Wafer abnormity monitoring method and system, storage medium and electronic equipment
By acquiring images after wafer casting and counting, and adjusting equipment parameters using preset rules and model training, the problem of abnormal monitoring during wafer slab preparation, wafer casting, and counting was solved, enabling timely and accurate monitoring of wafer status and efficient equipment maintenance.
Patent Information
- Application Number
- CN202511116789.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-11
- Publication Date
- 2025-11-21
AI Technical Summary
Existing technologies lack timely, accurate, and efficient monitoring of abnormal conditions during wafer slicing, casting, and counting, resulting in late detection of wafer abnormalities, causing batch losses, and making it difficult to guarantee the efficiency and accuracy of manual re-inspection.
By acquiring the wafer images after casting and the counted images, the wafer status is determined using preset rules, generating qualified or unqualified marks. The equipment operating parameters are adjusted according to the wafer status parameters, and combined with model training to optimize maintenance and adjustment, timely monitoring of wafer anomalies is achieved.
It has improved the efficiency and accuracy of wafer status monitoring, enabled timely detection and handling of abnormal conditions, avoided batch losses, and improved the pertinence and efficiency of equipment maintenance.
Smart Images
Figure CN120998795A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor manufacturing technology, and in particular to a wafer anomaly monitoring method, system, storage medium, and electronic device. Background Technology
[0002] A wafer is a silicon wafer used to make silicon semiconductor circuits. Its raw material is silicon. High-purity polycrystalline silicon is dissolved and mixed with silicon crystal seed crystals. Then it is slowly pulled out to form a cylindrical single crystal silicon. After grinding, polishing and slicing, the silicon crystal rod is formed into a silicon wafer, which is a wafer. Domestic wafer production lines are mainly 8-inch and 12-inch.
[0003] During production, wafers are typically sorted into square wafers according to different grades, with the die electrode face upwards. Since flip-chip products require the die light-emitting face to be shipped upwards, a wafer lamination process is usually required, followed by label printing. However, wafer abnormalities can occur during sorting, lamination, and counting. These abnormalities are often only detected during manual re-inspection after counting and labeling, leading to delayed detection and batch losses. Furthermore, the efficiency and accuracy of manual re-inspection are difficult to guarantee. Summary of the Invention
[0004] Based on this, the purpose of the present invention is to provide a wafer anomaly monitoring method, system, storage medium, and electronic device, aiming to solve the problem in the prior art of lacking a timely, accurate, and efficient method for monitoring abnormal conditions that exist during wafer dicing, wafer casting, and counting.
[0005] A wafer anomaly monitoring method according to an embodiment of the present invention is characterized in that the method includes: The wafers are sorted according to the preset wafer grade to determine the corresponding wafers, and the wafers are then laminated using a lamination machine. The image of the first wafer after lamination, the wafer grade, and the machine node parameters are then obtained. The square pieces after the film is poured are counted using an automatic counter, and the image of the second square piece after counting and the parameters of the counting nodes are determined. The wafer status parameters are determined based on the wafer grade, the first wafer image, and the second wafer image, and the target wafer is determined based on the wafer status parameters and the counting node parameters, so as to control the shipment or locking of the target wafer; The wafer casting data for a preset time period is acquired to adjust the equipment operating parameters based on the wafer casting data. The wafer casting data includes the wafer status parameters, the machine node parameters, and the counting node parameters.
[0006] In addition, the wafer anomaly monitoring method according to the above embodiments of the present invention may also have the following additional technical features: Furthermore, the step of determining wafer state parameters based on the wafer grade, the first wafer image, and the second wafer image includes: The target proofreading image is determined based on the square plate grade; The first difference region and the first wafer state are determined based on the first slice image and the target calibration image; The second difference region and the second wafer state are determined based on the second slice image and the target calibration image; The system determines whether both the first wafer state and the second wafer state are in a normal wafer state by using preset rules. If so, a pass mark is generated, and the wafer state parameters include the first wafer state, the second wafer state, and the pass mark.
