Polysilicon silicon rod production information tracing method, device, equipment and medium

Abstract

The present application relates to a kind of polycrystalline silicon silicon rod production information traceability method, device, equipment and medium, including the basic information of the silicon powder for producing polycrystalline silicon silicon rod as data field, import to label code generation system, and based on the query condition of pre-setting, first traceability information is generated according to data field by label code generation system;According to the growth information of polycrystalline silicon silicon rod in the production process, and the growth information of the polycrystalline silicon silicon rod generated under the same growth condition with polycrystalline silicon silicon rod history, determine second traceability information;According to the event information of polycrystalline silicon silicon rod in the process of transportation and packaging line circulation, generate third traceability information, according to third traceability information, the event information corresponding to the transportation and packaging of polycrystalline silicon silicon rod is prompted in the pop-up window of inspection system;According to the first traceability information, second traceability information and third traceability information of polycrystalline silicon silicon rod, generate and print the production information identification of polycrystalline silicon silicon rod.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of polycrystalline silicon production, in particular to a polycrystalline silicon silicon rod production information tracing method, device, equipment and medium. BACKGROUND

[0002] In the polycrystalline silicon production process, the quality tracing of the silicon rod is usually performed by manually searching from various business systems such as WMS (Warehouse Management System), MES (Manufacturing Execution System), inspection system and logistics video monitoring system according to batch information after customer feedback.

[0003] Firstly, the quality inspection and production team of the batch are found, and then the relevant monitoring information of the corresponding time is queried according to the production time of the batch. In this way, on the one hand, it is time-consuming and laborious to find the target scene in numerous video resources, and it is easy to miss, mistake and disorder; on the other hand, there is a serious disconnection with the process management of the enterprise. SUMMARY

[0004] The purpose of the present application is to provide a polycrystalline silicon silicon rod production information tracing method, device, equipment and medium, which aims to solve the above problems, by combining video monitoring of the production process, labeling the video at each key node, pop-up processing of abnormal events in the silicon rod flow process of the product transportation and packaging line, and quickly retrieving the target video according to the label information when querying the video. Thus, the query cost is reduced and the relevance of management is improved.

[0005] In order to achieve the above purpose, the first aspect of the embodiment of the present disclosure provides a polycrystalline silicon silicon rod production information tracing method, which comprises:

[0006] The basic information of the silicon powder for producing the polycrystalline silicon silicon rod is imported into a label code generation system as a data field, and a first tracing information is generated by the label code generation system according to the data field based on a pre-set query condition;

[0007] The second tracing information of the polycrystalline silicon silicon rod is determined according to the growth information of the polycrystalline silicon silicon rod in the production process and the growth information of the polycrystalline silicon silicon rod generated under the same growth condition;

[0008] The third tracing information of the polycrystalline silicon silicon rod is generated according to the event information of the polycrystalline silicon silicon rod in the production, transportation and packaging line flow process, and the event information corresponding to the transportation and packaging of the polycrystalline silicon silicon rod is prompted in the inspection system according to the third tracing information;

[0009] According to the first traceability information, the second traceability information and the third traceability information of the polysilicon silicon rod, production information identification of the polysilicon silicon rod is generated and printed, wherein the production information identification is used to query production information of the polysilicon silicon rod.

[0010] In a possible implementation manner, the second traceability information of the polysilicon silicon rod is determined according to the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod historically generated under the same growth condition as the polysilicon silicon rod, and the second traceability information of the polysilicon silicon rod comprises:

[0011] a growth Pearson correlation coefficient between the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod historically generated under the same growth condition as the polysilicon silicon rod.

[0012] determining whether the growth of the polysilicon silicon rod is at a critical node according to the growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition.

[0013] if the growth of the polysilicon silicon rod is at the critical node, a critical node label is added to the corresponding monitoring video.

[0014] generating the second traceability information according to the production information of the polysilicon silicon rod in the production process and the monitoring video corresponding to the critical node label.

[0015] In a possible implementation manner, the growth information comprises a growth diameter and a growth length, and the growth Pearson correlation coefficient between the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod historically generated under the same growth condition as the polysilicon silicon rod comprises:

[0016] discretizing the growth diameter and the growth length of the polysilicon silicon rod in the production process according to production time points to obtain the growth diameter and the growth length corresponding to different production time points;

[0017] a first growth Pearson correlation coefficient between the growth diameter of the polysilicon silicon rod at different production time points in the production process and the growth diameter of the polysilicon silicon rod historically generated under the same growth condition as the polysilicon silicon rod.

[0018] a second growth Pearson correlation coefficient between the growth length of the polysilicon silicon rod at different production time points in the production process and the growth length of the polysilicon silicon rod historically generated under the same growth condition as the polysilicon silicon rod.

[0019] In a possible implementation, the determining whether the growth of the polysilicon silicon rod is at a critical point according to a growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition comprises:

[0020] obtaining a first threshold set and a second threshold set, wherein the first threshold set and the second threshold set respectively include a plurality of first thresholds and a plurality of second thresholds;

[0021] comparing a first growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition with the plurality of first thresholds in the first threshold set to obtain a first comparison result;

[0022] comparing a second growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition with the plurality of second thresholds in the second threshold set to obtain a second comparison result;

[0023] determining whether the growth of the polysilicon silicon rod is at a critical point according to the first comparison result and the second comparison result.

[0024] In a possible implementation, the generating third trace information of the polysilicon silicon rod according to event information of the polysilicon silicon rod in a production, transportation and packaging line flow process comprises:

[0025] generating a production event according to the event information of the polysilicon silicon rod in the production, transportation and packaging line flow process;

[0026] determining whether the production event of the polysilicon silicon rod is an abnormal event;

[0027] if the production event of the polysilicon silicon rod is an abnormal event, labeling monitoring video in a production, transportation and packaging process of the polysilicon silicon rod as an abnormal label;

[0028] generating third trace information of the polysilicon silicon rod according to the abnormal label of the polysilicon silicon rod and the monitoring video labeled with the abnormal label.

