Process data anomaly analysis method, device, equipment, medium and program product

Abstract

The invention belongs to the technical field of silicon single crystal rod preparation, and particularly relates to a process data anomaly analysis method, device and equipment, a medium and a program product, process time sequence data generated in the pulling process of a silicon single crystal rod is obtained, and the process time sequence data comprises a plurality of process steps and a plurality of process time sequence parameters corresponding to the process steps; according to the plurality of process steps, the plurality of process time sequence parameters corresponding to each process step and a preset feature library, determining a process time sequence data score of each process step, and according to the process time sequence data score of each process step and the plurality of groups of process abnormality score intervals of each process step, determining an abnormality type of pulling of the single crystal silicon rod, the anomaly types corresponding to each group of process anomaly scoring intervals are different; according to the method, by dynamically associating process time sequence data with production conditions (crucible feeding and feeding information) and combining staged scoring and abnormal interval matching, the accuracy of abnormal type identification in the single crystal silicon rod drawing process is improved.

Description

Technical Field

[0001] This application relates to the field of single-crystal silicon rod preparation technology, and in particular to a method, apparatus, equipment, medium, and program product for analyzing process data anomalies. Background Technology

[0002] Monocrystalline silicon rods are a key basic material in the photovoltaic cell industry, and their crystal quality directly determines the photoelectric conversion efficiency and long-term operational stability of photovoltaic modules. During the monocrystalline silicon pulling process, the pulling furnace needs to precisely control hundreds of process parameters, including pulling speed, crystal rotation speed, crucible rotation speed, and heater power, and undergoes multiple steps such as lead-in, necking, equal-diameter shaping, and finishing. Each stage is extremely sensitive to dynamic changes in parameters, requiring high-precision real-time monitoring and control capabilities throughout the entire process to ensure the uniformity and integrity of the crystal structure.

[0003] Currently, the industry mainly relies on human experience to analyze the massive amounts of high-dimensional time-series data generated during crystal pulling, and uses feature extraction methods based on fixed rules to identify potential anomalies. These methods typically set thresholds or logical rules for specific operating conditions or historical data to identify typical anomalies such as crystal defects and thermal imbalances, thus supporting process control and quality assurance to a certain extent.

[0004] However, with the expansion of the photovoltaic industry and the increase in process complexity, factors such as batch differences in raw materials and fluctuations in equipment operating environment have led to complex characteristics in process data, including strong nonlinearity, high dimension, and time-series coupling. Traditional manual analysis and fixed rule methods are difficult to adapt to dynamically changing process conditions, resulting in low accuracy in identifying abnormal types during the pulling process of monocrystalline silicon rods. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, medium, and program product for analyzing process data anomalies, in order to solve the problem of low accuracy in identifying anomaly types during the pulling process of monocrystalline silicon rods.

[0006] In a first aspect, embodiments of this application provide a method for analyzing process data anomalies, including:

[0007] Acquire process timing data generated during the single-crystal silicon rod pulling process, the process timing data including multiple process steps and multiple process timing parameters corresponding to each process step;

[0008] Based on the multiple process steps, the multiple process timing parameters corresponding to each process step, and the preset feature library, the process timing data score of each process step is determined. The preset feature library includes the correlation between the process steps and the process timing parameters determined based on the crucible feeding information and material feeding information of the single crystal silicon rod pulling process.

[0009] Based on the process timing data score for each process step and the multiple sets of process anomaly score intervals for each process step, the anomaly type of the single crystal silicon rod pulling is determined, wherein the anomaly type corresponding to each set of process anomaly score intervals is different.

[0010] In one optional implementation, determining the process timing data score for each process step based on the plurality of process steps, the plurality of process timing parameters corresponding to each process step, and a preset feature library includes:

[0011] Based on the multiple process steps and the preset feature library, multiple preset process timing parameters for each process step are determined;

[0012] Based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step, the process timing data score for each process step is determined.