[0007] Furthermore, after the step of determining whether both the first wafer state and the second wafer state are normal wafer states according to preset rules, the following steps are included: If not, determine the image parameters at the first and / or second difference regions, and determine the missing depth and grain tilt angle based on the image parameters; The missing area is determined based on multiple missing depths, and a third wafer state is determined based on the missing area, the grain skew angle, the first wafer state, and the second wafer state, and a defect mark is generated. The wafer state parameters include the defect mark and the third wafer state.
[0008] Furthermore, both the first square image and the first square image comprise multiple images, and the step of determining the missing depth and grain tilt angle based on the image parameters includes: A grain distribution model for the corresponding region is constructed based on multiple image parameters, and the standard grain model is compared with the grain distribution model to determine the grain missing depth and grain tilt angle at different locations.
[0009] Furthermore, the first and second square images are hyperspectral images, and the steps of determining the missing depth and grain tilt angle based on the image parameters include: The missing depth and grain tilt angle are determined using a preset formula based on the image parameters. The preset formula is:
[0010] in, The grayscale value of a pixel. The incident light intensity The material absorption coefficient, For missing depth, For the first One wavelength, Optical distortion coefficient, This is the die-bonding tilt angle.
[0011] Furthermore, the step of acquiring wafer casting data over a preset time period and adjusting equipment operating parameters based on the wafer casting data includes: The changing trends of each abnormal state of the wafer are determined based on the wafer state parameters, the machine node parameters, and the counting node parameters. The degree of wafer casting and counting abnormality is determined based on the changing trends. The corresponding wafer abnormality status item is determined based on the wafer casting and the degree of count abnormality. The corresponding abnormal data is determined based on the wafer abnormality status item, the machine node parameters, and the count node parameters, so that maintenance personnel can perform maintenance and adjustment on the equipment based on the abnormal data and upload the maintenance and adjustment data. The abnormal data includes abnormal equipment and abnormal equipment process parameters.
[0012] Further, the step of acquiring wafer casting data over a preset time period, and adjusting equipment operating parameters based on the wafer casting data, includes: The model is trained and constructed based on the maintenance and adjustment data and its corresponding wafer state parameters, machine node parameters and counter node parameters, so that the trained model can determine the optimal maintenance and adjustment parameters based on the wafer state parameters, machine node parameters and counter node parameters.
[0013] Another object of the present invention is a wafer anomaly monitoring system, the system comprising: The first image determination module is used to sort the wafers according to the preset wafer grade to determine the corresponding wafers, and to perform wafer casting on the wafers through a wafer casting machine, and to obtain the first wafer image, wafer grade and machine node parameters after casting. The second image determination module is used to count the square pieces after the film is poured using an automatic counter, and to determine the image of the second square piece after counting and the parameters of the counting nodes; The target wafer determination module is used to determine wafer status parameters based on the wafer grade, the first wafer image, and the second wafer image, and to determine the target wafer based on the wafer status parameters and the counting node parameters, so as to control the shipment or locking of the target wafer; The data statistics module is used to acquire wafer casting data for a preset time period, so as to adjust the equipment operating parameters according to the wafer casting data. The wafer casting data includes the wafer status parameters, the machine node parameters, and the counting node parameters.
[0014] Another objective of this invention is to provide a storage medium on which a computer program is stored, which, when executed by a processor, implements the steps of the wafer anomaly monitoring method described above.
[0015] Another objective of this invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the wafer anomaly monitoring method described above.
[0016] This invention acquires corresponding wafer images after wafer casting and counting, analyzes these images to determine the wafer's condition, and then controls wafer locking or shipment based on the wafer's condition. This ensures that the wafers delivered to the customer meet requirements and allows for timely identification and adjustment of wafer defects. During image analysis, the images after casting and counting are first preliminarily judged and filtered to identify areas of difference. A second wafer condition confirmation is then performed on these areas, ensuring both efficiency and accuracy in wafer condition monitoring. Therefore, this invention solves the problem of the lack of a timely, accurate, and efficient method for monitoring abnormal conditions during wafer dicing, casting, and counting in the prior art. Attached Figure Description
[0017] Figure 1 This is a flowchart of the wafer anomaly monitoring method in the first embodiment of the present invention; Figure 2 This is a schematic diagram of the results of the wafer anomaly monitoring system in the second embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the electronic device in the third embodiment of the present invention; The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation
[0018] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
[0019] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.