[0029] In a possible implementation, the determining whether the production event of the polysilicon silicon rod is an abnormal event comprises:

[0030] configuring a historical operation curve model according to a temperature in a reduction furnace, a concentration of a feed gas, a material flow and an impurity content of the polysilicon silicon rod in a previous process, and predicting a predicted impurity content of the polysilicon silicon rod in a subsequent process;

[0031] determine whether the polysilicon silicon rod has abnormal impurity content in each of the processes according to the predicted impurity content of each of the processes and the corresponding impurity content threshold set for each of the processes;

[0032] store real-time pictures of the polysilicon silicon rod in the transportation and packaging line, historical personnel operation records, quality information retrieval records and equipment state monitoring information into an event information database;

[0033] identify the real-time pictures, the historical personnel operation records, the quality information retrieval records and the equipment state monitoring information to determine whether there is information abnormality;

[0034] determine whether the production event of the polysilicon silicon rod is an abnormal event according to the first result of whether the polysilicon silicon rod has abnormal impurity content in each of the processes and the second result of whether there is information abnormality.

[0035] In a possible implementation manner, the historical operation curve model is obtained by training in the following manner:

[0036] construct a regression equation: I = β0 + β1T + β2C + β3*F + ε, wherein I is the impurity content, T is the temperature in the reduction furnace, C is the concentration of the feed gas, F is the material flow of the polysilicon silicon rod in the previous process, β0 is the intercept term, β1, β2 and β3 are regression coefficients, and ε is an error term;

[0037] estimate the intercept term β0, the regression coefficient β1, the regression coefficient β2 and the regression coefficient β3 of the regression equation by using the least square method according to the temperature in the reduction furnace, the concentration of the feed gas, the material flow of the polysilicon silicon rod in the previous process and the impurity content of the polysilicon silicon rod corresponding to the historical production;

[0038] after the intercept term β0, the regression coefficient β1, the regression coefficient β2 and the regression coefficient β3 of the regression equation are determined, adjust the model parameters by cross-validation to obtain the historical operation curve model.

[0039] A second aspect of the embodiments of the present disclosure provides a polysilicon silicon rod production information tracing device, which comprises:

[0040] The first generation module is configured to import basic information of silicon powder for producing the polysilicon silicon rod as a data field into a label code generation system, and generate first tracing information according to the data field by the label code generation system based on a pre-set query condition;

[0041] a second generation module configured to determine second trace information of the polysilicon silicon rod according to growth information of the polysilicon silicon rod in a production process and growth information of the polysilicon silicon rod historically generated under the same growth condition as the polysilicon silicon rod;

[0042] a third generation module configured to generate third trace information of the polysilicon silicon rod according to event information of the polysilicon silicon rod in a transportation and packaging line circulation process, and make a pop-up prompt in an inspection system to event information corresponding to transportation and packaging of the polysilicon silicon rod according to the third trace information;

[0043] a fourth generation module configured to generate and print production information identification of the polysilicon silicon rod according to the first trace information, the second trace information and the third trace information of the polysilicon silicon rod, wherein the production information identification is used to query production information of the polysilicon silicon rod.

[0044] In a third aspect of the embodiments of the present disclosure, an electronic device is provided, comprising:

[0045] a memory having a computer program stored thereon;

[0046] a processor configured to execute the computer program in the memory to implement the steps of the method according to any one of the first aspect.

[0047] In a fourth aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, having a computer program stored thereon, which, when executed by a processor, implements the steps of the method according to any one of the first aspect.

[0048] The present disclosure provides a polysilicon silicon rod production information traceability method, device, equipment and medium. Compared with the prior art, the following beneficial effects are achieved:

[0049] By integrating the basic information of the silicon powder, the growth information of the silicon rod and the event information in the production, transportation and packaging process, the method can comprehensively cover the whole chain of polysilicon silicon rod production, ensuring the completeness and accuracy of the information. The first trace information is generated by the label code generation system, and the second trace information is determined based on the growth information and historical data. The third trace information is generated in combination with the event information in the production circulation process, and finally the production information identification is generated and printed, thereby enhancing the convenience of information traceability.

[0050] Not only the basic information and growth process of the polysilicon silicon rod are recorded, but also every important event in the production, transportation and packaging process is recorded in detail. This helps the enterprise to realize the refinement of production management and timely find and solve problems. The refinement of production management is improved. By comparing the growth information of the current silicon rod with the historical silicon rod, the abnormality in the production process can be found in time, so as to strengthen the quality control. At the same time, the event information in the transportation and packaging process is prompted by pop-up window, which helps the enterprise to find and deal with potential risks in time, and promotes the informatization and intelligent production.

[0051] In summary, the digitalization, networking and intelligent management of the polysilicon silicon rod production information are realized. This helps the enterprise to improve production efficiency, reduce cost and enhance market competitiveness. Through comprehensive, convenient and fine information tracing method, the level and quality of polysilicon silicon rod production management are effectively improved, and the informatization and intelligent production of the enterprise are promoted.

[0052] Other features and advantages of the present disclosure will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0053] The accompanying drawings are included to provide a further understanding of the present disclosure and constitute a part of the specification, which together with the specific embodiments below, serve to explain the present disclosure but do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 is a flow chart of a polysilicon silicon rod production information tracing method according to an embodiment of the specification.

[0055] Figure 2 is a block diagram of a polysilicon silicon rod production information tracing device according to an embodiment of the specification.

[0056] Figure 3 is a block diagram of another polysilicon silicon rod production information tracing device according to an embodiment of the specification. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0058] The specific embodiments of the present disclosure will be described in detail below with reference to the drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure, and are not used to limit the present disclosure.

[0059] The disclosure provides a polysilicon silicon rod production information tracing method, Figure 1 is a flowchart of a polysilicon silicon rod production information tracing method according to an embodiment. Specifically, the method comprises:

[0060] In step S11, the basic information of the silicon powder for producing the polysilicon silicon rod is imported into the label code generation system as a data field, and based on the pre-set query condition, the first tracing information is generated by the label code generation system according to the data field;

[0061] Wherein, the polysilicon silicon rod is a columnar silicon material generated by high-purity silicon powder at high temperature, such as polysilicon silicon rod applied to solar cells, semiconductors, etc. The label code generation system is a computer system capable of generating a label code (such as a bar code, a two-dimensional code, etc.) with uniqueness or specific rules based on the input data field (such as the basic information of the silicon powder).

[0062] In the embodiment of the disclosure, the basic information of the silicon powder (such as batch number, purity, source, etc.) is imported into the label code generation system as a data field. The system generates a label code corresponding to the silicon powder information according to the pre-set query condition (such as batch number, production date, etc.) and coding rule, as the first tracing information. The information of the source, batch, quality, etc. of the silicon powder is recorded to ensure the traceability of the raw material. Through the unique identification of the raw material, the related information can be quickly queried to provide reliable raw material guarantee for the production process.