[0013] In one optional implementation, determining the process timing data score for each process step based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step includes:

[0014] Based on the multiple process timing parameters and multiple preset process timing parameters corresponding to each process step, determine the process timing parameter score for each process timing parameter;

[0015] The process timing data score for each process step is determined based on the process timing parameter score and the weight of each process timing parameter.

[0016] In one optional implementation, determining the anomaly type in the single-crystal silicon rod pulling process based on the process timing data score for each process step and the multiple sets of process anomaly score intervals for each process step includes:

[0017] If the process timing data score of any step falls within the target process anomaly score range of that step, then the anomaly type of the single crystal silicon rod pulling is determined to include the anomaly type corresponding to the target process anomaly score range of that step, where the target process anomaly score range is any one of the multiple sets of process anomaly score ranges.

[0018] In one optional embodiment, the method further includes: if the process timing data scores of all steps are not within the target process anomaly score range for all steps, then the process of pulling the monocrystalline silicon rod is determined to be normal. In one optional embodiment, after determining the anomaly type of the monocrystalline silicon rod pulling based on the process timing data scores of each step and multiple sets of process anomaly ranges, the method further includes:

[0019] Multiple process timing parameters of any one of the process steps are identified as process anomaly data;

[0020] The anomaly type and the process anomaly data are sent to the client to provide an anomaly notification.

[0021] Secondly, embodiments of this application provide a process data anomaly analysis device, comprising:

[0022] The acquisition module is used to acquire process timing data generated during the pulling process of single crystal silicon rods. The process timing data includes multiple process steps and multiple process timing parameters corresponding to each process step.

[0023] The determination module is used to determine the process timing data score of each process step based on the multiple process steps, the multiple process timing parameters corresponding to each process step, and the preset feature library. The preset feature library includes the correlation between the process steps and the process timing parameters determined based on the crucible feeding information and material feeding information of the single crystal silicon rod pulling.

[0024] The determining module is further configured to determine the abnormality type of the single crystal silicon rod pulling based on the process timing data score of each process step and the multiple sets of process abnormality score intervals of each process step, wherein the abnormality type corresponding to each set of process abnormality score intervals is different.

[0025] In an optional implementation, the determining module is further configured to determine multiple preset process timing parameters for each process step based on the multiple process steps and the preset feature library;

[0026] The determining module is further configured to determine the process timing data score for each process step based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step.

[0027] In an optional implementation, the determining module is further configured to determine the process timing parameter score of each process timing parameter based on multiple process timing parameters corresponding to each process step and multiple preset process timing parameters.

[0028] The determining module is further configured to determine the process timing data score of each process step based on the process timing parameter score of each process timing parameter and the weight of each process timing parameter.

[0029] In an optional implementation, the determining module is further configured to determine that if the process timing data score of any step is within the target process anomaly score range of any step, the anomaly type of the single crystal silicon rod pulling includes the anomaly type corresponding to the target process anomaly score range of any step, wherein the target process anomaly score range is any one of the multiple sets of process anomaly score ranges.

[0030] In an optional embodiment, the determining module is further configured to determine that the single-crystal silicon rod pulling process is normal if the process timing data scores of all processes are not within the target process anomaly score range of all processes. In an optional embodiment, the process data anomaly analysis device further includes: a sending module;

[0031] The determining module is also used to determine multiple process timing parameters of any one of the process steps as process abnormal data;

[0032] The sending module is used to send the exception type and the process exception data to the client to provide an exception notification.

[0033] Thirdly, embodiments of this application provide a process data anomaly analysis device, including: a memory and a processor;

[0034] The memory stores computer-executed instructions;

[0035] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0036] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0037] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0038] The process data anomaly analysis method provided in this application accurately acquires multiple steps and corresponding process timing parameters in the single-crystal silicon rod pulling process. It integrates a preset feature library of crucible charging information and material feeding information to establish a dynamic correlation between the steps and process timing parameters. Then, it determines the anomaly type by scoring each step and matching multiple sets of differentiated anomaly scoring intervals. This method not only achieves a deep binding between process timing data and actual production conditions, but also effectively avoids the limitations of a single evaluation standard through phased refined evaluation and targeted anomaly interval matching, thereby improving the accuracy of anomaly type identification in the single-crystal silicon rod pulling process. Attached Figure Description

[0039] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0040] Figure 1The flowchart of the process data anomaly analysis method provided in this application Figure 1 ;

[0041] Figure 2 The flowchart of the process data anomaly analysis method provided in this application Figure 2 ;

[0042] Figure 3 A schematic diagram of the process data anomaly analysis device provided in this application;

[0043] Figure 4 A schematic diagram of the process data anomaly analysis equipment provided in this application.

