Wafer cutting parameter optimization method and system

Abstract

The invention discloses a wafer cutting parameter optimization method and system, and the method comprises the steps: collecting and processing data, carrying out the function fitting and machine learning model training based on the collected data, determining the interval range of a modification region and the initial parameters of a cutting process according to the wafer characteristic data required by a project, and carrying out the optimization of the cutting process. Wherein the initial parameters of the cutting process are determined by adopting a K-means clustering algorithm, and finally the initial parameters of the cutting process are optimized and adjusted based on a machine learning model and a fitting function. Compared with the prior art, according to the technical scheme, the stability of the cutting process and the product quality can be remarkably improved, and meanwhile the production cost is reduced. Through application of the machine learning model, the automation degree is greatly improved, dependence on artificial experience is reduced, parameter optimization of the semiconductor invisible cutting process is more efficient and accurate, and increasingly strict industrial production requirements can be met.

Description

Technical Field

[0001] The present invention belongs to the technical field of parameter control of semiconductor stealth cutting process, and in particular relates to a wafer cutting parameter optimization method and system. Background Art

[0002] The invisible cutting process in the semiconductor manufacturing process is a key technology, which is widely used in the cutting of ultra-thin wafers, the manufacture of optoelectronic devices, and micro sensors. With the increasing requirements of the integrated circuit industry for wafer size and precision, the invisible cutting process has become the mainstream choice in modern semiconductor processing because it can avoid the surface damage of the wafer caused by traditional cutting technology. This process forms a modified layer by focusing the laser beam at a specific depth inside the wafer, which produces microcracks or stress concentration areas inside the material. When external force acts on the wafer, these cracks will expand and eventually cause the wafer to separate, thereby achieving wafer segmentation. However, the realization of the invisible cutting process depends on the optimization of multiple process parameters. Since these process parameters have complex interrelationships and may be affected by factors such as wafer material, thickness, and surface quality in the actual production process, how to effectively control and adjust these parameters to ensure cutting quality and output has become an urgent problem to be solved in semiconductor processing.

[0003] Traditional invisible cutting process debugging usually relies on experienced engineers to conduct repeated experiments and adjustments. In actual operation, engineers determine the appropriate process parameter window through experiments and blind adjustments. However, since the relationship between process parameters and cutting effects is often nonlinear and difficult to predict, this experience-based adjustment method is not only inefficient, but may also lead to long downtime and increased production costs.

[0004] In addition, the optimization process of the invisible cutting process usually requires a lot of processing equipment, measurement tools and experimental time. Each parameter adjustment may involve a lot of testing and verification, which further increases the time cost and resource consumption in the production process. Therefore, how to quickly and efficiently adjust and optimize process parameters while ensuring cutting quality has become a key issue to improve production efficiency, reduce costs and enhance product stability.

[0005] In order to achieve this goal, traditional manual debugging methods urgently need to develop in a more intelligent and automated direction. In recent years, with the development of machine learning and artificial intelligence technologies, these emerging technologies have made breakthroughs in many fields and have begun to be applied in semiconductor manufacturing. At present, in the semiconductor manufacturing process, although there are some process optimization methods based on machine learning, most methods still rely on a large amount of training data and show certain limitations under complex and changing process conditions. Summary of the invention

[0006] In order to solve the problem of uncontrollable parameter optimization and trial cutting time cost of invisible cutting process, the present invention provides a wafer cutting parameter optimization method and system, which can effectively cope with complex and changeable wafer incoming materials and design requirements, and improve the production efficiency of proofing and the consistency of product quality.

[0007] The technical solution provided by the present invention is as follows: A method for optimizing wafer cutting parameters comprises the following steps: Obtain actual wafer cutting data, including wafer characteristic data, cutting process parameters and corresponding cutting result data; wafer characteristic data includes target thickness, safety distance and wafer thickness, cutting process parameters include laser mode, laser defocus point and laser power, and cutting result data includes modified area length, modified area distance BCD and cutting effect BHC; Pre-process the collected data, and quantify the scores of the corresponding wafer characteristic data and the cutting quality under the cutting process parameters according to the cutting result data; The relationship between the laser defocus point, laser power and the length of the modified area generated under different laser modes is fitted with a function to obtain a modified area length fitting function; the relationship between the modified area distance BCD and the laser defocus point is fitted with a function to obtain a laser defocus point fitting function; Training a plurality of machine learning models respectively, wherein the input of the machine learning model is a cutting process parameter, and the output of the machine learning model is a cutting quality score; The modified area range and the initial parameters of the cutting process are determined based on the wafer characteristic data required for the project, and then the initial parameters of the cutting process are optimized and adjusted based on the machine learning model and fitting function.

