UV light source is adaptively adjusted according to the mapping relationship.

[0010] Optionally, in a first implementation of the first aspect of this application, the step of performing variational mode decomposition on the first spectral data to determine the spectral characteristic parameters of the first spectral data includes:

[0011] The first spectral data is decomposed into N modal components using a variational mode decomposition algorithm; where N is an integer greater than or equal to 1.

[0012] Each modal component is analyzed to determine its modal characteristic parameters;

[0013] The modal characteristic parameters of the N modal components are integrated to determine the spectral characteristic parameters of the first spectral data.

[0014] Optionally, in the second implementation of the first aspect of this application, the step of determining the spectral feature vector of the spectral feature parameters by processing the spectral characteristic parameters through refined composite multi-scale weighted multi-dimensional feature entropy includes:

[0015] The multi-scale spectral features corresponding to the spectral characteristic parameters are obtained by determining the entropy values ​​of the spectral characteristic parameters at different scales.

[0016] By weighting the multi-scale spectral features, the weighted spectral features corresponding to the spectral characteristic parameters are obtained.

[0017] By extracting features from the weighted spectral features, the spectral feature vector of the spectral characteristic parameter is determined.

[0018] Optionally, in a third implementation of the first aspect of this application, the step of determining the mapping relationship between spectral features and wafer material properties by analyzing the spectral feature vector includes:

[0019] Spectral feature clusters are generated by clustering the spectral feature vectors; wherein, the spectral feature clusters are clusters of wafer materials with the same type of spectral features;

[0020] Statistical analysis is performed on the spectral feature vectors corresponding to the spectral feature clusters to determine the wafer material properties corresponding to the spectral feature clusters.

[0021] Based on the spectral feature clusters and the corresponding wafer material properties, a mapping relationship between spectral features and wafer material properties is established.

[0022] Optionally, in the fourth implementation of the first aspect of this application, the step of adaptively adjusting the spectrum of the UV light source according to the mapping relationship includes:

[0023] The spectral requirements of the target wafer are determined based on the mapping relationship;

[0024] The spectrum of the UV light source is initially adjusted according to the spectral requirements;

[0025] Obtain the second spectral data of the UV light source after preliminary adjustment;

[0026] Based on the second spectral data and the mapping relationship, determine whether the second spectral data meets the spectral requirements of the target wafer;

[0027] If the second spectral data does not meet the spectral requirements, the spectrum of the UV light source is adjusted a second time until the second spectral data meets the spectral requirements.

[0028] Optionally, in the fifth implementation of the first aspect of this application, the method further includes:

[0029] The second spectral data is compared with the spectral requirements to determine the corresponding spectral deviation data;

[0030] The spectral deviation feature vector is obtained by performing feature extraction on the spectral deviation data;

[0031] The spectral deviation feature vector is input into a machine learning model, and the optimal spectral adjustment strategy is generated through model calculation.

[0032] The spectrum of the UV light source is adjusted according to the optimal spectral adjustment strategy.

[0033] Optionally, in a sixth implementation of the first aspect of this application, the method further includes:

[0034] The spectral feature vector is generated by fuzzification;

[0035] Fuzzy rules are formulated based on the mapping relationship;

[0036] The fuzzy spectral feature vector is compared with the fuzzy rule to determine the fuzzy control output;

[0037] The fuzzy control output is converted into spectral adjustment parameters through defuzzification processing.

[0038] The spectrum of the UV light source is adjusted according to the spectral adjustment parameters.

[0039] A second aspect of this application provides an adaptive spectral adjustment device for a wafer inspection UV light source, the adaptive spectral adjustment device for the wafer inspection UV light source comprising:

[0040] The acquisition module is used to acquire the first spectral data for preliminary spectral detection of the target wafer;

[0041] The first determining module is used to perform variational mode decomposition processing on the first spectral data to determine the spectral characteristic parameters of the first spectral data.

[0042] The second determining module is used to process the spectral characteristic parameters by refining the composite multi-scale weighted multi-dimensional feature entropy to determine the spectral feature vector of the spectral characteristic parameters;

[0043] The third determining module is used to determine the mapping relationship between spectral features and wafer material properties by analyzing the spectral feature vector;

[0044] The control module is used to adaptively adjust the spectrum of the UV light source according to the mapping relationship.

[0045] A third aspect of this application provides an electronic device, including a memory and a processor, wherein the processor is configured to execute a computer program stored in the memory, and when the processor executes the computer program, it implements the steps of the adaptive spectral adjustment method for a wafer inspection UV light source provided in the first aspect of this application.

[0046] The fourth aspect of this application provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the steps in the adaptive spectral adjustment method for a wafer inspection UV light source provided in the first aspect of this application.