[0063] For example: assuming that there is a batch of silicon powder with batch number 20230101, purity 99.999%, and source from a certain silicon material supplier. After importing these information into the label code generation system, the system generates a unique two-dimensional code label containing the basic information of the silicon powder such as batch number, purity, source, etc., which is convenient for subsequent tracing.

[0064] In step S12, the second tracing information of the polysilicon silicon rod is determined according to the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod historically generated under the same growth conditions as the polysilicon silicon rod;

[0065] Wherein, the growth information can be the temperature, pressure, gas flow, etc. of the silicon rod in the growth furnace during the production process of the polysilicon silicon rod, as well as the growth period and other information.

[0066] In the embodiment of the disclosure, the growth information of the polysilicon silicon rod in the production process (such as temperature, pressure, gas flow, etc.) and the growth information of the silicon rod historically generated under the same growth conditions as the silicon rod are analyzed and compared to determine the growth state and quality characteristics of the current silicon rod as the second tracing information.

[0067] For example, during the growth of a silicon rod, the system records the temperature, pressure and other parameters in the growth furnace in real time. At the same time, the system retrieves the silicon rod growth information under the same growth conditions from the historical data for comparison and analysis. If it is found that the growth parameters of the current silicon rod deviate greatly from the historical data, the system will issue a warning to prompt the operator to make adjustments.

[0068] In step S13, according to the event information of the polysilicon silicon rod in the production, transportation and packaging line flow process, the third traceability information of the polysilicon silicon rod is generated, and according to the third traceability information, the event information corresponding to the transportation and packaging of the polysilicon silicon rod is prompted in the inspection system.

[0069] The inspection system is a computer system for monitoring and checking the running state of the production line or equipment, and timely discovering and handling exceptions.

[0070] In the embodiments of the present disclosure, during the production, transportation and packaging line flow process, the system records each important event (such as warehousing, delivery, transportation, packaging, etc.) experienced by the silicon rod, and generates corresponding event information. These information as the third traceability information, is used to track the flow process of the silicon rod. At the same time, the system will pop up a window in the inspection system to prompt the event information corresponding to the transportation and packaging of the silicon rod according to the third traceability information, so that the operator can timely discover and handle abnormal situations.

[0071] For example, when the silicon rod is taken out of the growth furnace, the system records the warehousing time and storage location information. When the silicon rod needs to be transported to the packaging line, the system records the transportation time, transportation method, transportation personnel and other information. During the packaging process, the system records the packaging time, packaging materials, packaging personnel and other information. If an abnormality occurs in a certain link (such as the packaging material does not meet the requirements), the system will pop up a prompt window in the inspection system to remind the operator to handle it in time.

[0072] In step S14, according to the first traceability information, the second traceability information and the third traceability information of the polysilicon silicon rod, the production information identification of the polysilicon silicon rod is generated and printed, wherein the production information identification is used to query the production information of the polysilicon silicon rod.

[0073] The production information identification is information that can track and record the whole process of a product from raw materials to finished products, including raw material information, production process information, transportation information, quality detection information, etc.

[0074] In the embodiments of the present disclosure, the first trace information, the second trace information and the third trace information are integrated and processed to generate a production information identifier containing all production information of the silicon rod. The identifier can be in the form of a bar code, a two-dimensional code or the like, facilitating subsequent query and trace. Then, the system will print the production information identifier and attach it to the silicon rod or deliver it together with the packaging of the silicon rod to the customer.

[0075] For example, on the basis of the completion of the first three steps, the system generates a production information identifier containing the basic information of the silicon powder, the growth information, the flow event information and the like. The identifier is presented in the form of a two-dimensional code and printed on the packaging of the silicon rod. Customers or subsequent users can quickly obtain all production information of the silicon rod by scanning the two-dimensional code.

[0076] The above technical solution integrates the basic information of the silicon powder, the growth information of the silicon rod and the event information in the production, transportation and packaging process. The method can comprehensively cover the whole chain of production of the multicrystalline silicon rod, ensuring the completeness and accuracy of the information. The first trace information is generated by the label code generation system, the second trace information is determined based on the growth information and historical data, the third trace information is generated by combining the event information in the production flow process, and finally the production information identifier is generated and printed, enhancing the convenience of information trace.

[0077] Not only the basic information and the growth process of the multicrystalline silicon rod are recorded, but also every important event in the production, transportation and packaging process is recorded in detail. This helps enterprises to realize fine production management and timely find and solve problems. The fine production management is improved. By comparing the growth information of the current silicon rod with that of the historical silicon rod, abnormalities in the production process can be found in time, so as to strengthen quality control. At the same time, the event information in the transportation and packaging process is prompted in a pop-up window, which helps enterprises to timely find and respond to potential risks, promoting informationization and intelligent production.

[0078] In summary, the digital, networked and intelligent management of the production information of the multicrystalline silicon rod is realized. This helps enterprises to improve production efficiency, reduce cost and enhance market competitiveness. Through comprehensive, convenient and fine information trace, the level and quality of production management of the multicrystalline silicon rod are effectively improved, promoting the informationization and intelligent production of enterprises.

[0079] In one possible implementation, in step S12, the second trace information of the multicrystalline silicon rod is determined according to the growth information of the multicrystalline silicon rod in the production process and the growth information of the multicrystalline silicon rod generated under the same growth conditions in history, comprising:

[0080] In step S121, a growth Pearson correlation coefficient between the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod historically generated under the same growth condition is calculated according to the growth information of the polysilicon silicon rod in the production process.

[0081] The Pearson correlation coefficient is a statistical measure of the linear correlation between two variables, with a value ranging from -1 to 1. In the context of polysilicon silicon rod production, the growth Pearson correlation coefficient is used to measure the linear correlation between the growth information of the current silicon rod and the growth information of the silicon rod under the same historical condition. A high correlation coefficient indicates that the growth trends of the two are similar, and a low correlation coefficient may indicate that the growth of the current silicon rod is abnormal.

[0082] In the embodiments of the present disclosure, the Pearson correlation coefficient formula in statistics is used to calculate the correlation coefficient between the growth information (such as temperature, pressure, gas flow, etc.) of the current polysilicon silicon rod and the growth information of the silicon rod under the same historical condition. The calculation of the correlation coefficient is based on the covariance and standard deviation of the two sets of data, which reflects the strength of the linear relationship between the two.