[0008] Furthermore, the optimization adjustment includes: inputting the initial parameters of the cutting process into multiple machine learning models to obtain the average cutting quality score under the parameters. If the cutting quality score meets the requirements, the parameters are used as the final cutting process parameters; otherwise, the length of the modified area is calculated according to the modified area length fitting function under the current parameters, and based on this, while satisfying the modified area interval range, the deflection point value in the parameter is adjusted according to the laser deflection point fitting function, and the adjusted parameters are input into the machine learning model to obtain the average cutting quality score until it meets the requirements.

[0009] Furthermore, the cutting effect BHC reflects the overall quality of the wafer after cutting through electronic measurement photos of the wafer circuit surface and electronic measurement photos of the laser scattering distance, and the scoring criteria of the quantitative scoring include the state of the wafer cutting marks.

[0010] Furthermore, the modified region length fitting function is a polynomial function obtained based on least squares fitting, and the laser defocus point fitting function is: BCD down = T - ( n 2+ k) L ， BCD up = BCD down + h, Among them, BCD down Indicates the lowest vertical height of the modified area, BCD up represents the maximum height of the modified area in the vertical direction, h represents the length of the modified area, T Indicates the wafer thickness, n 2 represents the refractive index of the laser incident on the wafer, k is the laser refraction correction coefficient, L Indicates the laser defocus point.

[0011] Furthermore, the calculation formula for the modified area interval range is: BCD down >De + 0.5D, BCD up <De - 0.5D, Wherein, De represents the target thickness in the wafer characteristic data, and D represents the safety distance.

[0012] Furthermore, the initial parameters of the cutting process are determined based on a K-means clustering algorithm, including the steps of: First, cluster the obtained cutting result data to obtain the cluster centers of each category: S1. Select m groups of data from the cutting result data as the initial cluster centers, and the dimensions of each sample point include the cutting quality score, the length of the modified area h, and the distance BCD of the modified area; S2. Traverse each sample in the cutting result data, calculate its distance to the m initial cluster centers respectively, and divide it into the category corresponding to the cluster center with the smallest distance; S3. Calculate the centroid of all samples contained in each category and use it as the cluster center of the category; S4. Repeat S2 and S3 until the iteration termination condition is reached; Then, the Euclidean distances between the wafer characteristic data corresponding to the cluster centers of each category and the wafer characteristic data required for the project are calculated respectively, and the cutting process parameters corresponding to the cluster center with the smallest Euclidean distance are used as the initial process parameters.

[0013] Furthermore, the iteration termination condition is that the cluster center of each category no longer changes or the change error is less than a set threshold or the number of iterations reaches a set value; if the cluster center is not an existing sample data, the sample data with the smallest distance to the cluster center is selected as the replacement value.

[0014] Preferably, the machine learning model adopts a support vector machine.

[0015] A wafer cutting parameter optimization system based on the above method comprises: A data acquisition and storage module is used to acquire and store actual wafer cutting data, including wafer characteristic data, cutting process parameters and corresponding cutting result data; The data processing module is used to pre-process the collected data and quantify the scores of the corresponding wafer characteristic data and the cutting quality under the cutting process parameters according to the cutting result data; Function fitting module, used for performing function fitting on the relationship between laser defocus point, laser power and length of the generated modified area under different laser modes, and performing function fitting on the relationship between modified area distance BCD and laser defocus point; A model training module, used to train multiple machine learning models respectively, wherein the input of the machine learning model is a cutting process parameter, and the output of the machine learning model is a cutting quality score; An initial parameter determination module is used to determine the initial parameters of the cutting process according to the wafer characteristic data required by the project; The parameter optimization module is used to optimize and adjust the initial parameters of the cutting process according to the machine learning model and fitting function while meeting the range of the modified area.

[0016] Furthermore, it also includes a model update module for updating and optimizing the machine learning model using continuously collected actual wafer cutting data.