[0047] In summary, the adaptive spectral adjustment method, apparatus, and device for a UV light source for wafer inspection provided in this application obtain first spectral data for preliminary spectral inspection of a target wafer; perform variational mode decomposition on the first spectral data to determine spectral characteristic parameters; process the spectral characteristic parameters using refined composite multi-scale weighted multi-dimensional feature entropy to determine spectral feature vectors; analyze the spectral feature vectors to determine the mapping relationship between spectral features and wafer material properties; and adaptively adjust the spectrum of the UV light source based on the mapping relationship. By implementing this application, the adaptive adjustment of the UV light source spectrum based on the mapping relationship between spectral features and wafer material properties ensures that the spectral characteristics of the light source match the wafer inspection requirements, avoids unnecessary spectral redundancy or omissions, improves spectral utilization, shortens inspection time, and thus significantly improves inspection efficiency. Attached Figure Description

[0048] Figure 1 A schematic flowchart illustrating the adaptive spectral adjustment method for a wafer inspection UV light source provided in this application embodiment;

[0049] Figure 2 This is a schematic diagram of the program module of the adaptive spectral adjustment device for a wafer inspection UV light source provided in an embodiment of this application;

[0050] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0051] To make the inventive objectives, features, and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] To address the problem in related technologies that cannot dynamically adjust the UV light source spectrum based on wafer material characteristics, embodiments of this application provide an adaptive spectrum adjustment method for a wafer inspection UV light source, such as... Figure 1 This is a flowchart illustrating the adaptive spectral adjustment method for a wafer inspection UV light source provided in this embodiment. The adaptive spectral adjustment method for a wafer inspection UV light source includes the following steps:

[0053] Step 110: Obtain the first spectral data for preliminary spectral detection of the target wafer.

[0054] Specifically, in this embodiment, a target wafer is selected and placed on the inspection platform, ensuring that the wafer surface is clean and free of obvious defects. The selection of the target wafer should be based on actual production needs and can be wafers of different materials, processes, or batches. The target wafer is initially irradiated using a UV light source, the spectral range of which should cover the wavelength range required for wafer inspection. During irradiation, the stability and uniformity of the light source are maintained to ensure the representativeness of the acquired spectral data. A spectrometer is used to collect the wafer's reflectance spectral data; the spectrometer's resolution should be high enough to capture subtle changes in the spectral signal. During the acquisition process, information such as the wavelength range and intensity of the spectral data is recorded to form preliminary first spectral data.

[0055] Step 120: Perform variational mode decomposition on the first spectral data to determine the spectral characteristic parameters of the first spectral data.

[0056] Specifically, in this embodiment, the preliminary spectral data undergoes preprocessing, including smoothing and normalization, to remove noise and outliers and improve data quality. The variational mode decomposition (VMD) algorithm is then applied to decompose the preprocessed spectral data. The VMD algorithm decomposes the spectral data into multiple modal components, determining the spectral characteristic parameters of each modal component, including center frequency, bandwidth, and amplitude. The center frequency reflects the dominant frequency component of the modal component, the bandwidth reflects the frequency range of the modal component, and the amplitude reflects the intensity of the modal component. These spectral characteristic parameters are crucial for subsequent spectral feature extraction and analysis, providing key information about the properties of the wafer material.

[0057] In one optional implementation of this embodiment, the step of performing variational mode decomposition on the first spectral data to determine the spectral characteristic parameters of the first spectral data includes: decomposing the first spectral data into N modal components using a variational mode decomposition algorithm; analyzing each modal component to determine the modal characteristic parameters of each modal component; and integrating the modal characteristic parameters of the N modal components to determine the spectral characteristic parameters of the first spectral data.

[0058] Specifically, in this embodiment, the first spectral data acquired during wafer inspection typically contains rich spectral information, but this information may be masked by noise and complex background signals. To effectively extract spectral features, a variational mode decomposition (VMD) algorithm can be used to decompose the first spectral data. In this process, the VMD algorithm uses the alternating direction multiplier method (ADMM) to solve this variational problem. After decomposition, the first spectral data will be finely divided into several modal components. Different modal components represent the performance of the first spectral data within different frequency bandwidths and have unique physical meanings and characteristics. During the analysis, the modal characteristic parameters of each modal component are determined, including center frequency, bandwidth, and amplitude. The characteristic parameters of each modal component are then combined to form a parameter set that can comprehensively describe the spectral characteristics of the first spectral data. During integration, the importance of each modal component and its proportion in the overall data need to be considered to ensure that the integrated spectral characteristic parameters can accurately reflect the comprehensive characteristics of the original spectral data.

[0059] It should be noted that the VMD algorithm decomposes the data by constructing a mathematical model. The objective function of this model is shown below:

[0060] ,

[0061] Where K represents the number of modal components in the objective decomposition, k is the index of the modal component, and t represents the time variable. It is the k-th modal component. It is the center frequency of the k-th modal component. These are constraint parameters used to balance the bandwidth between modal components. This is the first spectral data. These are smoothing parameters used to control the smoothness of modal components. The time step is used to calculate the continuity of modal components.