[0044] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0045] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0046] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0047] As a core material in the photovoltaic cell industry, the crystal quality of monocrystalline silicon rods directly determines the photoelectric conversion efficiency and long-term operational stability of photovoltaic modules. During the pulling process, the crystal pulling furnace requires high-precision coordinated control of hundreds of process parameters, including pulling speed, crystal rotation speed, crucible rotation speed, and heater power, sequentially undergoing several key steps such as lead-in, necking, equal-diameter shaping, and finishing. Each stage is extremely sensitive to parameter changes, requiring real-time, precise monitoring and dynamic adjustment capabilities throughout the entire process to ensure a highly uniform and intact crystal structure.

[0048] Currently, the industry generally relies on the experience of operators, combined with feature extraction methods based on fixed rules, to analyze the massive amounts of high-dimensional time-series data generated during crystal pulling to identify typical anomalies such as crystal defects or thermal imbalances. These methods typically set thresholds and logical judgment rules based on specific operating conditions or historical data, which to some extent supports process control and product quality assurance.

[0049] However, with the continuous expansion of the photovoltaic industry and the increasing complexity of its processes, factors such as batch differences in raw materials and fluctuations in equipment environment have led to complex characteristics in process data, including strong nonlinearity, high dimensionality, and temporal coupling. Traditional methods relying on human experience and static rules are difficult to effectively adapt to these dynamically changing process conditions, resulting in low accuracy in identifying anomalies during the pulling of monocrystalline silicon rods, which restricts further improvements in product quality and production efficiency.

[0050] To address the aforementioned issues, the process data anomaly analysis method provided in this application first acquires process timing data containing multiple process steps and corresponding sets of process timing parameters. Then, it calculates the process timing data score for each process step based on a preset feature library (constructing the correlation between process steps and process timing parameters based on crucible charging information and material feeding information). Finally, it matches this score with multiple sets of process anomaly score intervals corresponding to different anomaly types for each process step. Based on the logic of dynamically associating production conditions with process timing data, process step scores, and precise anomaly interval matching, it achieves accurate identification of anomaly types in monocrystalline silicon rod pulling.

[0051] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0052] Figure 1 The flowchart of the process data anomaly analysis method provided in this application Figure 1 The executing entity in this embodiment is, for example, a process data anomaly analysis system. Figure 1 As shown in this embodiment, the process data anomaly analysis method includes:

[0053] S101: Obtain the process timing data generated during the pulling process of monocrystalline silicon rods. The process timing data includes multiple process steps and multiple process timing parameters corresponding to each process step.

[0054] The crystal growth process includes the following steps: nucleation, necking, diameter equalization, and finishing. Nucleation is the initial stage for forming the crystal rudiment; necking is used to eliminate defects in the initial crystal; diameter equalization is the core step to ensure stable crystal diameter growth; and finishing is the step that concludes crystal growth.

[0055] Specifically, the process timing parameters include pulling speed (the speed of crystal growth), crystal rotation speed (the rotation speed of the crystal itself), crucible rotation speed (the rotation speed of the crucible holding the silicon material), heater power (the power of the heating device to maintain the temperature inside the furnace) and argon flow rate (the flow rate of the protective gas inside the furnace). The process timing parameters change in real time as the process steps progress, and their dynamic data in each process step are recorded.