[0017] Compared with the prior art, the present invention significantly improves the efficiency and accuracy of the optimization of the process parameters of ultra-thin wafer stealth cutting by introducing a machine learning model. The traditional stealth cutting process parameter optimization mainly relies on the experience of engineers and a large number of trial and error processes, resulting in a long debugging cycle and high labor costs. However, the present invention, by adopting small sample learning, can establish an effective model with relatively few data samples, quickly capture the complex relationship between different process parameters, and thus greatly reduce the number of experiments and optimization time required.

[0018] Traditional process optimization methods are usually unable to accurately identify the nonlinearity and interactions between process parameters, resulting in unstable results during the optimization process, and even in some cases, the inability to find the optimal process window. This is because traditional optimization methods cannot effectively handle the interdependence between different parameters, and their debugging process is usually blind and tentative, making it difficult to cope with complex process adjustment requirements. The present invention, through a machine learning model, can make effective predictions based on existing small sample data. The model can learn from historical data the impact of each process parameter on the cutting effect, and can dynamically optimize the parameter combination, avoiding the blindness of manual debugging.

[0019] In addition, the present invention can find the appropriate process parameter window for newly imported wafers or new materials in a relatively short time. In actual production, changes in wafer materials, sizes, and thicknesses may cause the effectiveness of process parameters under traditional methods to decrease, requiring a large number of readjustments and tests. Through machine learning, the system can automatically adjust parameters based on a small amount of test data, quickly adapt to the production needs of new products, and greatly improve the flexibility and production efficiency of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0021] Figure 1 is a schematic diagram of parameters related to the modified area provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of a process of function fitting and model training provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of a process flow for optimizing process parameters provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, other embodiments obtained by ordinary technicians in this field without making creative work are all within the scope of protection of the present invention.

[0023] Embodiment 1 This embodiment provides a method for optimizing wafer cutting parameters, comprising the steps of: 1. Data Collection Conduct DOE (Design of Experiments) experiments to collect actual wafer cutting data sets, including wafer characteristic data and cutting process parameters, as well as corresponding cutting result data.

[0024] Wafer characteristic data include target thickness De, safety distance D and wafer thickness T. Target thickness De refers to the ideal thickness of the wafer after cutting, safety distance D refers to the thickness of the material that needs to be removed during the grinding stage, and wafer thickness T refers to the thickness of the wafer before cutting. Cutting process parameters include laser mode SC, laser defocus point L and laser power P. The choice of laser mode SC affects the cutting path and efficiency, such as continuous mode or pulse mode. The defocus point L determines the relative distance between the laser beam and the material surface, and the laser power determines the energy input during the cutting process, thereby affecting the cutting speed and quality.

[0025] The cutting result data include cutting effect BHC, modified area distance BCD and modified area length h (the relevant parameters of the modified area are as follows Figure 1 These data are used to evaluate the cutting quality. Figure 1 In the figure, De1~De3 are the target thicknesses when cutting with multiple target thickness requirements, D1~D3 are the safety distances corresponding to the target thicknesses, P is the laser power used during cutting, and d is the modified layer interval. The cutting effect BHC can be evaluated through electronic measurement photos of the wafer circuit surface and electronic measurement photos of the laser scattering distance, which are used to reflect the overall quality of the wafer after cutting, including defects such as cracks and edge collapse. Photos of the wafer circuit surface after cutting are obtained through an electron microscope or other equipment, which can be used to analyze the surface morphology and defects of the circuit surface. The laser scattering distance reflects the scattering of the laser beam on the surface of the material, which can indirectly evaluate the cutting depth and quality.

[0026] 2. Data processing The collected data is preprocessed, including data cleaning and normalization, to remove outliers and noise data, and convert data of different dimensions into a unified scale to facilitate subsequent analysis and calculation.

[0027] Statistical analysis methods (such as variance analysis, regression analysis, etc.) are used to identify key process parameters that are highly correlated with product quality. For example, laser power, focus position, and laser mode have a significant impact on cutting quality. At the same time, the features in the electronic measurement photos of the wafer circuit surface and the electronic measurement photos of the laser scattering distance, that is, the BHC, are quantitatively scored from 1 to 10 according to the cutting quality of the front side of the wafer. The scoring criteria include whether the cutting marks are separated, whether there are serpentine bifurcations, etc. Generally speaking, the cutting scores gradually increase from never separating the cutting marks, the presence of serpentine bifurcations, a small amount of serpentine bifurcations to straight lines. The cutting quality under each process parameter is evaluated through data scoring and used for subsequent machine learning model training, such as Figure 2 shown.

