Production execution system early warning method, production execution system, device and medium

Abstract

The embodiment of the application discloses a kind of production execution system early warning method, production execution system, equipment and medium, it is related to photovoltaic cell technical field.The method comprises: collecting the efficiency raw data of each test line;According to the abnormal data index, the abnormal data in the efficiency raw data is rejected, and target data is obtained;According to the abnormal analysis data model and target data, abnormal process and abnormal proportion are obtained;Based on the preset warning rule, according to abnormal process and abnormal proportion, warning is carried out.In this way, the test line data collected by the production execution system is analyzed by the abnormal analysis data model to analyze the abnormal process and abnormal proportion, and automatic warning is carried out, which can reduce the problem of product yield efficiency and poor consistency caused by the fact that single-machine abnormality is not found in time in the case of long process flow and multiple single-process machines in photovoltaic manufacturing.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic cell technology, and in particular to an early warning method, production execution system, equipment, and medium for a production execution system. Background Technology

[0002] Semiconductor manufacturing requires corresponding processes to achieve different performance requirements. In existing photovoltaic (semiconductor) manufacturing workshops, the manufacturing process is lengthy, and each process step is tied to several manufacturing machines, programs, and process data specifications. To achieve better yield, efficiency, and consistency in finished products, existing semiconductor manufacturing workshops require a certain number of process engineers to classify and troubleshoot anomalies in the finished products. This troubleshooting requires manpower, which increases production costs, and the longer the troubleshooting takes, the more defective products are generated. Summary of the Invention

[0003] In view of this, the purpose of this invention is to overcome the shortcomings of the prior art and provide a method, system, equipment and medium for early warning of a Manufacturing Execution System (MES). This method connects the MES to all process equipment, imports a trained anomaly analysis data model into the MES, uses the anomaly analysis data model to locate abnormal processes and anomaly ratios, and analyzes abnormal equipment to achieve automatic alarm of the MES, thereby reducing losses caused by defective output.

[0004] This invention provides the following technical solution:

[0005] In a first aspect, the present invention proposes an early warning method for a production execution system, applied to a production execution system, the method comprising:

[0006] Collect raw efficiency data for each test line;

[0007] Abnormal data is removed from the original efficiency data based on abnormal data indicators to obtain the target data;

[0008] Based on the anomaly analysis data model and the target data, the abnormal process and the anomaly ratio are obtained;

[0009] Based on preset early warning rules, early warnings are issued according to the abnormal process and the abnormality ratio.

[0010] In one embodiment, the step of removing abnormal data from the original efficiency data based on abnormal data indicators to obtain target data includes:

[0011] Verify the original efficiency data for correctness based on the abnormal data indicators;

[0012] If the raw efficiency data is correct, then the raw efficiency data will be used as the target data.

[0013] If the raw efficiency data is incorrect, then the raw efficiency data will be considered as the abnormal data.

[0014] In one embodiment, the step of issuing an early warning based on preset early warning rules, according to the abnormal process and the abnormality ratio, includes:

[0015] The abnormal equipment is determined based on the abnormal process and the abnormal ratio.

[0016] Based on the preset early warning rules, early warnings are issued according to the abnormal process and the abnormal equipment.

[0017] In one embodiment, determining the abnormal equipment based on the abnormal process and the abnormality ratio includes:

[0018] Acquire shared camera location data and calculate the shared camera location percentage based on the shared camera location data;

[0019] The abnormal equipment is determined based on the shared machine position ratio, the abnormal process, and the abnormal ratio.

[0020] In one embodiment, the step of issuing an early warning based on the preset early warning rules, according to the abnormal process and the abnormal equipment, includes:

[0021] Based on the preset early warning rules, early warning information is determined according to the abnormal process and the abnormal equipment;

[0022] Issue a warning based on the aforementioned warning information.

[0023] In one embodiment, the anomaly analysis data model includes an EL defect classification model, and the method further includes:

[0024] Obtain historical EL test results for photovoltaic cells;

[0025] The EL defect classification model is obtained by training based on the historical EL detection results using machine learning.

[0026] In one embodiment, the anomaly analysis data model includes a single-cell electrical performance analysis model, and the method further includes:

[0027] Obtain historical electrical performance parameters of a single solar cell;

[0028] Based on the historical electrical performance parameters, data analysis results are obtained, including parameter correlation information and performance impact information.