[0083] For example, assume that the growth temperature of the current silicon rod is T1, the pressure is P1, and the gas flow is F1; while the growth temperature of the silicon rod under the same historical condition is T2, the pressure is P2, and the gas flow is F2. By calculating the Pearson correlation coefficient between the two sets of data, it can be evaluated whether the growth trend of the current silicon rod is consistent with the historical data.

[0084] In step S122, it is determined whether the growth of the polysilicon silicon rod is at a critical node according to the growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition.

[0085] The critical node is a production stage or time point in the production process of the polysilicon silicon rod that has a significant impact on the quality, performance or production efficiency of the silicon rod. These nodes are usually related to significant changes in growth parameters, potential quality problems or critical operations in the production process.

[0086] In the embodiments of the present disclosure, a threshold is set according to the calculated growth Pearson correlation coefficient to determine whether the growth of the current silicon rod is at a critical node. If the correlation coefficient is lower than the threshold, it indicates that the growth trend of the current silicon rod is significantly different from the historical data, and it may be at a critical node.

[0087] For example, if the set threshold is 0.8, and the calculated growth Pearson correlation coefficient between the current silicon rod and the historical silicon rod is 0.6, which is lower than the threshold, it is determined that the growth of the current silicon rod is at a critical node.

[0088] In step S123, if the growth of the polysilicon silicon rod is at the critical node, a critical node label is added to the corresponding monitoring video.

[0089] The monitoring video critical node label is a label added to the monitoring video during production to record and analyze detailed information of the critical node. These labels are called critical node labels, which are used to indicate specific time periods or frames in the video that contain information about the critical node.

[0090] In the embodiments of the present disclosure, the monitoring system continuously records the growth of the silicon rod during production. When it is determined that the silicon rod is at a critical node, the system automatically or manually adds a critical node label to the corresponding monitoring video for subsequent analysis and tracing.

[0091] For example, when the system determines that the current silicon rod is growing at a critical node, it automatically finds the time period corresponding to the node in the monitoring video and adds critical node labels such as "critical node_start" and "critical node_end" before and after the time period.

[0092] In step S124, the second tracing information is generated based on the production information of the polysilicon silicon rod during production and the monitoring video corresponding to the critical node label.

[0093] In the embodiments of the present disclosure, the second tracing information is generated based on the production information of the current silicon rod during production (such as batch number, growth time, growth parameters, etc.) and the monitoring video corresponding to the critical node label. These information are used to record the key stages and abnormal conditions of the silicon rod during growth in detail.

[0094] For example, suppose the batch number of the current silicon rod is 20230401, the growth time is from April 1 to April 5, and it is at a critical node on April 3. Then, the second tracing information will include the batch number, the growth time, the critical node label on April 3, and the corresponding monitoring video link or summary. These information will be integrated into the production information identifier for subsequent query and tracing.

[0095] The above technical solution can accurately identify whether the growth state of the current silicon rod is significantly different from the historical data by calculating the Pearson correlation coefficient of the growth of the current silicon rod and the silicon rod under the same conditions. This difference may indicate potential growth abnormalities or quality problems, providing a basis for timely corrective action.

[0096] According to the comparison of the growth Pearson correlation coefficient with the preset threshold value, the technology can intelligently locate the key nodes in the growth process of the silicon rod. These nodes are usually related to significant changes in growth parameters, potential quality problems or key operations in the production process, and are the focus of monitoring and tracing. The key nodes can be intelligently located.

[0097] For the silicon rod at the key node, the technology can automatically label the corresponding monitoring video with the key node label. This not only facilitates subsequent video retrieval and analysis, but also improves the management efficiency of monitoring video. Combined with the production information of the silicon rod in the production process and the corresponding monitoring video of the key node label, the technology can comprehensively generate the second trace information. These information details the key stages and abnormal conditions of the silicon rod in the growth process, providing strong data support for product quality tracing and problem analysis.

[0098] In summary, by introducing the growth Pearson correlation coefficient and the key node label, the technology realizes the fine monitoring and tracing of the production process of the polycrystalline silicon silicon rod. This not only improves the production efficiency and quality level, but also provides strong technical support for the continuous improvement and intelligent production of enterprises.

[0099] In one possible implementation, the growth information includes growth diameter and growth length, wherein the growth diameter refers to the cross-sectional diameter of the silicon rod at a certain growth time point in the production process of the polycrystalline silicon silicon rod. It is one of the important indicators for measuring the growth speed and morphology of the silicon rod. The growth length refers to the total length of the silicon rod from the beginning of growth to a certain growth time point. It reflects the growth progress and efficiency of the silicon rod.

[0100] In step S121, the growth Pearson correlation coefficient between the growth information of the polycrystalline silicon silicon rod in the production process and the growth information of the polycrystalline silicon silicon rod historically generated under the same growth conditions comprises:

[0101] In step S1211, the growth diameter and the growth length of the polycrystalline silicon silicon rod in the production process are discretized according to the production time point to obtain the growth diameter and the growth length corresponding to different production time points;

[0102] Wherein, discretization refers to the process of dividing continuous data set into several discrete intervals or points. In the present technology, discretization is used to divide the growth diameter and growth length of the polycrystalline silicon silicon rod according to the production time point, so as to calculate the correlation coefficient subsequently.

[0103] In the embodiments of the present disclosure, in order to calculate the growth Pearson correlation coefficient, the growth diameter and the growth length of the polysilicon silicon rod in the production process need to be discretized according to the production time points first. This means that the continuous growth process is divided into several discrete time points, and the corresponding growth diameter and growth length are recorded for each time point.

[0104] For example, assuming that the production process of the polysilicon silicon rod lasts for one week, and the growth diameter and the growth length are recorded once a day. Then, the production process of this week can be discretized into 7 time points (i.e. 7 days), and the corresponding growth diameter and growth length are recorded for each time point.

[0105] In step S1212, a first growth Pearson correlation coefficient is calculated between the growth diameter of the polysilicon silicon rod at different production time points in the production process and the growth diameter of the polysilicon silicon rod generated historically under the same growth conditions.

[0106] In the embodiments of the present disclosure, the first growth Pearson correlation coefficient is used to measure the degree of linear correlation between the growth diameter of the current silicon rod at different production time points and the growth diameter of the silicon rod under the same historical conditions. By calculating the correlation coefficient between the two sets of data, it can be evaluated whether the growth diameter of the current silicon rod is consistent with the historical data or there is a significant difference.