[0028] 3. Function fitting Function fitting is performed on the relationship between the wafer cutting result data and the cutting process parameter data.

[0029] Different laser heads and their working modes generate different shapes of modified areas, so the collected data can be used to fit the corresponding relationship, namely: Fitting function 1:

[0030] Among them, the out-of-focus L and laser power P These are the input process parameters of the laser cutting equipment. h is the length of the modified area. SC is the laser mode, which is a beam mode with fixed optical path characteristics and cannot be quantified. f That is, the function expression obtained by fitting, and its specific form depends on the specific application scenario, for example:

[0031] , , …, For fitting coefficients, the accuracy of the fitting polynomial function can meet engineering requirements, and the fitting method can adopt least squares and other algorithms.

[0032] During laser cutting, the light beam is incident from the gaseous medium to the backside polished surface of the wafer. Its projection angle can be calculated according to the law of refraction of light: ,in n 1 and n 2 are the refractive indices of the incident and refracting media, respectively, θ 1 and θ 2 are the angle of incidence and the angle of refraction respectively.

[0033] In stealth dicing, the laser processes the wafer to generate single or multiple modified areas. The cracks in these modified areas will extend perpendicular to the laser focus direction, eventually leading to wafer separation. In order to estimate the distance BCD of the modified area, the measured value can be used for fitting: Fitting function 2: BCD down = T - ( n 2+ k) L ， BCD up = BCD down + h Where k is the laser refraction correction coefficient, which is usually obtained through experimental calibration and parameter fitting. Since the wafer has losses such as absorption, reflection, and scattering of the laser, and the difference in refractive index will change the actual focal position, this will cause the geometric shape of the modified area (such as depth and width) to deviate from the theoretical calculation. The role of k is to correct the errors caused by these optical effects through experience. BCD down Indicates the lowest vertical height of the modified area, BCD up It represents the maximum height of the modified area in the vertical direction.

[0034] 4. Model training Since the amount of data that can be generated by each DOE experiment is limited, a suitable machine learning algorithm must be used. Support vector machine (SVM) is a supervised learning method that is particularly suitable for small sample, nonlinear and high-dimensional pattern recognition problems. Multiple groups of SVM models are trained separately using a small amount of historical data, with the cutting process parameters as the model input and the cutting result data (BHC score) as the output, so that the trained SVM model can predict the wafer cutting state under a given combination of process parameters.

[0035] 5. Process parameter optimization The cutting process parameters are optimized by training multiple sets of machine learning models and combining fitting functions.

[0036] 5.1. Determine the improvement area To ensure that the wafer will not be ground to the modified area after cutting, the modified area needs to avoid the target thickness position and leave a certain safety distance. That is: BCD down >De + 0.5D, BCD up <De - 0.5D。

[0037] 5.2. Determine the initial value of process parameters According to the wafer characteristic data required by the project (target thickness De and safety distance D), the K-means algorithm is used to find matching parameters in the existing cutting process parameters (obtained through DOE experiments) as the initial process parameters, and a complete match is not required.

[0038] Specifically, first, cluster the cutting result data obtained from the DOE experiment to obtain the cluster centers of each category; (1) Select m groups of data from the cutting result data as the initial cluster centers. The dimensions of each sample point include the BHC score, the distance from the modified area to the BCD, and the length h of the modified area; (2) Traverse each sample in the cutting result data, calculate its distances to the m initial cluster centers respectively, and assign it to the category corresponding to the cluster center with the minimum distance; (3) Calculate the centroid of all samples included in each category respectively and use it as the cluster center of this category; (4) Repeat the above two steps (2) and (3) until the termination condition is reached, such as the cluster center no longer changes, or the change error is less than the set threshold, or the number of iterations reaches a certain value.

[0039] Then, calculate the Euclidean distances between the wafer characteristic data corresponding to the cluster centers of each category and the wafer characteristic data required by the project respectively, and use the cutting process parameters corresponding to the cluster center with the minimum Euclidean distance as the initial process parameters. If the cluster center is not the sample data existing in the DOE experiment, select the sample with the minimum distance from this cluster center as the substitute value.