[0029] Based on the data analysis results, an electrical performance analysis model for the single-cell battery is established.

[0030] Secondly, the present invention proposes a production execution system, the system comprising:

[0031] The data acquisition module is used to collect raw efficiency data for each test line.

[0032] The processing module is used to remove abnormal data from the raw efficiency data to obtain the target data;

[0033] The determination module is used to obtain the abnormal process and the abnormality ratio based on the anomaly analysis data model and the target data;

[0034] The early warning module is used to issue early warnings based on preset early warning rules, according to the abnormal process and the abnormal ratio.

[0035] Thirdly, the present invention proposes a computer device including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the production execution system early warning method as described in the first aspect.

[0036] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the production execution system early warning method as described in the first aspect.

[0037] This invention discloses a method, system, equipment, and medium for early warning in a production execution system (MES). The method involves collecting raw efficiency data from each test line; removing abnormal data from the raw efficiency data based on abnormal data indicators to obtain target data; determining abnormal processes and proportions based on an anomaly analysis data model and the target data; and issuing warnings based on preset warning rules and the abnormal processes and proportions. By using an anomaly analysis data model to analyze abnormal processes and proportions in the test line data collected by the MES and issuing automatic warnings, this method can reduce problems of poor product yield and consistency caused by the failure to detect anomalies in individual machines during photovoltaic manufacturing processes with long workflows and many machines per process. Attached Figure Description

[0038] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope of protection of the present invention. In the various drawings, similar components are numbered similarly.

[0039] Figure 1 A flowchart illustrating the advance warning method for the production execution system proposed in this embodiment is shown.

[0040] Figure 2 Another flowchart of the production execution system early warning method proposed in this embodiment is shown;

[0041] Figure 3 This illustration shows another flowchart of the production execution system early warning method proposed in this embodiment;

[0042] Figure 4 This illustration shows another flowchart of the production execution system early warning method proposed in this embodiment;

[0043] Figure 5 This illustration shows another flowchart of the production execution system early warning method proposed in this embodiment;

[0044] Figure 6 A schematic diagram of the early warning device for the production execution system proposed in this embodiment is shown.

[0045] Explanation of reference numerals in the attached diagram:

[0046] 600 - Production Execution System; 601 - Data Acquisition Module; 602 - Processing Module; 603 - Determination Module; 604 - Early Warning Module. Detailed Implementation

[0047] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0048] The components of the embodiments of the invention described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0049] In the following, the terms “comprising,” “having,” and their cognates, which may be used in various embodiments of the invention, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as excluding, firstly, the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations thereof.

[0050] Furthermore, the terms "first," "second," and "third" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0051] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of the invention pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be interpreted as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of the invention.

[0052] Example 1

[0053] This disclosure provides an early warning method for a Manufacturing Execution System (MES), which connects the MES to all process equipment, imports a trained anomaly analysis data model into the MES, uses the anomaly analysis data model to locate abnormal processes and anomaly ratios, thereby analyzing abnormal equipment and implementing automatic alarms in the MES to reduce losses caused by defective output.

[0054] Please see Figure 1 A method for early warning in a production execution system is provided, which is applied to the production execution system. The method includes steps S101 to S104, and each step is described in detail below.

[0055] Step S101: Collect raw efficiency data for each test line.

[0056] In this embodiment, raw efficiency data is collected in real time or periodically from each test line using various sensors, counters, or other data acquisition devices, and saved to a database. The MES system uses a clock to scan and record data from the databases of the test machines on each production line, providing backups.

[0057] Step S102: Remove abnormal data from the original efficiency data according to the abnormal data index to obtain the target data.

[0058] In this embodiment, a series of abnormal data indicators, such as data range and rate of change, are set to identify and remove abnormal data in the original efficiency data, so as to obtain more accurate and reliable target data.

[0059] Please see Figure 2 In one specific embodiment, step S102 includes steps S1021 to S1023, and each step is described in detail below.

[0060] Step S1021: Verify whether the original efficiency data is correct based on the abnormal data indicators.

[0061] In this embodiment, a series of abnormal data indicators need to be set before starting the verification. These indicators may include the reasonable range of the data, the rate of change of the data, and the stability of the data. These indicators are used to measure whether the data meets expectations, thereby determining the correctness of the data.