[0062] It should be noted that the bandwidth constraint is as follows:

[0063] ,

[0064] This term is used to limit the bandwidth of each modal component by penalizing the modal component. This is achieved through high-frequency changes in parameters. This is used to balance the relationship between bandwidth constraints and reconstruction errors. In spectral data processing, this helps to highlight important spectral peaks while suppressing high-frequency noise.

[0065] Smoothness constraint term:

[0066] ,

[0067] This ensures the continuity and smoothness of each modal component over time t. Parameter It is a smoothing parameter used to adjust the intensity of smoothness, while It is the time step, representing the interval between consecutive time points. In spectral data, this helps maintain the shape of spectral peaks and prevents excessive frequency jumps.

[0068] Reconstruction error term:

[0069] ,

[0070] This item is for calculating the raw data. Sum of all modal components The difference between them is squared and integrated to ensure that the sum of all modal components can accurately reconstruct the first spectral data.

[0071] Step 130: Determine the spectral feature vector of the spectral feature parameters by processing the spectral characteristic parameters through refined composite multi-scale weighted multi-dimensional feature entropy.

[0072] Specifically, in this embodiment, multi-scale analysis is performed on the spectral characteristic parameters of each modal component, and the spectral features are extracted using the Refined Composite Multiscale Weighted Multidimensional Feature Entropy (RCMWMFE) algorithm. The RCMWMFE algorithm captures the complex features of the spectral signal by calculating the entropy values ​​of the spectral characteristic parameters at different scales. The entropy value of each modal component at different scales is calculated to form a multi-scale entropy matrix. The entropy value can be calculated using methods such as Shannon entropy, Renyi entropy, or Tsallis entropy, and a suitable entropy definition is selected according to actual needs. The multi-scale entropy matrix is ​​weighted, using the standard deviation as the weight, to weight and fuse the entropy values ​​at different scales. Weighting highlights the main features, suppresses secondary features, and improves the representativeness of the feature vector. The weighted entropy values ​​are integrated into a spectral feature vector, which can comprehensively describe the complex features of the spectral signal. The dimension of the spectral feature vector is determined according to actual needs and can contain feature information from multiple modal components.

[0073] In one optional implementation of this embodiment, the step of determining the spectral feature vector of the spectral characteristic parameters by refining the composite multi-scale weighted multi-dimensional feature entropy processing of the spectral characteristic parameters includes: obtaining the multi-scale spectral features corresponding to the spectral characteristic parameters by determining the entropy values ​​of the spectral characteristic parameters at different scales; obtaining the weighted spectral features corresponding to the spectral characteristic parameters by weighting the multi-scale spectral features; and determining the spectral feature vector of the spectral characteristic parameters by performing feature extraction on the weighted spectral features.

[0074] Specifically, in this embodiment, spectral analysis is an important technique in wafer inspection. By analyzing the spectral response of a wafer at different wavelengths, its internal structure and defect information can be obtained. Multi-scale spectral feature extraction involves analyzing spectral data at different scales to obtain more comprehensive feature information. For example, a multi-scale entropy algorithm can be used to calculate entropy values ​​at different scales, thereby assessing the complexity and irregularity of the spectral signal. By weighting multi-scale spectral features, important features can be highlighted and unimportant features suppressed, thus improving the representativeness and discriminative power of the spectral features. In wafer inspection, spectral features at different scales have different sensitivities to defects. Weighting can enhance the defect detection capability. For example, spectral features at different scales can be weighted according to the magnitude of the multi-scale entropy value. Scales with larger entropy values ​​indicate higher complexity and irregularity of the spectral signal at that scale, and therefore should be given a larger weight. By extracting features from the weighted spectral features, high-dimensional spectral data can be reduced to low-dimensional feature vectors, retaining the main spectral feature information, which facilitates subsequent analysis and application. In wafer inspection, features can be extracted using methods such as principal component analysis (PCA), linear discriminant analysis (LDA), or independent component analysis (ICA). For example, PCA can be used to calculate the covariance matrix of weighted spectral features, solve for the eigenvalues ​​and eigenvectors of the covariance matrix, and select the eigenvectors with larger eigenvalues ​​as principal components, thereby forming spectral feature vectors.

[0075] Step 140: Determine the mapping relationship between spectral features and wafer material properties by analyzing spectral feature vectors.

[0076] Specifically, in this embodiment, after obtaining the spectral feature vector, it is analyzed in depth to determine the mapping relationship between spectral features and wafer material properties. By establishing the connection between spectral features and material properties, the behavior of the wafer under different spectral conditions can be better understood. For example, certain spectral features may correspond to characteristics such as wafer defects, impurities, or crystal structure. Through the accumulation and analysis of a large amount of data, an accurate mapping model is established to provide guidance for adaptive adjustment of the spectrum.