[0056] S102: Based on multiple process steps, multiple process timing parameters corresponding to each process step, and a preset feature library, determine the process timing data score for each process step. The preset feature library includes the correlation between process steps and process timing parameters determined based on crucible feeding information and material feeding information during the pulling of monocrystalline silicon rods.

[0057] The preset feature library is designed based on the specific production conditions of single-crystal silicon rod pulling, namely crucible feeding information (such as crucible size, material, and number of uses) and material feeding information (such as silicon purity and feeding amount). This information directly affects the correlation between each step and the process timing parameters. For example, in the "equal diameter" step, the reasonable range of heater power may differ for larger crucibles compared to smaller crucibles; and the standard range of pulling speed will be adjusted accordingly when the feeding amount increases.

[0058] Specifically, under specific crucible and feeding conditions, what are the reasonable ranges for the process timing parameters (such as pulling speed and crystal rotation speed) for each step (e.g., lead-in, necking)? What rules should the matching relationships between parameters (e.g., the coordinated changes in pulling speed and heater power) follow? In actual evaluation, the real-time parameter data for each step is compared with the corresponding standards in the feature library. By calculating the degree of deviation (e.g., the number of times parameters exceed the reasonable range, the magnitude of deviation, etc.), the final result is converted into a process timing data score for each step.

[0059] For example, for the "necking" step, if the feature library specifies that the pulling speed should be stable at 0.6-0.8 mm / min and the ratio of crystal rotation speed to crucible rotation speed should be maintained between 1.5-2.0 based on the currently used 8-inch crucible and 300 kg feed rate, then during scoring, the fluctuation of the actual pulling speed and the degree to which the rotation speed ratio conforms will be checked. If it meets the standard most of the time, it may be awarded 90 points; if it exceeds the range multiple times, the score will be reduced, such as to 60 points.

[0060] S103: Based on the process timing data score for each step and the multiple sets of process anomaly score intervals for each step, determine the anomaly type in the pulling of monocrystalline silicon rods. The anomaly type corresponds to each set of process anomaly score intervals.

[0061] For different process steps such as crystal release, necking, constant diameter formation, and finishing, multiple sets of corresponding scoring intervals were set, and each interval uniquely corresponds to one type of anomaly. This is because the process objectives and key parameters of different process steps are different, and the manifestations of anomalies also differ, requiring targeted interval division. For example, anomalies in the "constant diameter" process may be related to diameter stability, while anomalies in the "crystal release" process may be related to the integrity of the initial crystal formation; naturally, their scoring intervals and corresponding anomaly types are different.

[0062] Specifically, first check which abnormal score range each step's score falls into. Then, based on the correspondence between the ranges and the abnormality types, determine the abnormality present in that step. If a step's score does not fall into any abnormal range, it indicates that the process for that step is normal. If it falls into a range, it is determined that a corresponding abnormality exists. If it falls into multiple ranges simultaneously (this needs to be determined in conjunction with the actual logic), there may be multiple related abnormalities.

[0063] For example, for the "equal diameter" step, the preset abnormal score ranges might include: 80-90 points corresponding to "minor diameter fluctuations", 60-80 points corresponding to "insufficient diameter stability", 40-60 points corresponding to "diameter deviation risk", and below 40 points corresponding to "diameter out of control". If the score for this step is 70 points, falling into the 60-80 point range, then the abnormality type "insufficient diameter stability" can be determined for this step; if the score for the "finishing" step is 30 points, falling into its corresponding "finishing speed too fast" range, then this type of abnormality is determined to exist. Finally, the abnormality types of each step are summarized to obtain the abnormal situation of the entire drawing process.

[0064] The process data anomaly analysis method provided in this embodiment acquires process timing data generated during the pulling of monocrystalline silicon rods. This data includes multiple steps and corresponding process timing parameters. Based on these steps, the corresponding process timing parameters, and a preset feature library, a score for each step's process timing data is determined. The preset feature library includes the correlation between steps and process timing parameters determined based on crucible charging and material feeding information during monocrystalline silicon rod pulling. Based on the process timing data score for each step and multiple sets of process anomaly score intervals for each step, the anomaly type in monocrystalline silicon rod pulling is determined. Each set of process anomaly score intervals corresponds to a different anomaly type. This method improves the accuracy of anomaly type identification during monocrystalline silicon rod pulling by dynamically associating process timing data with production conditions (crucible charging and material feeding information) and combining staged scoring with anomaly interval matching.