[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0021] Example 1 Please see Figure 1 The figure shows a wafer anomaly monitoring method in the first embodiment of the present invention, which specifically includes steps S01-S04.
[0022] S01, the wafers are sorted according to the preset wafer grade to determine the corresponding square wafers, and the square wafers are laminated using a lamination machine, and the image of the first square wafer after lamination, the square wafer grade and the machine node parameters are obtained. Specifically, when sorting wafers into square pieces, rectangular products are typically divided to identify multiple circular target products. Due to the shape difference between the rectangular products and the target products, various arc-shaped target products may appear to ensure as many target products as possible are obtained. Therefore, during wafer sorting, it is necessary to determine the wafer grade of the corresponding sorted target products, i.e., the state of the target product—whether it is a normal circular product or an arc-shaped product of different sizes. This is to accurately monitor and judge the wafer state during subsequent wafer casting and counting. After casting, the image and grade of the first square piece after casting need to be acquired, along with the machine node parameters, including the casting time, the wafer number, and the machine parameters during casting. This is to facilitate subsequent identification of target wafers and adjustments to the machine equipment, parameters, or processes.
[0023] S02, the square pieces after the film is poured are counted by an automatic counter, and the image of the second square piece after counting and the parameters of the counting nodes are determined; Specifically, after wafer casting, the wafers need to be counted to ensure accurate data on the number of dies on each wafer. This is crucial for determining the wafer's true functionality, classifying the wafers, and avoiding customer complaints due to overcounting or losses due to undercounting. After counting, a second-division image is acquired to verify the accuracy of the count and to collect corresponding counting node parameters, including the counting time, wafer number, and equipment parameters during casting. This facilitates subsequent identification of the target wafer and adjustments to equipment, parameters, or processes.
[0024] S03, determine wafer status parameters based on the wafer grade, the first wafer image, and the second wafer image, and determine the target wafer based on the wafer status parameters and the counting node parameters, so as to control the shipment or locking of the target wafer; Specifically, the target calibration image is determined based on the wafer grade; a first difference region and a first wafer state are determined based on the first wafer image and the target calibration image; a second difference region and a second wafer state are determined based on the second wafer image and the target calibration image; and it is determined whether the first wafer state and the second wafer state are both normal wafer states according to preset rules. If so, a pass mark is generated, and the wafer state parameters include the first wafer state, the second wafer state, and the pass mark. In specific implementation, the wafer grade is used to determine the corresponding grade state of the wafer, that is, to determine whether the wafer is circular or an arc shape of different sizes, thereby selecting the corresponding target calibration image to achieve accurate judgment of the wafer state. In addition, when performing difference comparison, due to the large number of wafers and the large amount of data, and the low probability of wafer anomalies, a relatively strict screening threshold is usually adopted. A simple and direct image comparison method is used for preliminary screening. This method can quickly screen a large number of images, but the screening accuracy is limited. Furthermore, a relatively strict screening threshold is used to directly exclude some wafers with ambiguous states as unqualified states, thereby ensuring efficiency and avoiding misjudgments.
[0025] Furthermore, if not, determine the image parameters at the first and / or second difference regions, and determine the missing depth and grain skew angle based on the image parameters; determine the missing area based on multiple missing depths, and determine the third wafer state based on the missing area, the grain skew angle, the first wafer state, and the second wafer state, and generate a non-conforming or conforming mark, wherein the wafer state parameters include the non-conforming or conforming mark and the third wafer state. Specifically, when making a conformity judgment, it is necessary to ensure that the wafer state after casting and counting is in a normal state. That is, it is possible that after the initial comparison judgment, the wafer image after casting or the wafer image after counting does not meet the requirements, or both the wafer image after casting and the wafer image after counting do not meet the requirements. Then, based on the specific state, determine the difference region on the corresponding image for a more refined and accurate judgment, that is, accurately judge the missing status of wafer grains after casting, the missed count status of wafers after counting, and whether there is a die bonding abnormality in the wafer based on the missing area and the grain skew, etc. Furthermore, by precisely determining the wafer condition in specific areas, the accuracy of the assessment is ensured. This allows for the identification of the actual wafer condition, such as wafer block damage, wafer strip damage, wafer counting anomalies, wafer undercount anomalies, and wafer bonding anomalies. This facilitates subsequent, accurate adjustments to equipment or process technology based on identified wafer anomalies and trends.