[0077] In one optional embodiment of this example, the step of determining the mapping relationship between spectral features and wafer material properties by analyzing spectral feature vectors includes: generating spectral feature clusters by performing cluster analysis on the spectral feature vectors; wherein, the spectral feature clusters are clusters of wafer materials with the same type of spectral features; performing statistical analysis on the spectral feature vectors corresponding to the spectral feature clusters to determine the wafer material properties corresponding to the spectral feature clusters; and establishing a mapping relationship between spectral features and wafer material properties based on the spectral feature clusters and the corresponding wafer material properties.

[0078] Specifically, in this embodiment, the spectral feature vector is obtained by extracting and processing the spectral data of the wafer material, containing the spectral response information of the wafer material at different wavelengths. Through cluster analysis, wafer materials with similar spectral characteristics can be grouped into one category, forming spectral feature clusters. Cluster analysis can employ various algorithms, such as the K-Means algorithm and hierarchical clustering algorithms. These algorithms divide the spectral feature vectors into different clusters by calculating the similarity or distance between them.

[0079] Statistical analysis is performed on the spectral eigenvectors corresponding to the generated spectral feature clusters to determine the wafer material properties. This statistical analysis may include calculating statistics such as the mean, variance, and standard deviation of the spectral eigenvectors. These statistics reflect the central tendency and dispersion of the wafer material's spectral response. For example, the mean represents the average spectral response intensity of the wafer material at different wavelengths, while the variance reflects the fluctuations in the spectral response. By analyzing these statistics, further characteristics of the wafer material, such as purity and defect types, can be determined.

[0080] Based on spectral feature clusters and corresponding wafer material properties, a mapping relationship between spectral features and wafer material properties is established. This mapping relationship can be implemented using machine learning algorithms, such as Support Vector Machines (SVM), decision trees, and random forests. These algorithms take spectral feature vectors as input and wafer material properties as output, learning the mapping relationship between spectral features and wafer material properties through training. For example, the SVM algorithm separates spectral feature vectors of different categories by finding an optimal hyperplane, thereby achieving the classification and prediction of wafer material properties. Once the mapping relationship between spectral features and wafer material properties is established, the material properties of unknown wafer materials can be predicted by analyzing their spectral characteristics, providing strong support for wafer inspection and quality control.

[0081] It should be noted that the cluster analysis formula is expressed as:

[0082] ,

[0083] in, The k-th cluster represents a class of wafer materials with similar spectral characteristics. Let be the cluster center of the k-th cluster, and represent the mean of all spectral eigenvectors within that cluster. Let be the spectral feature vector, representing the spectral features of the i-th sample. For weight parameters, Let be the parameters that minimize the objective function, and let be the cluster that minimizes the sum of squared distances between samples within the cluster and the cluster center. Here, the clustering objective function is used to minimize the sum of squared distances between samples within a cluster and the cluster center. This introduces a weight parameter. This term is used to balance the relative importance of the objective function in the automatic spectral adjustment process. By adding this term, the dynamic changes in spectral features can be considered during clustering, thus better adapting to the actual needs in wafer inspection.

[0084] The statistical analysis formula is expressed as follows:

[0085] ,

[0086] ,

[0087] in, The mean of the k-th cluster represents the average value of all spectral eigenvectors within that cluster. The variance of the k-th cluster represents the average of the sum of squared deviations of all spectral eigenvectors within that cluster from the mean. This introduces a weight parameter. This term is used to balance the relative importance of the objective function in the automatic spectral adjustment process. By adding this term, the dynamic changes in spectral characteristics can be taken into account when calculating the mean, thus better reflecting the properties of the wafer material. This term introduces a weighting parameter δ to balance the relative importance of the objective function in the automatic spectral adjustment process. By adding this term, the dynamic changes in spectral characteristics can be taken into account when calculating the variance, thereby better assessing the stability of wafer material properties.

[0088] Step 150: Adaptively adjust the spectrum of the UV light source according to the mapping relationship.

[0089] Specifically, in this embodiment, the spectrum of the UV light source is adaptively adjusted according to the established mapping relationship. By adjusting the spectral characteristics of the UV light source, such as center frequency, bandwidth, and amplitude, it is matched to the characteristics of the wafer material. This adaptive adjustment can improve the accuracy and efficiency of wafer inspection and optimize the inspection process. For example, for wafers with specific defect characteristics, the detection capability for these defects can be enhanced by adjusting the spectrum of the UV light source, thereby improving the sensitivity and reliability of the inspection.