[0040] 5.3. Optimize the process parameters To meet the requirements of multiple chip thicknesses, use machine learning models and fitting functions to optimize the initial process parameters. That is, input the initial values of different process parameters and output the optimized and adjusted process parameters based on the initial values.

[0041] Specifically, first input the initial process parameters into multiple trained machine learning models to obtain the average BHC score under this process parameter. If the average BHC score meets the requirements, use this process parameter as the final process parameter. If the average BHC score does not meet the requirements, then within the range of the modified area interval (BCD down > De + 0.5D, BCD up < De - 0.5D), adjust the off-focus L value in the process parameters according to fitting function 1 and fitting function 2, and input the adjusted process parameters into the machine learning model to obtain the BHC score until it meets the requirements. The entire optimization process of the process parameters is as Figure 3 shown.

[0042] 6. Model update By continuously collecting the latest data in the production process, continuously update the machine learning model and the actual wafer cutting data set to adapt to the changes in process conditions and improve the robustness and adaptability of the model.

[0043] Embodiment 2 Based on the above method, this embodiment provides a wafer cutting parameter optimization system, which mainly includes the following modules: A data acquisition and storage module is used to acquire and store actual wafer cutting data, including wafer characteristic data, cutting process parameters and corresponding cutting result data; The data processing module is used to pre-process the collected data and quantify the scores of the corresponding wafer characteristic data and the cutting quality under the cutting process parameters according to the cutting result data; Function fitting module, used for performing function fitting on the relationship between laser defocus point, laser power and length of the generated modified area under different laser modes, and performing function fitting on the relationship between modified area distance BCD and laser defocus point; The model training module is used to train multiple machine learning models separately. The input of the machine learning model is the cutting process parameters, and the output of the machine learning model is the cutting quality score. For large-scale, high-dimensional data, a deep neural network model is preferred, while for small and medium-sized data sets, traditional machine learning algorithms such as random forests or support vector machines can be selected.

[0044] An initial parameter determination module is used to determine the initial parameters of the cutting process according to the wafer characteristic data required by the project; The parameter optimization module is used to optimize and adjust the initial parameters of the cutting process according to the machine learning model and fitting function while meeting the range of the modified area.

[0045] In some embodiments, the system also includes a model updating module for updating and optimizing the machine learning model using continuously collected actual wafer cutting data.

[0046] The above-mentioned system can execute the wafer cutting parameter optimization method described in Example 1, and has the corresponding functional modules and beneficial effects of the method. For technical details not described in detail in this embodiment, please refer to the wafer cutting parameter optimization method provided in Example 1 of the present invention.

[0047] In summary, compared with the traditional experience-based process parameter optimization scheme, the present invention can use machine learning algorithms to analyze the relationship between complex process parameters, shorten the optimization time, and improve the process optimization efficiency; by optimizing the process parameter combination, it can maintain a high level of product quality within a wider process parameter window range and reduce the defective product rate. Adaptive optimization is achieved by updating and iterating the new model, reducing human intervention and reducing manpower and time investment.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may also be combined, the steps may be implemented in any order, and there are many other changes in different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for optimizing wafer cutting parameters, characterized in that: Includes steps: Obtain actual wafer cutting data, including wafer characteristic data, cutting process parameters and corresponding cutting result data; wafer characteristic data includes target thickness, safety distance and wafer thickness, cutting process parameters include laser mode, laser defocus point and laser power, and cutting result data includes modified area length, modified area distance BCD and cutting effect BHC; Preprocess the collected data, and quantify the scores of the corresponding wafer characteristic data and the cutting quality under the cutting process parameters according to the cutting result data; The relationship between the laser defocus point, laser power and the length of the modified area generated under different laser modes is fitted with a function to obtain a modified area length fitting function; the relationship between the modified area distance BCD and the laser defocus point is fitted with a function to obtain a laser defocus point fitting function; Training a plurality of machine learning models respectively, wherein the input of the machine learning model is a cutting process parameter, and the output of the machine learning model is a cutting quality score; The modified area range and the initial parameters of the cutting process are determined based on the wafer characteristic data required for the project, and then the initial parameters of the cutting process are optimized and adjusted based on the machine learning model and fitting function.