[0026] Furthermore, a grain distribution model for the corresponding region is constructed based on multiple image parameters, and the standard grain model is compared with the grain distribution model to determine the grain missing depth and grain skew angle at different locations. In specific implementation, the condition of each grain can be determined sequentially by building a model of the grains in the different regions based on the images and comparing it with the normal grain model. That is, it can be determined whether there are missing grains in each region, the depth of the missing grains, and the skewness of each grain, thereby accurately judging the specific condition of the wafer.
[0027] Additionally, the missing depth and grain tilt angle are determined using a preset formula based on the image parameters. The preset formula is:
[0028] in, The grayscale value of a pixel. The incident light intensity The material absorption coefficient, For missing depth, For the first One wavelength, Optical distortion coefficient, This refers to the die tilt angle. In practical implementation, based on the light attenuation model and the angular distortion, the above formula can be constructed to link wafer defects and die tilt with the gray values (light intensity) of the corresponding areas in the image. This quantifies the abstract state phenomenon with specific parameters, allowing for accurate determination of the wafer state. Furthermore, in practical implementation, since two unknowns need to be determined, and semiconductor materials exhibit drastically different absorption and reflection characteristics for different wavelengths of light, it is necessary to solve an overdetermined system of equations using the above formula at multiple different wavelengths, followed by least-squares optimization to obtain an accurate and realistic wafer state. This method avoids the steps of wafer modeling and sequential comparison of individual dies; the wafer state can be determined based on the overall state of the image, resulting in higher efficiency and less data processing.
[0029] Furthermore, in practical implementation, the wafer state can be initially determined based on the formula, and then this initially determined wafer state can provide a foundation and assistance for subsequent wafer modeling, improving the efficiency and accuracy of wafer modeling. Thus, by combining these two methods, both accuracy and efficiency in wafer state determination can be achieved.
[0030] S04, acquire wafer casting data for a preset time period, and adjust equipment operating parameters according to the wafer casting data. The wafer casting data includes the wafer status parameters, the machine node parameters, and the counting node parameters.
[0031] Specifically, the changing trends of various abnormal wafer states are determined based on the wafer state parameters, machine node parameters, and counting node parameters. The degree of wafer casting and counting anomalies is then determined based on these trends. Corresponding wafer abnormal state items are determined based on the wafer casting and counting anomaly degrees. Corresponding abnormal data is then determined based on the wafer abnormal state items, machine node parameters, and counting node parameters. This allows maintenance personnel to perform maintenance adjustments on the equipment based on the abnormal data and upload the maintenance adjustment data. The abnormal data includes abnormal equipment and abnormal equipment process parameters. In practical implementation, statistical analysis of wafer data over a certain period can help determine the causes of abnormal wafer states, facilitating targeted improvements and maintenance by maintenance personnel.
[0032] Furthermore, following step S04, a model is trained and constructed based on the maintenance and adjustment data and its corresponding wafer state parameters, machine node parameters, and counter node parameters. This allows the trained model to determine the optimal maintenance and adjustment parameters based on these parameters. In practice, statistical learning is performed on multiple abnormal state data and corresponding processing data to train a corresponding analysis model. This enables systematic analysis during subsequent wafer anomalies, determining corresponding optimization parameters for the equipment manager to expedite or automatically adjust equipment parameters and promptly resolve wafer anomaly issues.