[0062] The raw data is validated using predefined outlier metrics. This typically involves examining the data one by one, or performing batch analysis using some algorithm or model.

[0063] Step S1022: If the original efficiency data is correct, then the original efficiency data is used as the target data.

[0064] Step S1023: If the raw efficiency data is incorrect, then the raw efficiency data is treated as the abnormal data.

[0065] In this embodiment, the accuracy of the original efficiency data is determined based on the verification results. If the data meets the requirements of all abnormal data indicators, the data is considered correct; if the data does not meet the requirements of any one or more abnormal data indicators, the data is considered incorrect.

[0066] Based on the accuracy of the data, classify it into target data or outlier data. If the data is accurate, retain it as target data for subsequent analysis and decision-making. If the data is incorrect, mark it as outlier data and mask it.

[0067] It should be noted that in actual production, the testing machine may produce erroneous data due to factors such as incorrect selection of the test formula, hardware failure of the testing machine, or incorrect external input products (e.g., the tested solar cells have missing corners or variations in area, resulting in significant differences in the test data). A large amount of erroneous data can lead to distortion in the model analysis and incorrect conclusions. Therefore, it is necessary to mask outlier data.

[0068] Step S103: Obtain the abnormal process and abnormal ratio based on the abnormal analysis data model and the target data.

[0069] In this embodiment, a pre-established anomaly analysis data model is used to conduct in-depth analysis of the target data to identify specific abnormal processes and calculate the proportion of these abnormal processes in the total production process.

[0070] It should be noted that abnormal processes refer to processes (operations) in the production process that do not conform to expected standards or specifications. These processes may manifest as low production efficiency, unstable product quality, or potential safety hazards. Through data analysis and monitoring, these abnormal processes can be identified, and their specific location and scope of impact in the production process can be determined.

[0071] The anomaly ratio refers to the proportion of defective processes in the entire production process. This ratio reflects the severity and prevalence of the problem. By calculating the anomaly ratio, the impact of the problem on overall production can be assessed, and corresponding early warning strategies can be developed accordingly.

[0072] Please see Figure 3 In one specific embodiment, the anomaly analysis data model includes the EL defect classification model, and the method further includes steps S301 to S302, which are described in detail below.

[0073] Step S301: Obtain historical EL test results of photovoltaic cells.

[0074] In this embodiment, historical EL inspection results of photovoltaic cells are obtained. EL inspection is a non-destructive testing technique that uses voltage to make the cell emit light, thereby detecting internal defects such as cracks, broken grids, and black cores. Historical EL inspection results typically contain a large amount of image data, with each image corresponding to the EL image of a single cell, along with possible annotations indicating the presence of defects in the image.

[0075] Step S302: Based on machine learning, the model is trained according to the historical EL detection results to obtain the EL defect classification model.

[0076] In this embodiment, machine learning techniques are used to train the model using historical EL detection results. Image recognition or classification algorithms, such as convolutional neural networks (CNNs), can be used during model training. The training process includes data preprocessing (such as image enhancement and normalization), model architecture design, parameter tuning, and training iterations. The final result is an EL defect classification model that can accurately identify defective features in EL images.

[0077] It should be noted that the trained EL defect classification model can be used to detect newly produced solar cells in real time, quickly identify defective solar cells, and thus take corresponding measures.

[0078] Please see Figure 4 In one specific embodiment, the anomaly analysis data model includes a single-cell electrical performance analysis model, and the method further includes steps S401 to S403, which are described in detail below.

[0079] Step S401: Obtain the historical electrical performance parameters of a single solar cell.

[0080] In this embodiment, historical electrical performance parameters of a single solar cell are obtained. These parameters include open-circuit voltage (Voc), short-circuit current (Isc), fill factor (FF), and conversion efficiency (η). These parameters are typically measured under laboratory conditions using specialized testing equipment and reflect the electrical performance and photoelectric conversion capability of the solar cell.

[0081] Step S402: Analyze the historical electrical performance parameters to obtain data analysis results, which include parameter correlation information and performance impact information.