[0107] For example, assuming that the growth diameter of the current silicon rod on the first day is D1, the growth diameter on the second day is D2,..., and the growth diameter on the seventh day is D7; while the growth diameter of the silicon rod under the same historical conditions on the first day is D1_hist, the growth diameter on the second day is D2_hist,..., and the growth diameter on the seventh day is D7_hist. Then, the Pearson correlation coefficient between the two sets of data can be calculated to evaluate whether the growth diameter of the current silicon rod is consistent with the historical data.

[0108] In step S1213, a second growth Pearson correlation coefficient is calculated between the growth length of the polysilicon silicon rod at different production time points in the production process and the growth length of the polysilicon silicon rod generated historically under the same growth conditions.

[0109] In the embodiments of the present disclosure, similar to the first growth Pearson correlation coefficient, the second growth Pearson correlation coefficient is used to measure the degree of linear correlation between the growth length of the current silicon rod at different production time points and the growth length of the silicon rod under the same historical conditions. By calculating the correlation coefficient between the two sets of data, it can be evaluated whether the growth length of the current silicon rod is consistent with the historical data or there is a significant difference.

[0110] For example, assume that the current silicon rod has a growth length of L1 on day 1, a growth length of L2 on day 2,..., and a growth length of L7 on day 7, while the historical silicon rod under the same conditions has a growth length of L1_hist on day 1, a growth length of L2_hist on day 2,..., and a growth length of L7_hist on day 7. Then, the consistency of the growth length of the current silicon rod with the historical data can be evaluated by calculating the Pearson correlation coefficient between the two sets of data.

[0111] In other words, the growth Pearson correlation coefficient between the polycrystalline silicon rod and the historical polycrystalline silicon rod grown under the same conditions includes a first growth Pearson correlation coefficient and a second growth Pearson correlation coefficient. In this way, by calculating the first growth Pearson correlation coefficient and the second growth Pearson correlation coefficient, the consistency of the growth state of the current silicon rod with the historical data can be comprehensively evaluated, thereby providing strong data support for subsequent key node determination and traceability information generation.

[0112] In one possible implementation, in step S122, determining whether the growth of the polycrystalline silicon rod is at a key node according to the growth Pearson correlation coefficient between the polycrystalline silicon rod and the historical polycrystalline silicon rod grown under the same conditions includes:

[0113] In step S1221, a first threshold set and a second threshold set are obtained, where the first threshold set and the second threshold set respectively include a plurality of first thresholds and a plurality of second thresholds.

[0114] The threshold set is a plurality of correlation coefficient thresholds set. These thresholds are used to judge the similarity between the growth state of the current silicon rod and the historical data. Generally, these thresholds are set according to actual production experience and needs to ensure that the key node can be accurately identified.

[0115] In step S1222, the first growth Pearson correlation coefficient between the polycrystalline silicon rod and the historical polycrystalline silicon rod grown under the same conditions is compared with the plurality of first thresholds in the first threshold set, to obtain a first comparison result.

[0116] In the embodiments of the present disclosure, the calculated first growth Pearson correlation coefficient (i.e., the correlation coefficient between the growth diameter of the current silicon rod at different production time points and the historical data) is compared with the set thresholds respectively. If the correlation coefficient is lower than a certain threshold, it may mean that the growth diameter of the current silicon rod is relatively rare or abnormal in the historical data, and therefore it may be at a key node.

[0117] In step S1223, a second growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition is compared with a plurality of second threshold values in the second threshold set, to obtain a second comparison result;

[0118] Similarly, the second growth Pearson correlation coefficient (i.e., the correlation coefficient between the growth length of the current silicon rod at different production time points and the historical data) is compared with the set threshold value. If the correlation coefficient is also below a certain threshold value, it may further confirm that the growth state of the current silicon rod is abnormal, thereby increasing the possibility that it is at a critical node.

[0119] In step S1224, according to the first comparison result and the second comparison result, it is determined whether the growth of the polysilicon silicon rod is at a critical node.

[0120] In the embodiments of the present disclosure, after comparing the first growth Pearson correlation coefficient and the second growth Pearson correlation coefficient with the threshold value, a comprehensive judgment is made. If both correlation coefficients are below the threshold value, or one of the correlation coefficients is significantly below the threshold value and the other also shows a certain abnormality (such as close to the threshold value but below the historical average level), it can be considered that the growth of the current silicon rod is at a critical node.

[0121] In a possible implementation manner, in step S13, the third traceability information of the polysilicon silicon rod is generated according to the event information of the polysilicon silicon rod in the production, transportation and packaging line flow process, including:

[0122] In step S131, a production event is generated according to the event information of the polysilicon silicon rod in the production, transportation and packaging line flow process.

[0123] In the embodiments of the present disclosure, according to the preset rules and algorithms, the key production events are extracted from the event information collected from the polysilicon silicon rod in the production, transportation and packaging line flow process. These event information can come from various data sources such as sensors, monitoring cameras, RFID tags, etc. The system will integrate and analyze these information to generate a series of event records related to the production process.

[0124] For example, in the production process, the system can record events such as the start of silicon rod growth, the end of growth, entering the cooling stage, being transported to the next process, etc. In the transportation process, the system can record the transfer time of the silicon rod from the production line to the warehouse, the information of the transportation vehicle, etc. In the packaging line flow process, the system can record events such as the packaging of the silicon rod, the labeling, the warehousing, etc.

[0125] In step S132, it is determined whether the production event of the polysilicon silicon rod is an abnormal event.

[0126] In this embodiment of the disclosure, generated production events are screened and judged according to preset abnormal event definitions and judgment criteria to determine which events belong to abnormal events. These abnormal events may include equipment failures during the production process, deviations of process parameters from standards, damage or contamination of silicon rods, etc.

[0127] For example, if the system detects in the monitoring data that the growth rate of the silicon rod suddenly slows down or stops, or that obvious cracks or stains appear on the surface of the silicon rod, these may be considered abnormal events.

[0128] In step S133, if the production event of the polycrystalline silicon rod is an abnormal event, then an abnormal tag is added to the monitoring video of the polycrystalline silicon rod during production, transportation and packaging.

[0129] In this embodiment of the disclosure, once an abnormal event is identified, the system automatically retrieves and tags related surveillance videos based on the time and location information of the abnormal event. These videos will be tagged with abnormal tags for subsequent analysis and tracing.

[0130] For example, if the system detects an abnormal event such as a crack appearing on the surface of the silicon rod during the growth process, the system will automatically find and record the monitoring videos before and after the crack appears, and label these videos as "crack anomaly".

[0131] In step S134, third traceability information of the polycrystalline silicon rod is generated based on the abnormal label on the polycrystalline silicon rod and the monitoring video with the abnormal label.