[0065] Figure 2 The flowchart of the process data anomaly analysis method provided in this application Figure 2 .like Figure 2 As shown, in this embodiment... Figure 1Based on the examples, the method for analyzing process data anomalies is described in detail, including:

[0066] S201: Obtain the process timing data generated during the pulling process of monocrystalline silicon rods. The process timing data includes multiple process steps and multiple process timing parameters corresponding to each process step.

[0067] The core steps in pulling single-crystal silicon rods include lead-in, necking, equal diameter, and tailing. Each step has different process objectives (e.g., necking requires eliminating initial crystal defects) and depends on real-time adjustments of parameters such as pulling speed, crystal rotation speed, crucible rotation speed, heater power, and argon flow rate. It is necessary to accurately record the time interval of each step and the dynamic curves of the corresponding parameters to ensure that the data can completely reproduce the process execution.

[0068] For example, in the "equal diameter" step (lasting 120 minutes), data is collected every 10 seconds to record the real-time changes in pulling speed (fluctuating between 1.0-1.2 mm / min), crystal rotation speed (15 r / min), crucible rotation speed (10 r / min), heater power (60 kW ± 2 kW), and argon flow rate (50 L / min), forming a time-series dataset for this step.

[0069] S202: Based on multiple process steps and a preset feature library, determine multiple preset process timing parameters for each process step.

[0070] The preset feature library combines crucible feeding information (such as 8-inch crucible, 3 uses) and material feeding information (such as 300kg of high-purity silicon material) to customize parameter standards for different processes.

[0071] Specifically, the "release" step is crucial for the formation of the crystal prototype, and its preset parameters need to be more stringent; while the parameter range of the "finishing" step needs to be adapted to the characteristics of the end of crystal growth. Preset parameters for each step that match the current production conditions are retrieved from the feature library.

[0072] For example, for the current 8-inch crucible and 300kg feed amount, the preset process timing parameters for the "necking" step can be retrieved from the feature library as follows: pulling speed 0.6-0.8mm / min, crystal rotation speed / crucible rotation speed ratio 1.5-2.0, and heater power 55-60kW.

[0073] S203: Determine the process timing data score for each process step based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step.

[0074] Specifically, by comparing the differences between the process timing parameters and the preset process timing parameters, the overall process execution quality of each step is quantitatively evaluated, forming an intuitive score that can be used for anomaly detection.

[0075] The process quality of each step is determined by multiple parameters (such as the drawing speed and heater power in the equal diameter step). It is necessary to first determine whether each parameter meets the preset range (e.g., whether the drawing speed is between 1.0-1.2 mm / min), and then calculate the total score for the step based on the compliance of all parameters. A higher score indicates that the process execution of that step is closer to the ideal state; conversely, a lower score indicates a deviation.

[0076] For example, for the "constant diameter" step, the preset parameters are a drawing speed of 1.0-1.2 mm / min and a heater power of 58-62 kW. In actual data, the drawing speed was within the range 90% of the time, and the heater power was within the range 100%. Considering the degree of compliance of both, the final score for this step is 95 points.

[0077] Optionally, based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step, a process timing data score for each process step is determined. Specific implementation methods include:

[0078] Based on the multiple process timing parameters and multiple preset process timing parameters corresponding to each process step, determine the process timing parameter score for each process timing parameter;

[0079] The process timing data score for each process timing parameter is determined based on the process timing parameter score and the weight of each process timing parameter.

[0080] Specifically, firstly, each parameter is scored individually (e.g., pulling speed is scored 90 points, argon flow rate is scored 85 points), based on the percentage of time the parameter is within the preset range or the deviation range; then, the parameters are weighted and summed according to their importance (e.g., pulling speed weight 0.5, argon flow rate weight 0.1), highlighting the impact of core parameters (e.g., pulling speed of equal diameter step) on the quality of the step.