[0090] In one optional embodiment of this example, the step of adaptively adjusting the spectrum of the UV light source according to the mapping relationship includes: determining the spectral requirements of the target wafer according to the mapping relationship; making a preliminary adjustment to the spectrum of the UV light source according to the spectral requirements; obtaining the second spectral data of the UV light source after the preliminary adjustment; determining whether the second spectral data meets the spectral requirements of the target wafer according to the second spectral data and the mapping relationship; if the second spectral data does not meet the spectral requirements, making a secondary adjustment to the spectrum of the UV light source until the second spectral data meets the spectral requirements.

[0091] Specifically, in this embodiment, the spectral requirements of the target wafer are first determined based on the established mapping relationship. Spectral requirements refer to the specific requirements for the spectral characteristics (such as wavelength range, intensity distribution, etc.) of the UV light source in a specific application scenario, based on the material properties of the target wafer and the detection purpose. This ensures that the UV light source can provide a spectrum that matches the material properties of the target wafer, thereby achieving effective detection and analysis. For example, in wafer inspection, if the target wafer's material property is high-purity silicon, the spectral requirement might be a high-intensity spectrum within a specific wavelength range to better excite the spectral characteristics of the silicon material. Then, the wavelength and intensity of the UV light source are adjusted according to the spectral requirements to make its output spectrum closer to the spectrum required by the target wafer. For example, if the target wafer requires a spectrum in the 300nm to 400nm wavelength range, the output wavelength of the UV light source needs to be adjusted to this range, and the intensity also needs to be adjusted accordingly. After the initial adjustment, the second spectral data of the UV light source is acquired using equipment such as a spectrometer. Based on the second spectral data, it is determined whether the spectrum of the UV light source meets the spectral requirements of the target wafer. If there is a deviation between the second spectral data and the spectral requirements of the target wafer, the spectrum of the UV light source needs to be adjusted a second time. Through iterative adjustment and judgment, the spectral output of the UV light source is finally made to meet the spectral requirements of the target wafer.

[0092] In one optional implementation of this embodiment, the second spectral data is compared with the spectral requirements to determine the corresponding spectral deviation data; a spectral deviation feature vector is obtained by extracting features from the spectral deviation data; the spectral deviation feature vector is input into a machine learning model, and the optimal spectral adjustment strategy is generated by the model calculation; the spectrum of the UV light source is adjusted according to the optimal spectral adjustment strategy.

[0093] Specifically, in this embodiment, the process of comparing the second spectral data with the spectral requirements to determine the spectral deviation data first requires a detailed analysis of the second spectral data. By comparing it with the preset spectral requirements, the difference between the two is calculated, i.e., the spectral deviation data. This comparison process can use various methods, such as Euclidean distance and Mahalanobis distance. For example, Euclidean distance can be obtained by calculating the square root of the sum of the squares of the differences between the two spectral data at each wavelength. In this way, the difference between the second spectral data and the spectral requirements can be quantified, thereby determining the spectral deviation data. Feature extraction can reduce the dimensionality of the high-dimensional spectral deviation data to a low-dimensional feature space, extracting the most representative features. For example, PCA calculates the covariance matrix of the spectral deviation data, finds the direction with the largest variance in the data, i.e., the principal components, and then projects the data onto these principal components to obtain the spectral deviation feature vector. This process not only reduces the dimensionality of the data but also retains the main information of the data, providing effective features for the subsequent machine learning model input. The machine learning model can be a support vector machine (SVM), a neural network (NN), a decision tree (DT), etc. These models, by learning the relationship between spectral deviation feature vectors and spectral adjustment strategies, can predict the optimal spectral adjustment strategy based on the input feature vectors. For example, neural networks, through multiple layers of neurons and activation functions, can learn complex nonlinear relationships, thus more accurately predicting the optimal spectral adjustment strategy. The optimal spectral adjustment strategy generated by the model is then applied to the control system of a UV light source to precisely adjust its spectrum. For instance, if the optimal spectral adjustment strategy indicates a need to increase the spectral intensity of a specific wavelength, the UV light source control system will adjust corresponding parameters, such as current and voltage, to achieve this increase. In this way, it can be ensured that the spectral output of the UV light source meets the spectral requirements of the target wafer, thereby improving the accuracy and efficiency of wafer inspection.

[0094] In one optional implementation of this embodiment, a fuzzy spectral feature vector is generated by fuzzification of the spectral feature vector; fuzzy rules are formulated according to the mapping relationship; the fuzzy spectral feature vector is compared with the fuzzy rules to determine the fuzzy control output; the fuzzy control output is converted into spectral adjustment parameters by defuzzification; and the spectrum of the UV light source is adjusted according to the spectral adjustment parameters.