[0082] In this embodiment, the acquired historical electrical performance parameters are analyzed to obtain data analysis results. These results include correlation analysis between parameters (such as the relationship between Voc and Isc), performance impact analysis (such as the impact of a change in a certain parameter on other parameters or overall performance), etc.

[0083] Step S403: Establish the electrical performance analysis model of the single battery cell based on the data analysis results.

[0084] In this embodiment, a single-cell electrical performance analysis model is established based on the data analysis results. This model can be a statistical regression model, classification model, or clustering model, used to predict cell performance, identify performance anomalies, or optimize performance.

[0085] It should be noted that the model building process may require consideration of various factors, such as the interaction between parameters and the impact of environmental factors on performance.

[0086] Please see again Figure 1 Step S104: Based on preset alarm rules, issue an early warning according to the abnormal process and the abnormal ratio.

[0087] In this embodiment, preset alarm rules are used to issue early warnings based on abnormal information reflected by abnormal processes and abnormal proportions, thereby realizing automatic early warning of the production execution system and improving the efficiency of investigation and early warning. The preset alarm rules are shown in Table 1:

[0088] Table 1:

[0089] Warning level illustrate plan action Normal P1 Green light, mass production in progress Normal, no action required. No action taken Level 3 warning P2 Yellow light indicates a conversion efficiency of <26.3% or a yield of <95%. Manual alarm cancellation is in mass production; inspection and filing required. Manual judgment, early warning Level 2 expected P3 The red light indicates that 1000 consecutive samples were found to be inefficient, with an average photoelectric conversion efficiency of <26%. EMS system locates abnormal equipment and does not accept feeding. Single-machine shutdown maintenance Level 2 warning P3 A red light indicates that 1000 wafers with abnormal yield were detected consecutively, and the first-pass yield was less than 90%. EMS system locates abnormal equipment and does not accept feeding. Single-machine shutdown maintenance Level 1, expected P4 The average photoelectric conversion efficiency of red-light audible and visual alarm factories is <25%. Problems with the process formula, abnormalities in the entire process. Shut down the machine and switch to a different production process. Level 1 warning P4 The average yield and first-pass yield of factories using red-light audible and visual alarms are less than 80%. Problems with the process formula, abnormalities in the entire process. Shut down the machine and switch to a different production process.

[0090] Please see Figure 5 In one specific embodiment, step S104 includes steps S1041 to S1042, and each step is described in detail below.

[0091] Step S1041: Determine the abnormal equipment based on the abnormal process and the abnormal ratio.

[0092] In this embodiment, production data related to abnormal processes are collected and analyzed, including equipment operating status, operation records, maintenance history, etc.

[0093] Furthermore, by combining the characteristics of abnormal processes (such as frequency of occurrence, duration, and scope of impact) and the proportion of abnormal processes (i.e. the proportion of abnormal processes in the entire production process), the abnormal processes are associated with specific production equipment.

[0094] Furthermore, based on the correlation results, it is determined which devices are identified as abnormal devices, i.e. devices that may have malfunctions or performance degradation.

[0095] In one specific embodiment, step S1041 includes: acquiring shared camera position data and calculating the shared camera position ratio based on the shared camera position data; and determining the abnormal equipment based on the shared camera position ratio, the abnormal process, and the abnormal ratio.

[0096] In this embodiment, shared machine location data is obtained from data sources such as production management systems, automated monitoring equipment, or manually recorded production logs.

[0097] Furthermore, dividing the number of shared workstations by the total number of workstations yields the shared workstation percentage. This percentage reflects the relative quantity and importance of shared equipment in the production system. The shared workstation percentage can be used to assess risks related to the flexibility, reliability, and maintenance costs of the production system.

[0098] Furthermore, based on the known shared machine space data, and by combining abnormal processes and the proportion of abnormal processes, the scope of abnormal equipment can be further narrowed down. Specifically, by combining the proportion of shared machine spaces, the characteristics of abnormal processes, and the proportion of abnormal processes, a comprehensive judgment can be made as to which equipment is identified as abnormal. These devices may be identified as abnormal due to frequent failures, excessive load, performance degradation, or other reasons.

[0099] Step S1042: Based on the preset early warning rules, issue an early warning according to the abnormal process and the abnormal equipment.

[0100] In this embodiment, an early warning mechanism is triggered based on the identified abnormal process and abnormal equipment, as well as preset early warning rules.