[0081] For example, in the "equal diameter" step, the actual compliance rate of the pulling speed is 90% (score 90 points, weight 0.5), the compliance rate of the heater power is 100% (score 100 points, weight 0.4), and the compliance rate of the argon flow rate is 80% (score 80 points, weight 0.1). Therefore, the individual parameter scores are 90, 100, and 80 points respectively, and the total score for the step = 90 × 0.5 + 100 × 0.4 + 80 × 0.1 = 93 points.

[0082] S204: If the process timing data score of any step falls within the target process anomaly score range of any step, then the anomaly type of single crystal silicon rod pulling is determined to include the anomaly type corresponding to the target process anomaly score range of any step, and the target process anomaly score range is any one of multiple sets of process anomaly score ranges.

[0083] Specifically, the anomaly type is determined by matching the process timing data score with a preset anomaly range.

[0084] Each work step corresponds to multiple sets of abnormal score intervals, and each interval is uniquely associated with one type of abnormality (e.g., "insufficient diameter stability" in the equal diameter work step corresponds to 60-80 points). By determining the interval in which the score falls, the abnormality type is directly mapped.

[0085] For example, if the score for the "equal diameter" step is 70 points, falling into the preset range of "60-80 points → insufficient diameter stability", then the step is judged to have an "insufficient diameter stability" anomaly; if the score for the "finishing" step is 35 points, falling into the range of "≤40 points → incomplete finishing", then the anomaly type "incomplete finishing" is added.

[0086] S205: Determine multiple process timing parameters for any step as process anomaly data.

[0087] Specifically, when a certain process step is determined to be abnormal, all process timing parameters (pulling speed, rotation speed, etc.) of that process step are key clues for anomaly analysis. The fluctuation trend of process timing parameters, the time of exceeding the tolerance, and other details can help determine the root cause of the problem (e.g., abnormal heater power may be due to equipment failure).

[0088] For example, after the "equal diameter" step is judged to have "insufficient diameter stability", the parameters of this step, such as the pulling speed curve (exceeding 1.2 mm / min multiple times), heater power fluctuation (±5 kW), and crucible rotation speed record (occasionally dropping to 8 r / min), are marked as abnormal process data.

[0089] S206: Send the exception type and process exception data to the client to provide an exception notification.

[0090] Specifically, abnormal information is pushed to relevant personnel in real time to ensure that process problems are detected and dealt with in a timely manner, and to reduce quality losses caused by the continued occurrence of abnormalities.

[0091] The client prompt should include two parts of information: the anomaly type (such as "insufficient diameter stability") to clarify the nature of the problem, and the abnormal process data (such as parameter curves, out-of-tolerance time points) to provide clues for investigation and help relevant personnel quickly locate the direction of adjustment (such as calibrating heater power).

[0092] For example, the system pushes a prompt to the client: "Abnormal type: equal diameter step - insufficient diameter stability; abnormal data: pulling speed reaches 1.3mm / min multiple times between 10:05-10:15, heater power fluctuates by ±5kW". Engineers can use this information to check the heating system.

[0093] S207: If the process timing data scores for all processes are not within the target process anomaly score range for all processes, then the process of pulling monocrystalline silicon rods is determined to be normal.

[0094] Specifically, when all process step scores do not fall within the target process anomaly score range, the process execution is confirmed to meet the standard, providing a basis for judging the quality of monocrystalline silicon rods and ensuring the stability of the normal production process.

[0095] Only when the process timing data scores of all steps (from introduction to completion) are within the normal range (e.g., ≥90 points) can it be determined that the overall process is without abnormalities, thus avoiding the impact of hidden problems in a single step on the final quality.

[0096] For example, if the scores for lead-in (95 points), necking (92 points), equal diameter (94 points), and tailing (93 points) are all higher than the lowest value of the abnormal interval of each step (such as the lowest score of 60 points in the equal diameter abnormal interval), then the entire pulling process is judged to be normal and the quality of the single crystal silicon rod meets the standards.