[0095] Specifically, in this embodiment, a fuzzy spectral feature vector is generated by fuzzification of the spectral feature vector. Fuzzification converts the precise spectral feature vector into a fuzzy set. The spectral feature vector can be represented as... The goal of fuzzing is to separate each feature parameter Convert to its corresponding fuzzy set, member functions of the fuzzy set It can be represented using triangles, trapezoids, or Gaussian functions. For example, using a Gaussian function as the fuzzy member function, it can be represented as:

[0096] ,

[0097] in, It is a characteristic parameter The central value, It is the standard deviation of the fuzzy set. This formula can be used to calculate the standard deviation of each feature parameter. Convert to its fuzzy value to generate a fuzzy spectral feature vector. Fuzzy rules define the relationship between fuzzy spectral feature vectors and output spectral adjustment parameters. Typically, fuzzy rules can be expressed in an "if-then" form, for example: "If..." It is high and If it's low, then the output spectrum adjustment parameter... "Medium". These rules can be formulated using expert knowledge or data-driven methods. In data-driven methods, historical data and machine learning algorithms can be used to automatically generate fuzzy rules. Then, the fuzzy spectral feature vector is compared with the fuzzy rules to determine the fuzzy control output. The comparison process involves comparing the fuzzy spectral feature vector... The rules are matched against those in the fuzzy rule base, and the activation degree of each rule is calculated. Assuming there are M fuzzy rules, the activation degree of each rule can be calculated using the intersection of fuzzy sets. For example, for the j-th fuzzy rule, its activation degree is... It can be represented as:

[0098] ,

[0099] By calculating the activation degree of each rule, the fuzzy control output can be determined. Defuzzification processing is the process of resolving fuzzy control outputs. This is converted into the precise spectral adjustment parameter Y. Commonly used defuzzification methods include the centroid method and the maximum membership method. The centroid method determines the precise value by calculating the centroid of the fuzzy set; its formula can be expressed as:

[0100] ,

[0101] in, This is a member function of the fuzzy control output. Using this formula, the precise spectral adjustment parameter Y can be calculated. Finally, the spectrum of the UV light source is adjusted based on the spectral adjustment parameter Y. The spectral adjustment parameter Y can include specific parameters such as the light source's intensity, wavelength, and bandwidth. By adjusting control parameters such as the UV light source's current and temperature, the output spectrum of the light source is adjusted in real time.

[0102] According to the adaptive spectral adjustment method for a UV light source for wafer inspection provided in this application, first spectral data for preliminary spectral detection of a target wafer is obtained; variational mode decomposition is performed on the first spectral data to determine the spectral characteristic parameters; the spectral characteristic parameters are processed by refined composite multi-scale weighted multi-dimensional feature entropy to determine the spectral feature vectors of the spectral characteristic parameters; the mapping relationship between the spectral features and the wafer material characteristics is determined by analyzing the spectral feature vectors; and the spectrum of the UV light source is adaptively adjusted according to the mapping relationship. By implementing this application, the adaptive adjustment of the UV light source spectrum based on the mapping relationship between spectral characteristics and wafer material characteristics ensures that the spectral characteristics of the light source match the wafer inspection requirements, avoids unnecessary spectral redundancy or omissions, improves spectral utilization, shortens inspection time, and thus significantly improves inspection efficiency.

[0103] Figure 2 An adaptive spectral adjustment device for a wafer inspection UV light source is provided as an embodiment of this application. This adaptive spectral adjustment device for a wafer inspection UV light source can be used to implement the adaptive spectral adjustment method for a wafer inspection UV light source in the aforementioned embodiments. Figure 2 As shown, the adaptive spectral adjustment device for the UV light source of the wafer inspection mainly includes:

[0104] The acquisition module 10 is used to acquire the first spectral data for preliminary spectral detection of the target wafer;

[0105] The first determining module 20 is used to perform variational mode decomposition on the first spectral data to determine the spectral characteristic parameters of the first spectral data.

[0106] The second determining module 30 is used to process the spectral characteristic parameters by refining the composite multi-scale weighted multi-dimensional feature entropy to determine the spectral feature vector of the spectral characteristic parameters;

[0107] The third determining module 40 is used to determine the mapping relationship between spectral features and wafer material properties by analyzing the spectral feature vector;

[0108] The control module 50 is used to adaptively adjust the spectrum of the UV light source according to the mapping relationship.

[0109] In one optional implementation of this embodiment, the first determining module is specifically used to: decompose the first spectral data into N modal components using a variational mode decomposition algorithm; wherein N is an integer greater than or equal to 1; analyze each modal component to determine the modal characteristic parameters of each modal component; and integrate the modal characteristic parameters of the N modal components to determine the spectral characteristic parameters of the first spectral data.

[0110] In one optional implementation of this embodiment, the second determining module is specifically used to: obtain multi-scale spectral features corresponding to the spectral characteristic parameters by determining the entropy values ​​of the spectral characteristic parameters at different scales; obtain weighted spectral features corresponding to the spectral characteristic parameters by weighting the multi-scale spectral features; and determine the spectral feature vector of the spectral characteristic parameters by performing feature extraction on the weighted spectral features.