2. The wafer cutting parameter optimization method according to claim 1, characterized in that: The optimization adjustment includes: inputting the initial parameters of the cutting process into multiple machine learning models to obtain the average cutting quality score under the parameters. If the cutting quality score meets the requirements, the parameters are used as the final cutting process parameters; otherwise, the length of the modified area is calculated according to the modified area length fitting function under the current parameters, and based on this, the defocus value in the parameter is adjusted according to the laser defocus fitting function while satisfying the modified area interval range, and the adjusted parameters are input into the machine learning model to obtain the average cutting quality score until it meets the requirements.

3. The wafer cutting parameter optimization method according to claim 1, characterized in that: The cutting effect BHC reflects the overall quality of the wafer after cutting through electronic measurement photos of the wafer circuit surface and electronic measurement photos of the laser scattering distance. The scoring criteria of the quantitative scoring include the state of the wafer cutting marks.

4. The wafer cutting parameter optimization method according to claim 1, characterized in that: The modified region length fitting function is a polynomial function obtained based on the least squares method, and the laser defocus fitting function is: BCD down = T - ( n 2 + k) L ， BCD up = BCD down + h, Among them, BCD down Indicates the lowest vertical height of the modified area, BCD up represents the maximum height of the modified area in the vertical direction, h represents the length of the modified area, T Indicates the wafer thickness, n 2 represents the refractive index of the laser incident on the wafer, k is the laser refraction correction coefficient, L Indicates the laser defocus point.

5. The wafer cutting parameter optimization method according to claim 4, characterized in that: The calculation formula of the modified area interval range is: BCD down > De + 0.5D, BCD up < De - 0.5D, where De represents the target thickness in the wafer characteristic data and D represents the safety distance.

6. The wafer cutting parameter optimization method according to claim 1, characterized in that: The initial parameters of the cutting process are determined based on the K-means clustering algorithm, including the steps of: First, cluster the obtained cutting result data to obtain the cluster centers of each category: S1. Select m groups of data from the cutting result data as the initial cluster centers, and the dimensions of each sample point include the cutting quality score, the length of the modified area h, and the distance BCD of the modified area; S2. Traverse each sample in the cutting result data, calculate its distance to the m initial cluster centers respectively, and divide it into the category corresponding to the cluster center with the smallest distance; S3. Calculate the centroid of all samples contained in each category and use it as the cluster center of the category; S4. Repeat S2 and S3 until the iteration termination condition is reached; Then, the Euclidean distances between the wafer characteristic data corresponding to the cluster centers of each category and the wafer characteristic data required for the project are calculated respectively, and the cutting process parameters corresponding to the cluster center with the smallest Euclidean distance are used as the initial process parameters.

7. The wafer cutting parameter optimization method according to claim 6, characterized in that: The iteration termination condition is that the cluster center of each category no longer changes or the change error is less than a set threshold or the number of iterations reaches a set value; if the cluster center is not an existing sample data, the sample data with the smallest distance to the cluster center is selected as a replacement value.

8. The wafer cutting parameter optimization method according to claim 1, characterized in that: The machine learning model adopts support vector machine.

9. A wafer cutting parameter optimization system based on the method according to any one of claims 1 to 8, characterized in that: include: A data acquisition and storage module is used to acquire and store actual wafer cutting data, including wafer characteristic data, cutting process parameters and corresponding cutting result data; The data processing module is used to pre-process the collected data and quantify the scores of the corresponding wafer characteristic data and the cutting quality under the cutting process parameters according to the cutting result data; Function fitting module, used for performing function fitting on the relationship between laser defocus point, laser power and length of the generated modified area under different laser modes, and performing function fitting on the relationship between modified area distance BCD and laser defocus point; A model training module, used to train multiple machine learning models respectively, wherein the input of the machine learning model is a cutting process parameter, and the output of the machine learning model is a cutting quality score; An initial parameter determination module is used to determine the initial parameters of the cutting process according to the wafer characteristic data required by the project; The parameter optimization module is used to optimize and adjust the initial parameters of the cutting process according to the machine learning model and fitting function while meeting the range of the modified area.

10. The wafer cutting parameter optimization system according to claim 9, characterized in that: It also includes a model update module, which is used to update and optimize the machine learning model using the actual wafer cutting data that is continuously collected.

Images