[0097] The process data anomaly analysis method provided in this embodiment first collects process timing data during the single-crystal silicon rod pulling process. This data includes multiple steps such as lead-in, necking, constant diameter, and finishing, as well as multiple process timing parameters corresponding to each step, such as pulling speed, crystal rotation speed, crucible rotation speed, heater power, and argon flow rate. Then, based on these steps and a preset feature library (which contains the correlation between steps and process timing parameters determined by crucible charging and material feeding information), multiple preset process timing parameters corresponding to each step are determined. Next, the actual process timing parameters for each step are compared with the preset process timing parameters to obtain a process timing data score for each step. If... If the score of any step falls into a target process anomaly score range corresponding to that step (each range corresponds to a unique anomaly type), then the anomaly type in the monocrystalline silicon rod pulling process is determined to include the anomaly type corresponding to that range. Simultaneously, multiple process timing parameters for that step are identified as process anomaly data, and the anomaly type and process anomaly data are sent to the client for notification. If the scores of all steps do not fall into the corresponding target process anomaly score range, then the process is considered normal. This method improves the accuracy of anomaly type identification during monocrystalline silicon rod pulling by dynamically associating process timing data with production conditions (crucible charging and material feeding information) and combining staged scoring with anomaly range matching.

[0098] Figure 3 A schematic diagram of the process data anomaly analysis device provided in this application. Figure 3 As shown, the process data anomaly analysis device 300 provided in this embodiment includes:

[0099] The acquisition module 301 is used to acquire process timing data generated during the single crystal silicon rod pulling process. The process timing data includes multiple process steps and multiple process timing parameters corresponding to each process step.

[0100] The determination module 302 is used to determine the process timing data score of each process step based on multiple process steps, multiple process timing parameters corresponding to each process step, and a preset feature library. The preset feature library includes the correlation between process steps and process timing parameters determined based on crucible feeding information and material feeding information for pulling single crystal silicon rods.

[0101] The determination module 302 is also used to determine the abnormality type of monocrystalline silicon rod pulling based on the process timing data score of each process step and the multiple sets of process abnormality score intervals for each process step, wherein the abnormality type corresponding to each set of process abnormality score intervals is different.

[0102] In one optional implementation, the determining module 302 is further configured to determine multiple preset process timing parameters for each process step based on multiple process steps and a preset feature library.

[0103] The determination module 302 is also used to determine the process timing data score of each process step based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step.

[0104] In an optional implementation, the determining module 302 is further configured to determine the process timing parameter score of each process timing parameter based on multiple process timing parameters corresponding to each process step and multiple preset process timing parameters.

[0105] The determination module 302 is also used to determine the process timing data score for each process step based on the process timing parameter score of each process timing parameter and the weight of each process timing parameter.

[0106] In an optional implementation, the determining module 302 is further configured to determine that if the process timing data score of any step is within the target process anomaly score range of any step, the anomaly type of the single crystal silicon rod pulling includes the anomaly type corresponding to the target process anomaly score range of any step, wherein the target process anomaly score range is any one of multiple sets of process anomaly score ranges.

[0107] In an optional implementation, the determining module 302 is further configured to determine that the process of pulling monocrystalline silicon rods is normal if the process timing data scores of all steps are not within the target process anomaly score range of all steps.

[0108] In one optional implementation, the process data anomaly analysis device further includes: a sending module 303;

[0109] The determination module 302 is also used to determine multiple process timing parameters of any step as process abnormal data;

[0110] The sending module 303 is used to send the exception type and process exception data to the client for exception notification.

[0111] Figure 4 A schematic diagram of the process data anomaly analysis equipment provided in this application. Figure 4 As shown, this application provides a process data anomaly analysis device. The process data anomaly analysis device 400 includes: a receiver 401, a transmitter 402, a processor 403, and a memory 404.