[0132] The third traceability information refers to comprehensive data that details events (especially abnormal events) that occur at each stage of the polycrystalline silicon rod's production, transportation, and packaging process, along with related monitoring video information. It is used to trace the specific circumstances of the silicon rod during production, transportation, and packaging, ensuring product quality and process compliance.

[0133] In this embodiment of the disclosure, third traceability information for the polycrystalline silicon rod is generated based on the anomaly tag and the tagged surveillance video. This information will include a description of the anomaly event, its occurrence time, location, and a link to or storage location of the relevant surveillance video. This information will be integrated into a unified traceability system for subsequent quality control and process optimization.

[0134] For example, for a polysilicon silicon rod, its third traceability information can include: during the production process, the silicon rod was detected to have a crack on the surface at the end of the third day of growth, which is an abnormal event; the abnormal event occurred in area B of the growth workshop; the monitoring video related to the abnormal event has been labeled with "crack abnormality" and stored in a specific location of the server. These information will help relevant personnel quickly locate the cause of the problem, take corrective measures, and prevent similar problems from happening again.

[0135] In one possible implementation manner, in step S132, the determining whether the production event of the polysilicon silicon rod is an abnormal event comprises:

[0136] In step S1321, a historical running curve model is configured according to the temperature in the reduction furnace, the concentration of the feed gas, the material flow and the impurity content of the polysilicon silicon rod in the previous process, and the predicted impurity content of the polysilicon silicon rod in the subsequent process is predicted.

[0137] Wherein, the reduction furnace is a device used to convert silicon raw materials (such as trichlorosilane) into polysilicon at high temperature through chemical reaction (such as thermal reduction method) in the polysilicon production process. The feed gas is the gas input into the reduction furnace to participate in the chemical reaction, such as hydrogen, trichlorosilane, etc. The material flow is the amount of material passing through a certain cross section per unit time in the production process, here specifically refers to the flow of silicon raw materials in the previous process of polysilicon production. The impurity content is the content of other components in addition to the main component in the material, for polysilicon, impurities can include metal elements, non-metal elements or compounds, etc., which directly affect the purity of polysilicon. The historical running curve model is a mathematical model based on historical data, used to predict the future state of the system or process, here refers to the model used to predict the impurity content of the polysilicon silicon rod in the subsequent process.

[0138] In the embodiments of the present disclosure, a machine learning model or a statistical model is trained using historical data (including reduction furnace temperature, feed gas concentration, previous process material flow and impurity content, etc.) to form a historical running curve model. The model can predict the impurity content of the polysilicon silicon rod in the subsequent process based on the current production conditions.

[0139] For example, suppose the historical data shows that when the reduction furnace temperature is 1200℃, the trichlorosilane concentration in the feed gas is 95%, the previous process material flow is 100kg / h, and the impurity content is less than 0.01%, the impurity content in the subsequent process is usually less than 0.005%. Based on these data, the model can predict the impurity content in the subsequent process under the current conditions.

[0140] In step S1322, it is determined whether the polysilicon silicon rod has impurity content abnormality in each process according to the predicted impurity content of each process and the corresponding impurity content threshold set for each process.

[0141] In the embodiments of the present disclosure, the predicted impurity content is compared with the impurity content threshold set for each process to determine whether the impurity content exceeds the acceptable range. For example, if the impurity content threshold of a process is 0.008%, and the predicted impurity content is 0.012%, it is determined that the impurity content of the process is abnormal.

[0142] In step S1323, the real-time pictures of the polysilicon silicon rod in the transportation and packaging line, the historical personnel operation records, the quality information retrieval records and the equipment state monitoring information are stored in the event information database.

[0143] The event information database is a database system for storing various data related to production events, such as real-time pictures, operation records, quality information, etc. It can communicate with WMS, MES, LIMS, packaging line, inspection, etc.

[0144] In the embodiments of the present disclosure, during the production process, the real-time pictures of the polysilicon silicon rod in the transportation and packaging line, the personnel operation records, the quality information retrieval records and the equipment state monitoring information are collected and recorded in real time, and stored in the event information database. For example, the transportation process of the silicon rod from the production line to the packaging area is recorded, including the time stamp, the operator, the equipment state (such as whether it is running normally), the quality detection result, etc., and stored in the database.

[0145] In step S1324, the real-time pictures, the historical personnel operation records, the quality information retrieval records and the equipment state monitoring information are identified to determine whether there is information abnormality.

[0146] In the embodiments of the present disclosure, the stored information is identified by using data analysis technology (such as abnormal detection algorithm) to find out whether there is abnormal pattern or behavior, such as abnormally high error rate, non-standard operation, equipment failure, etc. For example, if the data analysis finds that the equipment failure rate suddenly increases in a certain period of time, or the operator frequently performs non-standard operation, it is determined that there is information abnormality.

[0147] In step S1325, it is determined whether the production event of the polysilicon silicon rod is an abnormal event according to the first result of whether the polysilicon silicon rod has impurity content abnormality in each process and the second result of whether there is information abnormality.

[0148] In the embodiments of the present disclosure, if the impurity content of the polysilicon silicon rod is abnormal in a certain process or information is abnormal, it is determined that the production event is an abnormal event. For example, if the predicted impurity content exceeds the threshold value in a certain process, and at the same time, it is found that the equipment state monitoring information in this period shows that the equipment has abnormal fluctuation, it is determined that the production event is an abnormal event, which needs to be further checked and corresponding measures are taken.

[0149] In a possible implementation manner, the historical operation curve model is obtained by training in the following manner:

[0150] A regression equation I = β0+ β1T + β2C + β3*F + ε is constructed, where I is the impurity content, T is the temperature in the reduction furnace, C is the concentration of the feed gas, F is the material flow of the polysilicon silicon rod in the previous process, β0 is the intercept term, β1, β2 and β3 are regression coefficients, and ε is an error term.

[0151] Where I is the dependent variable (the impurity content of the polysilicon silicon rod), T is one of the independent variables (the temperature in the reduction furnace), C is the second independent variable (the concentration of the feed gas), and F is the third independent variable (the material flow of the polysilicon silicon rod in the previous process).

[0152] According to the temperature in the reduction furnace, the concentration of the feed gas, the material flow of the polysilicon silicon rod in the previous process and the impurity content corresponding to the polysilicon silicon rod produced in the history, the intercept term β0 of the regression equation, the regression coefficient β1, the regression coefficient β2 and the regression coefficient β3 are estimated by using the least square method.