[0101] In one specific embodiment, step S1042 includes: determining warning information based on the preset warning rules, according to the abnormal process and the abnormal equipment; and issuing a warning based on the warning information.

[0102] In this embodiment, pre-defined early warning rules are used to determine early warning information based on abnormal processes and abnormal equipment, and then early warnings are issued based on this information. The early warning information may include a detailed description of the abnormal process, the name and location of the abnormal equipment, possible causes of failure, suggested corrective measures, and the level of urgency. The early warning information is sent to relevant personnel in a pre-defined manner so that they can be informed of the situation and respond promptly.

[0103] It should be noted that there are two types of abnormalities in the test line: one is a single machine abnormality in the photovoltaic cell factory, and the other is an abnormality in the entire photovoltaic cell factory line.

[0104] For abnormal processes under single-machine malfunction, due to the mismatch in capacity between upstream and downstream processes, tape-out will be produced through different paths. The specific abnormal equipment will be automatically identified based on the abnormality ratio of each line and the proportion of shared machine positions used.

[0105] In the event of an abnormal process under abnormal conditions across the entire production line, due to the mismatch in capacity between upstream and downstream processes, the wafers will be fabricated from different paths. Based on the abnormal proportion of each line and the percentage of shared machine stations used, the system will automatically determine if the abnormal proportions across the entire production line are equal, and thus identify a process formulation problem.

[0106] The production execution system early warning method proposed in this embodiment collects raw efficiency data from each test line; removes abnormal data from the raw efficiency data based on abnormal data indicators to obtain target data; obtains abnormal processes and abnormal proportions based on an anomaly analysis data model and the target data; and issues warnings based on preset warning rules and the abnormal processes and abnormal proportions. In this way, by using an anomaly analysis data model to analyze abnormal processes and abnormal proportions in the test line data collected by the production execution system and issuing automatic warnings, it can reduce the problems of poor product yield and consistency caused by the failure to detect anomalies in individual machines in photovoltaic manufacturing, which has long process flows and many machines in each process.

[0107] Example 2

[0108] Furthermore, this disclosure provides a production execution system 600, please refer to [link to relevant documentation]. Figure 6 The system includes:

[0109] The data acquisition module 601 is used to collect raw efficiency data for each test line.

[0110] Processing module 602 is used to remove abnormal data from the raw efficiency data to obtain target data;

[0111] The determination module 603 is used to obtain the abnormal process and the abnormal ratio based on the abnormality analysis data model and the target data;

[0112] The early warning module 604 is used to issue an early warning based on preset early warning rules, according to the abnormal process and the abnormal ratio.

[0113] Optionally, the processing module 602 is used to verify whether the raw efficiency data is correct based on the abnormal data indicator; if the raw efficiency data is correct, then the raw efficiency data is used as the target data; if the raw efficiency data is incorrect, then the raw efficiency data is used as the abnormal data.

[0114] Optionally, the early warning module 604 is used to determine abnormal equipment based on the abnormal process and the abnormal ratio; and to issue an early warning based on the preset early warning rules and the abnormal process and the abnormal equipment.

[0115] Optionally, the early warning module 604 is used to acquire shared camera position data and calculate the shared camera position ratio based on the shared camera position data; and to determine the abnormal equipment based on the shared camera position ratio, the abnormal process, and the abnormal ratio.

[0116] Optionally, the early warning module 604 is used to determine early warning information based on the preset early warning rules, the abnormal process, and the abnormal equipment; and to issue an early warning based on the early warning information.

[0117] Optionally, the system also includes a training module, which is used to obtain historical EL detection results of photovoltaic cells; and to train the EL defect classification model based on machine learning according to the historical EL detection results.

[0118] Optionally, a training module is used to obtain historical electrical performance parameters of a single solar cell; analyze the historical electrical performance parameters to obtain data analysis results, the data analysis results including parameter correlation information and performance impact information; and establish an electrical performance analysis model for the single solar cell based on the data analysis results.

[0119] The system provided in this embodiment can execute the steps of the production execution system early warning method provided in Embodiment 1. To avoid repetition, the steps will not be repeated.