[0112] Receiver 401 is used to receive instructions and data;

[0113] Transmitter 402 is used to send commands and data;

[0114] Memory 404 is used to store instructions executed by the computer;

[0115] The processor 403 is used to execute computer execution instructions stored in the memory 404 to implement the various steps of the process data anomaly analysis method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing embodiments of the process data anomaly analysis method.

[0116] Alternatively, the memory 404 can be either standalone or integrated with the processor 403.

[0117] When the memory 404 is set up independently, the electronic device also includes a bus for connecting the memory 404 and the processor 403.

[0118] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the process data anomaly analysis method performed by the aforementioned process data anomaly analysis device.

[0119] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0120] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0121] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0122] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0123] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0126] If a function 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 invention, or the part that contributes to the prior art, or a 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 of the various embodiments of this invention. 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.

[0127] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0128] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for analyzing process data anomalies, characterized in that, include: Acquire process timing data generated during the single-crystal silicon rod pulling process, the process timing data including multiple process steps and multiple process timing parameters corresponding to each process step; Based on the multiple process steps, the multiple process timing parameters corresponding to each process step, and the preset feature library, the process timing data score of each process step is determined. The preset feature library includes the correlation between the process steps and the process timing parameters determined based on the crucible feeding information and material feeding information of the single crystal silicon rod pulling process. Based on the process timing data score for each process step and the multiple sets of process anomaly score intervals for each process step, the anomaly type of the single crystal silicon rod pulling is determined, wherein the anomaly type corresponding to each set of process anomaly score intervals is different.

2. The method according to claim 1, characterized in that, The step of determining the process timing data score for each process step based on the multiple process steps, the multiple process timing parameters corresponding to each process step, and a preset feature library includes: Based on the multiple process steps and the preset feature library, multiple preset process timing parameters for each process step are determined; Based on multiple process timing parameters and multiple preset process timing parameters corresponding to each process step, the process timing data score for each process step is determined.

3. The method according to claim 1, characterized in that, The step of determining the process timing data score for each process step based on multiple process timing parameters and multiple preset process timing parameters includes: Based on the multiple process timing parameters and multiple preset process timing parameters corresponding to each process step, determine the process timing parameter score for each process timing parameter; The process timing data score for each process step is determined based on the process timing parameter score and the weight of each process timing parameter.

4. The method according to claim 1, characterized in that, The method of determining the anomaly type in the single-crystal silicon rod pulling process based on the process timing data score for each process step and the multiple sets of process anomaly score intervals for each process step includes: If the process timing data score of any step falls within the target process anomaly score range of that step, then the anomaly type of the single crystal silicon rod pulling is determined to include the anomaly type corresponding to the target process anomaly score range of that step, where the target process anomaly score range is any one of the multiple sets of process anomaly score ranges.

5. The method according to claim 4, characterized in that, Also includes: If the process timing data scores for all processes are not within the target process anomaly score range for all processes, then the process of pulling the monocrystalline silicon rod is determined to be normal.

6. The method according to claim 1, characterized in that, After determining the anomaly type in the single-crystal silicon rod pulling process based on the process timing data score of each step and multiple sets of process anomaly intervals, the method further includes: Multiple process timing parameters of any one of the process steps are identified as process anomaly data; The anomaly type and the process anomaly data are sent to the client to provide an anomaly notification.

7. A process data anomaly analysis device, characterized in that, include: The acquisition module is used to acquire process timing data generated during the pulling process of single crystal silicon rods. The process timing data includes multiple process steps and multiple process timing parameters corresponding to each process step. The determination module is used to determine the process timing data score of each process step based on the multiple process steps, the multiple process timing parameters corresponding to each process step, and the preset feature library. The preset feature library includes the correlation between the process steps and the process timing parameters determined based on the crucible feeding information and material feeding information of the single crystal silicon rod pulling. The determining module is further configured to determine the abnormality type of the single crystal silicon rod pulling based on the process timing data score of each process step and the multiple sets of process abnormality score intervals of each process step, wherein the abnormality type corresponding to each set of process abnormality score intervals is different.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.

Images