[0153] Where a design matrix X is constructed, each row of which corresponds to a data point, and each column of which corresponds to an independent variable (including the constant 1 of the intercept term).

[0154] The regression coefficients β = [β0, β1, β2, β3] can be calculated by using the least square method formula: β = (X T ×X) -1 ×X T ×Y, where Y is the vector form of the dependent variable I.

[0155] After the intercept term β0 of the regression equation, the regression coefficient β1, the regression coefficient β2 and the regression coefficient β3 are determined, the model parameters are adjusted by cross-validation to obtain the historical operation curve model.

[0156] In the embodiments of the present disclosure, after the regression coefficients are obtained, the remaining data points or the cross-validation method can be used to verify the accuracy of the model.

[0157] In the embodiments of the present disclosure, the data set is randomly divided into k parts, for each part, the remaining k-1 parts are used as the training set to fit the regression model, and the current fold is used as the validation set to evaluate the performance of the model. This involves calculating the error (such as mean square error MSE) between the predicted value and the actual value.

[0158] According to the results of cross-validation, the parameters of the regression model can be adjusted (although in this particular case, we are mainly interested in the coefficients obtained using least squares, but cross-validation can be used to evaluate the effects of different model structures or preprocessing steps).

[0159] After cross-validation is completed, the model that performs best on all folds is usually selected as the final model. However, in the regression problem, since we have already obtained the optimal solution of the coefficients using least squares (under the given data and model structure), cross-validation is more used to evaluate the stability and generalization ability of the model, rather than directly adjusting the coefficients.

[0160] Finally, by combining the regression coefficients obtained by least squares and the evaluation results of cross-validation, we can obtain a historical run curve model that can predict the impurity content of the polysilicon silicon rod according to the given temperature, concentration and flow. This model can be used to guide future production decisions to optimize product quality and production costs.

[0161] To achieve the above-mentioned purposes of the invention, the present invention adopts:

[0162] 1. Data acquisition and integration: Establish a powerful data acquisition platform covering process, equipment, safety, energy, operation and other aspects of data. Real-time acquisition of various data in the production process, connection of data from MES, WMS, inspection, video monitoring, and suggestion of unified data platform.

[0163] 2. Traceability system model construction: Based on the production model, the silicon powder input, reduction furnace loading, furnace discharge, silicon rod transportation, crushing and packaging, warehouse in and out, and logistics transportation process are comprehensively monitored. Through two-dimensional code or RFID tag marking of silicon powder, silicon rod, video monitoring recording, through data analysis, the overall situation of the polysilicon production process is mastered, and for abnormal conditions, the model can be analyzed to give reasons and treatment measures.

[0164] 3. Visual traceability system:

[0165] Raw material traceability: The raw materials are uniquely coded, and information such as batch number, source, and quality is recorded to achieve traceability of the raw materials. Production process traceability: Data collection points are set up at each production stage to record process parameters, equipment status, and operator information, which are associated with the production model to achieve visual traceability of the production process. Finished product traceability: The finished product is uniquely coded and associated with various data during the production process to achieve full traceability from raw materials to finished product. Video monitoring traceability: By using video watermarking OSD overlay and slice video backup functions, key information such as traceability number, station name, and scene name is overlaid on the video, making the evidence information more comprehensive.

[0166] 4. Intelligent analysis and early warning: The system has built-in abnormal analysis models that can trace and analyze abnormal conditions in the production process and provide operation adjustment suggestions in combination with the expert database and knowledge base. At the same time, using online analysis instruments and LIMS historical impurity content curves, an impurity early warning model is provided to predict the impurity content of the next process based on the material flow and impurity content of the previous process, so as to provide early warning and control. The threshold value configuration range is synchronized with the video platform, and real-time reminders are provided through the video screen. By analyzing the equipment operating state data, the occurrence of equipment failure can be predicted, and maintenance or replacement can be performed in advance. The intelligent analysis and early warning module is associated with other production management modules (production planning, inventory management, quality management, and energy consumption management), and in combination with the key parameter warning thresholds set by the production management system, the data is analyzed, and the analysis results are fed back to the production management system, which can achieve optimized control of inventory, quality, and energy consumption.

[0167] 5. Mobile terminal and Internet of Things technology: In combination with technologies such as two-dimensional codes, RFID, video monitoring, and mobile terminal devices, raw materials and equipment involved in the production process are labeled, positioned, and operated to achieve visual scheduling and production traceability, and to improve post-processing efficiency.

[0168] The technical solution of the present disclosure monitors key parameters in the production process in real time, such as the temperature in the reduction furnace and the concentration of the feed gas, to ensure the stability and controllability of the production process. The system can automatically detect and record key data such as the growth diameter of the silicon rod to provide data support for the optimization of the production process. In combination with the LIMS system and online analysis instrument data, an impurity early warning model is provided to predict the impurity content of the next process based on the material flow and impurity content of the previous process, and to provide early warning and control. The threshold value configuration range is synchronized with the video platform, and the video screen can provide real-time reminders for abnormal conditions and events in the reduction furnace feed. Real-time inventory information is obtained, including the number of silicon rods, tray information, storage location, and production date. The transportation information of the silicon rods is recorded, including the shipment batch, transportation route, transportation time, and destination. Through GPS and other technical means, the transportation process is monitored in real time to ensure the safe delivery of the silicon rods.

[0169] In addition to online real-time monitoring and early warning of video visualization, rich visualization reports and charts are also provided, such as production progress chart, quality control chart, inventory distribution chart, etc. Related personnel can check these information at any time through mobile phones, computers and other terminal devices, improving work efficiency and decision-making accuracy.

[0170] Each step in the production process of the polysilicon silicon rod can be recorded and traced, including raw material input, production processing, quality detection, packaging and storage, etc. Through video, barcode or RFID technology, the full life cycle of the silicon rod is traced to ensure the accuracy and integrity of the product information.

[0171] The disclosure also provides a polysilicon silicon rod production information tracing device, as shown in Figure 2 The device includes:

[0172] The first generation module 210 is configured to import the basic information of the silicon powder for producing the polysilicon silicon rod as a data field into a label code generation system, and generate first tracing information based on the data field through the label code generation system based on the pre-set query condition;

[0173] The second generation module 220 is configured to determine the second tracing information of the polysilicon silicon rod according to the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod generated under the same growth condition in history;

[0174] The third generation module 230 is configured to generate the third tracing information of the polysilicon silicon rod according to the event information of the polysilicon silicon rod in the transportation and packaging line circulation process, and make a pop-up window prompt in the inspection system according to the event information corresponding to the transportation and packaging of the polysilicon silicon rod according to the third tracing information;

[0175] The fourth generation module 240 is configured to generate and print the production information identification of the polysilicon silicon rod according to the first tracing information, the second tracing information and the third tracing information of the polysilicon silicon rod, wherein the production information identification is used to query the production information of the polysilicon silicon rod.