[0033] In summary, the wafer anomaly monitoring method in the above embodiments of the present invention acquires corresponding wafer images after wafer casting and counting, analyzes the wafer state based on the wafer images, identifies the corresponding wafer, and then controls wafer locking or shipment based on the wafer state to ensure that the wafers delivered to customers meet their requirements and promptly identify and address wafer defects. During image analysis, a preliminary judgment and screening of the images after wafer casting and counting is performed to identify areas of difference. A secondary wafer state confirmation is then conducted on these areas of difference, thus ensuring both the efficiency and accuracy of wafer state monitoring. Therefore, the present invention solves the problem in the prior art of lacking a timely, accurate, and efficient method for monitoring abnormal conditions during wafer dicing, wafer casting, and counting.
[0034] Example 2 Please see Figure 2 The diagram shows a structural block diagram of a wafer anomaly monitoring system proposed in the second embodiment of the present invention. The wafer anomaly monitoring system 200 includes: a first image determination module 21, a second image determination module 22, a target wafer determination module 23, and a data statistics module 24, wherein: The first image determination module 21 is used to sort the wafers according to the preset wafer grade to determine the corresponding wafers, and to perform wafer casting on the wafers through a wafer casting machine, and to obtain the first wafer image, wafer grade and machine node parameters after wafer casting. The second image determination module 22 is used to count the square pieces after the film is poured using an automatic counter, and to determine the second square piece image and the counting node parameters after counting. The target wafer determination module 23 is used to determine wafer status parameters based on the wafer grade, the first wafer image, and the second wafer image, and to determine the target wafer based on the wafer status parameters and the counting node parameters, so as to control the shipment or locking of the target wafer; The data statistics module 24 is used to acquire wafer casting data for a preset time period, so as to adjust the equipment operating parameters according to the wafer casting data. The wafer casting data includes the wafer status parameters, the machine node parameters, and the counting node parameters.
[0035] The functions or operation steps implemented by the above modules are largely the same as those in the above method embodiments, and will not be repeated here.
[0036] Example 3 In another aspect, the present invention also proposes an electronic device, please refer to [link to relevant documentation]. Figure 3 The diagram shown is a schematic diagram of an electronic device in the third embodiment of the present invention, including a memory 20, a processor 10, and a computer program 30 stored in the memory and executable on the processor. When the processor 10 executes the computer program 30, it implements the wafer anomaly monitoring method as described above.
[0037] In some embodiments, the processor 10 may be a central processing unit (CPU), controller, microcontroller, microprocessor or other data processing chip, used to run program code stored in memory 20 or process data, such as executing access restriction programs.
[0038] The memory 20 includes at least one type of readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 20 can be an internal storage unit of an electronic device, such as the hard disk of the electronic device. In other embodiments, the memory 20 can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. Furthermore, the memory 20 can include both internal and external storage units of the electronic device. The memory 20 can be used not only to store application software and various types of data of the electronic device, but also to temporarily store data that has been output or will be output.
[0039] It should be pointed out that, Figure 3 The structure shown does not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than shown, or combine certain components, or have different component arrangements.
[0040] This invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the wafer anomaly monitoring method described above.
[0041] Those skilled in the art will understand that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.
[0042] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0043] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0044] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0045] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.
Claims
1. A wafer anomaly monitoring method, characterized in that, The method includes: The wafers are sorted according to the preset wafer grade to determine the corresponding wafers, and the wafers are then laminated using a lamination machine. The image of the first wafer after lamination, the wafer grade, and the machine node parameters are then obtained. The square pieces after the film is poured are counted using an automatic counter, and the image of the second square piece after counting and the parameters of the counting nodes are determined. The wafer status parameters are determined based on the wafer grade, the first wafer image, and the second wafer image, and the target wafer is determined based on the wafer status parameters and the counting node parameters, so as to control the shipment or locking of the target wafer; The wafer casting data for a preset time period is acquired to adjust the equipment operating parameters based on the wafer casting data. The wafer casting data includes the wafer status parameters, the machine node parameters, and the counting node parameters.