[0111] In one optional implementation of this embodiment, the third determining module is specifically used to: generate spectral feature clusters by performing cluster analysis on the spectral feature vectors; wherein the spectral feature clusters are clusters of wafer materials with the same spectral features; perform statistical analysis on the spectral feature vectors corresponding to the spectral feature clusters to determine the wafer material characteristics corresponding to the spectral feature clusters; and establish a mapping relationship between spectral features and wafer material characteristics based on the spectral feature clusters and the corresponding wafer material characteristics.

[0112] In one optional implementation of this embodiment, the control module is specifically used to: determine the spectral requirements of the target wafer according to the mapping relationship; make preliminary adjustments to the spectrum of the UV light source according to the spectral requirements; obtain the second spectral data of the UV light source after preliminary adjustment; determine whether the second spectral data meets the spectral requirements of the target wafer according to the second spectral data and the mapping relationship; if the second spectral data does not meet the spectral requirements, make a second adjustment to the spectrum of the UV light source until the second spectral data meets the spectral requirements.

[0113] In an optional implementation of this embodiment, the control module is further configured to: compare the second spectral data with the spectral requirement to determine the corresponding spectral deviation data; obtain a spectral deviation feature vector by performing feature extraction on the spectral deviation data; input the spectral deviation feature vector into a machine learning model and generate an optimal spectral adjustment strategy through model calculation; and adjust the spectrum of the UV light source according to the optimal spectral adjustment strategy.

[0114] In an optional embodiment of this example, the control module is further configured to: generate a fuzzy spectral feature vector by fuzzifying the spectral feature vector; formulate fuzzy rules according to the mapping relationship; compare the fuzzy spectral feature vector with the fuzzy rules to determine the fuzzy control output; convert the fuzzy control output into spectral adjustment parameters by defuzzification; and adjust the spectrum of the UV light source according to the spectral adjustment parameters.

[0115] According to the adaptive spectral adjustment device for a UV light source for wafer inspection provided in this application, first spectral data for preliminary spectral detection of a target wafer is acquired; variational mode decomposition is performed on the first spectral data to determine the spectral characteristic parameters; the spectral characteristic parameters are processed by refined composite multi-scale weighted multidimensional feature entropy to determine the spectral feature vector; the mapping relationship between the spectral characteristics and the wafer material characteristics is determined by analyzing the spectral feature vector; and the spectrum of the UV light source is adaptively adjusted according to the mapping relationship. By implementing this application, the adaptive adjustment of the UV light source spectrum based on the mapping relationship between spectral characteristics and wafer material characteristics ensures that the spectral characteristics of the light source match the wafer inspection requirements, avoids unnecessary spectral redundancy or omissions, improves spectral utilization, shortens inspection time, and thus significantly improves inspection efficiency.

[0116] According to the scheme provided in this application Figure 3 An electronic device is provided as an embodiment of this application. This electronic device can be used to implement the adaptive spectral adjustment method for the wafer inspection UV light source in the foregoing embodiments, and mainly includes:

[0117] The system includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and executable on the processor 302. The memory 301 and the processor 302 are connected via communication. When the processor 302 executes the computer program 303, it implements the adaptive spectral adjustment method for the wafer inspection UV light source described in the foregoing embodiments. The number of processors can be one or more.

[0118] The memory 301 can be a high-speed random access memory (RAM) or a non-volatile memory, such as a disk storage device. The memory 301 is used to store executable program code, and the processor 302 is coupled to the memory 301.

[0119] Furthermore, embodiments of this application also provide a computer-readable storage medium, which may be disposed in the electronic device described in the above embodiments, and the computer-readable storage medium may be as described above. Figure 3 The memory in the illustrated embodiment.

[0120] The computer-readable storage medium stores a computer program that, when executed by a processor, implements the adaptive spectral adjustment method for the wafer inspection UV light source described in the foregoing embodiments. Furthermore, the computer-readable storage medium can also be a USB flash drive, external hard drive, read-only memory (ROM), RAM, magnetic disk, or optical disk, or any other medium capable of storing program code.

[0121] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0122] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it 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 all or part 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.

[0123] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications 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.