[0120] The production execution system proposed in this embodiment collects raw efficiency data from each test line; it then removes abnormal data from the raw efficiency data based on abnormal data indicators to obtain target data; based on the anomaly analysis data model and the target data, it obtains abnormal processes and anomaly ratios; and based on preset early warning rules, it issues early warnings according to the abnormal processes and anomaly ratios. In this way, by using the anomaly analysis data model to analyze abnormal processes and anomaly ratios in the test line data collected by the production execution system and issuing automatic early warnings, it can reduce the problems of poor product yield and consistency caused by the failure to detect anomalies in individual machines in photovoltaic manufacturing, which has long process flows and many machines in each process.

[0121] Example 3

[0122] Furthermore, this disclosure provides a computer device including a memory and a processor. The memory stores a computer program, which, when executed by the processor, implements the production execution system early warning method described in Embodiment 1.

[0123] The device provided in this embodiment can execute the steps of the production execution system early warning method provided in Embodiment 1. To avoid repetition, the steps will not be repeated.

[0124] Example 4

[0125] This disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the production execution system early warning method described in Embodiment 1.

[0126] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0127] The computer-readable storage medium provided in this embodiment can implement the production execution system early warning method provided in Embodiment 1. To avoid repetition, it will not be described again here.

[0128] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limitations; therefore, other examples of exemplary embodiments may have different values.

[0129] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0130] The above-described embodiments are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the 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 modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A method for early warning in a production execution system, characterized in that, Applied to a production execution system, the method includes: Collect raw efficiency data for each test line; Abnormal data is removed from the original efficiency data based on abnormal data indicators to obtain the target data; Based on the anomaly analysis data model and the target data, the abnormal process and the anomaly ratio are obtained; Based on preset early warning rules, early warnings are issued according to the abnormal process and the abnormality ratio.

2. The method for early warning of a production execution system according to claim 1, characterized in that, The step of removing abnormal data from the original efficiency data based on abnormal data indicators to obtain target data includes: Verify the accuracy of the original efficiency data based on the abnormal data indicators. If the raw efficiency data is correct, then the raw efficiency data will be used as the target data. If the raw efficiency data is incorrect, then the raw efficiency data will be considered as the abnormal data.

3. The method for early warning of a production execution system according to claim 1, characterized in that, The method of issuing early warnings based on preset early warning rules, according to the abnormal process and the abnormality ratio, includes: The abnormal equipment is determined based on the abnormal process and the abnormal ratio. Based on the preset early warning rules, early warnings are issued according to the abnormal process and the abnormal equipment.

4. The early warning method for a production execution system according to claim 3, characterized in that, The step of determining abnormal equipment based on the abnormal process and the abnormal ratio includes: Acquire shared camera location data and calculate the shared camera location percentage based on the shared camera location data; The abnormal equipment is determined based on the shared machine position ratio, the abnormal process, and the abnormal ratio.

5. The early warning method for a production execution system according to claim 3, characterized in that, The step of issuing early warnings based on the preset early warning rules, according to the abnormal process and the abnormal equipment, includes: Based on the preset early warning rules, early warning information is determined according to the abnormal process and the abnormal equipment; A warning will be issued based on the aforementioned warning information.

6. The method for early warning of a production execution system according to claim 1, characterized in that, The anomaly analysis data model includes the EL defect classification model, and the method further includes: Obtain historical EL test results for photovoltaic cells; The EL defect classification model is obtained by training based on the historical EL detection results using machine learning.

7. The method for early warning of a production execution system according to claim 1, characterized in that, The anomaly analysis data model includes a single-cell electrical performance analysis model, and the method further includes: Obtain historical electrical performance parameters of a single solar cell; Based on the historical electrical performance parameters, data analysis results are obtained, including parameter correlation information and performance impact information. Based on the data analysis results, an electrical performance analysis model for the single-cell battery is established.

8. A production execution system, characterized in that, The system includes: The data acquisition module is used to collect raw efficiency data for each test line. The processing module is used to remove abnormal data from the raw efficiency data to obtain the target data; The determination module is used to obtain the abnormal process and the abnormality ratio based on the anomaly analysis data model and the target data; The early warning module is used to issue early warnings based on preset early warning rules, according to the abnormal process and the abnormal ratio.

9. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, implements the production execution system early warning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the production execution system early warning method as described in any one of claims 1 to 7.

Images