2. The wafer anomaly monitoring method according to claim 1, characterized in that, The steps for determining wafer state parameters based on the wafer grade, the first wafer image, and the second wafer image include: The target proofreading image is determined based on the square plate grade; The first difference region and the first wafer state are determined based on the first slice image and the target calibration image; The second difference region and the second wafer state are determined based on the second slice image and the target calibration image; The system determines whether both the first wafer state and the second wafer state are in a normal wafer state by using preset rules. If so, a pass mark is generated, and the wafer state parameters include the first wafer state, the second wafer state, and the pass mark.
3. The wafer anomaly monitoring method according to claim 2, characterized in that, After the step of determining whether both the first wafer state and the second wafer state are in a normal wafer state according to preset rules, the following is included: If not, determine the image parameters at the first and / or second difference regions, and determine the missing depth and grain tilt angle based on the image parameters; The missing area is determined based on multiple missing depths, and a third wafer state is determined based on the missing area, the grain skew angle, the first wafer state, and the second wafer state, and a defect mark is generated. The wafer state parameters include the defect mark and the third wafer state.
4. The wafer anomaly monitoring method according to claim 3, characterized in that, Both the first square image and the first square image include multiple images. The steps of determining the missing depth and grain tilt angle based on the image parameters include: A grain distribution model for the corresponding region is constructed based on multiple image parameters, and the standard grain model is compared with the grain distribution model to determine the grain missing depth and grain tilt angle at different locations.
5. The wafer anomaly monitoring method according to claim 3, characterized in that, The first and second square images are hyperspectral images. The steps for determining the missing depth and grain tilt angle based on the image parameters include: The missing depth and grain tilt angle are determined using a preset formula based on the image parameters. The preset formula is: in, The grayscale value of a pixel. The incident light intensity The material absorption coefficient, For missing depth, For the first One wavelength, Optical distortion coefficient, This is the die-bonding tilt angle.
6. The wafer anomaly monitoring method according to claim 1, characterized in that, The steps for obtaining wafer casting data over a preset time period and adjusting equipment operating parameters based on the wafer casting data include: The changing trends of each abnormal state of the wafer are determined based on the wafer state parameters, the machine node parameters, and the counting node parameters. The degree of wafer casting and counting abnormality is determined based on the changing trends. The corresponding wafer abnormality status item is determined based on the wafer casting and the degree of count abnormality. The corresponding abnormal data is determined based on the wafer abnormality status item, the machine node parameters, and the count node parameters, so that maintenance personnel can perform maintenance and adjustment on the equipment based on the abnormal data and upload the maintenance and adjustment data. The abnormal data includes abnormal equipment and abnormal equipment process parameters.
7. The wafer anomaly monitoring method according to claim 6, characterized in that, After the step of acquiring wafer casting data for a preset time period and adjusting equipment operating parameters based on the wafer casting data, the following steps are included: The model is trained and constructed based on the maintenance and adjustment data and its corresponding wafer state parameters, machine node parameters and counter node parameters, so that the trained model can determine the optimal maintenance and adjustment parameters based on the wafer state parameters, machine node parameters and counter node parameters.
8. A wafer anomaly monitoring system, characterized in that, For implementing the wafer anomaly monitoring method as described in any one of claims 1 to 7, the system comprises: The first image determination module is used to sort the wafers according to the preset wafer grade to determine the corresponding wafers, and to perform wafer casting on the wafers through a wafer casting machine, and to obtain the first wafer image, wafer grade and machine node parameters after casting. The second image determination module is used to count the square pieces after the film is poured using an automatic counter, and to determine the image of the second square piece after counting and the parameters of the counting nodes; The target wafer determination module is used to determine wafer status parameters based on the wafer grade, the first wafer image, and the second wafer image, and to determine the target wafer based on the wafer status parameters and the counting node parameters, so as to control the shipment or locking of the target wafer; The data statistics module is used to acquire wafer casting data for a preset time period, so as to adjust the equipment operating parameters according to the wafer casting data. The wafer casting data includes the wafer status parameters, the machine node parameters, and the counting node parameters.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the wafer anomaly monitoring method as described in any one of claims 1 to 7.
10. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the wafer anomaly monitoring method as described in any one of claims 1-7.