Claims

1. An adaptive spectral adjustment method for a UV light source for wafer inspection, characterized in that, The method comprises the following steps: acquiring first spectral data of a preliminary spectral detection of a target wafer; performing variational mode decomposition processing on the first spectral data to determine spectral characteristic parameters of the first spectral data; determining a spectral feature vector of the spectral characteristic parameters by performing fine composite multi-scale weighted multi-dimensional feature entropy processing on the spectral characteristic parameters, comprising: acquiring multi-scale spectral features corresponding to the spectral characteristic parameters by determining the entropy values of the spectral characteristic parameters at different scales; acquiring weighted spectral features corresponding to the spectral characteristic parameters by performing weighted processing on the multi-scale spectral features; and determining a spectral feature vector of the spectral characteristic parameters by performing feature extraction on the weighted spectral features; determining a mapping relationship between spectral features and wafer material characteristics by analyzing the spectral feature vector, comprising: generating spectral feature clusters by performing clustering analysis on the spectral feature vector; wherein the spectral feature clusters are clusters of wafer materials with the same spectral features; performing statistical analysis on the spectral feature vectors corresponding to the spectral feature clusters to determine the wafer material characteristics corresponding to the spectral feature clusters; and establishing a mapping relationship between spectral features and wafer material characteristics according to the spectral feature clusters and the corresponding wafer material characteristics; adaptively adjusting the spectrum of a UV light source according to the mapping relationship.

2. The method of claim 1, wherein the method further comprises: The step of performing variational mode decomposition processing on the first spectral data to determine the spectral characteristic parameters of the first spectral data comprises the following steps: decomposing the first spectral data into N modal components by a variational mode decomposition algorithm; wherein N is an integer greater than or equal to 1; analyzing each of the modal components to determine modal characteristic parameters of each of the modal components; integrating the modal characteristic parameters of the N modal components to determine the spectral characteristic parameters of the first spectral data.

3. The method of claim 1, wherein the method further comprises: The step of adaptively adjusting the spectrum of a UV light source according to the mapping relationship comprises the following steps: determining the spectral requirements of the target wafer according to the mapping relationship; preliminarily adjusting the spectrum of the UV light source according to the spectral requirements; acquiring second spectral data of the UV light source after the preliminary adjustment; judging whether the second spectral data meets the spectral requirements of the target wafer according to the second spectral data and the mapping relationship; if the second spectral data does not meet the spectral requirements, performing secondary adjustment on the spectrum of the UV light source until the second spectral data meets the spectral requirements.

4. The method of claim 3, wherein the method further comprises: The method further comprises the following steps: comparing the second spectral data with the spectral requirements to determine corresponding spectral deviation data; acquiring a spectral deviation feature vector by performing feature extraction on the spectral deviation data; inputting the spectral deviation feature vector into a machine learning model and generating an optimal spectral adjustment strategy by model calculation; adjusting the spectrum of the UV light source according to the optimal spectral adjustment strategy.

5. The method of claim 1, wherein the method further comprises: The method further comprises the following steps: generating a fuzzy spectral feature vector by fuzzifying the spectral feature vector; formulating fuzzy rules according to the mapping relationship; determining a fuzzy control output by comparing the fuzzy spectral feature vector with the fuzzy rules; transforming the fuzzy control output into a spectral adjustment parameter through de-fuzzification processing; adjusting a spectrum of the UV light source according to the spectral adjustment parameter.

6. An adaptive spectral adjustment device for a wafer inspection UV light source, characterized by, The wafer detection UV light source adaptive spectrum adjustment device comprises: an acquisition module configured to acquire first spectral data obtained by performing preliminary spectral detection on a target wafer; a first determination module configured to perform variational mode decomposition processing on the first spectral data to determine spectral characteristic parameters of the first spectral data; a second determination module configured to determine a spectral feature vector of the spectral characteristic parameters by performing fine composite multi-scale weighted multi-dimensional feature entropy processing on the spectral characteristic parameters, including: obtaining multi-scale spectral features corresponding to the spectral characteristic parameters by determining entropy values of the spectral characteristic parameters at different scales; obtaining weighted spectral features corresponding to the spectral characteristic parameters by performing weighted processing on the multi-scale spectral features; and determining the spectral feature vector of the spectral characteristic parameters by performing feature extraction on the weighted spectral features; a third determination module configured to determine a mapping relationship between spectral features and wafer material characteristics by analyzing the spectral feature vector, including: generating spectral feature clusters by performing clustering analysis on the spectral feature vector; wherein the spectral feature clusters are clusters of wafer materials with the same spectral features; performing statistical analysis on spectral feature vectors corresponding to the spectral feature clusters to determine wafer material characteristics corresponding to the spectral feature clusters; and establishing the mapping relationship between the spectral features and the wafer material characteristics according to the spectral feature clusters and the corresponding wafer material characteristics; a control module configured to perform adaptive adjustment on a spectrum of a UV light source according to the mapping relationship.

7. An electronic device, comprising: comprise a memory and a processor, wherein: the processor is configured to execute a computer program stored in the memory; the processor, when executing the computer program, implements the steps of the wafer detection UV light source adaptive spectrum adjustment method in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, the computer program, when executed by the processor, implements the steps of the wafer detection UV light source adaptive spectrum adjustment method in any one of claims 1 to 5.

Images