[0176] In one possible implementation, the second generation module 220 is configured to:

[0177] According to the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod generated under the same growth condition in history, the growth Pearson correlation coefficient between them is determined.

[0178] determine whether the growth of the polysilicon silicon rod is at a critical node according to a growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition;

[0179] if the growth of the polysilicon silicon rod is at the critical node, label the corresponding monitoring video with a critical node label;

[0180] generate second trace information according to the production information of the polysilicon silicon rod in the production process and the monitoring video corresponding to the critical node label.

[0181] In a possible implementation manner, the growth information includes a growth diameter and a growth length, and the second generation module 220 is configured to:

[0182] discretize the growth diameter and the growth length of the polysilicon silicon rod in the production process according to production time points to obtain growth diameters and growth lengths corresponding to different production time points;

[0183] a first growth Pearson correlation coefficient between the growth diameters of the polysilicon silicon rod at different production time points in the production process and the growth diameters of the polysilicon silicon rod historically generated under the same growth condition;

[0184] a second growth Pearson correlation coefficient between the growth lengths of the polysilicon silicon rod at different production time points in the production process and the growth lengths of the polysilicon silicon rod historically generated under the same growth condition.

[0185] In a possible implementation manner, the second generation module 220 is configured to:

[0186] obtain a first threshold set and a second threshold set set in advance, wherein the first threshold set and the second threshold set respectively include a plurality of first thresholds and a plurality of second thresholds;

[0187] compare the first growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition with the plurality of first thresholds in the first threshold set to obtain a first comparison result;

[0188] compare the second growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod historically generated under the same growth condition with the plurality of second thresholds in the second threshold set to obtain a second comparison result;

[0189] determine whether the growth of the polysilicon silicon rod is at a critical node according to the first comparison result and the second comparison result.

[0190] In a possible implementation, the third generation module 230 is configured to:

[0191] generate a production event according to event information of the polysilicon silicon rod in a production, transportation and packaging line flow process;

[0192] determine whether the production event of the polysilicon silicon rod is an abnormal event;

[0193] if the production event of the polysilicon silicon rod is an abnormal event, label monitoring video of the polysilicon silicon rod in the production, transportation and packaging process as abnormal;

[0194] generate third traceability information of the polysilicon silicon rod according to the abnormal label of the polysilicon silicon rod and the monitoring video labeled as the abnormal;

[0195] In a possible implementation, the third generation module 230 is configured to:

[0196] configure a historical operation curve model according to a temperature in a reduction furnace, a concentration of a feed gas, a material flow and an impurity content of the polysilicon silicon rod in a previous process, and predict a predicted impurity content of the polysilicon silicon rod in a subsequent process;

[0197] determine whether there is an impurity content abnormality of the polysilicon silicon rod in each of the processes according to the predicted impurity content of each of the processes and a corresponding impurity content threshold value set for each of the processes;

[0198] store real-time pictures, historical personnel operation records, quality information retrieval records and equipment state monitoring information of the polysilicon silicon rod in the transportation and packaging line flow process into an event information database;

[0199] identify the real-time pictures, the historical personnel operation records, the quality information retrieval records and the equipment state monitoring information, and determine whether there is an information abnormality;

[0200] determine whether the production event of the polysilicon silicon rod is an abnormal event according to a first result of whether there is an impurity content abnormality of the polysilicon silicon rod in each of the processes and a second result of whether there is an information abnormality.

[0201] In a possible implementation, the third generation module 230 is configured to train the historical operation curve model by the following method:

[0202] a regression equation I = β0 + β1T + β2C + β3*F + ε is constructed, wherein I is the impurity content, T is the temperature in the reduction furnace, C is the concentration of the feed gas, F is the material flow of the polysilicon silicon rod in the previous process, β0 is an intercept term, β1, β2, and β3 are regression coefficients, and ε is an error term;

[0203] According to the temperature in the reduction furnace, the concentration of the feed gas, the material flow of the polysilicon silicon rod in the previous process, and the impurity content corresponding to the polysilicon silicon rod produced historically, the intercept term β0, the regression coefficient β1, the regression coefficient β2, and the regression coefficient β3 of the regression equation are estimated using the least squares method.

[0204] After the intercept term β0, the regression coefficient β1, the regression coefficient β2, and the regression coefficient β3 of the regression equation are determined, the model parameters are adjusted through cross-validation to obtain the historical operation curve model.

[0205] The embodiments of the present disclosure further provide a computer readable storage medium having a computer program stored thereon, and the program is executed by a processor to implement the steps of the method in any one of the preceding embodiments.

[0206] The embodiments of the present disclosure further provide an electronic device, which comprises:

[0207] a memory having a computer program stored thereon;

[0208] a processor configured to execute the computer program in the memory to implement the steps of the method in any one of the preceding embodiments.

[0209] Figure 3 The polysilicon silicon rod production information tracing device 100 shown in the figure comprises a processor 1001 and a memory 1003. The processor 1001 and the memory 1003 are connected, for example, through a bus 1002. Optionally, the polysilicon silicon rod production information tracing device 100 can further comprise a communication component 1004, which can be used for data interaction, such as data sending and / or data receiving, between the device 100 and other devices. It should be noted that the communication component 1004 is not limited to one in actual scheduling, and the structure of the polysilicon silicon rod production information tracing device 100 does not constitute a limitation on the embodiments of the present disclosure.

[0210] The processor 1001 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, transistor logic device, hardware component or any combination thereof. It can implement or execute the various exemplary logical blocks, modules and circuits described in connection with the disclosure. The processor 1001 can also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0211] The bus 1002 can include a path for transmitting information between the above-mentioned components. The bus 1002 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 1002 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 3 Only one thick line is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0212] The memory 1003 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, an optical disk storage (including a compact disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium, other magnetic storage device, or any other medium capable of carrying or storing program code and capable of being read by a computer, without limitation.

[0213] The memory 1003 is configured to store program codes for implementing the embodiments of the present disclosure, and the processor 1001 is configured to execute the program codes stored in the memory 1003. The processor 1001 is configured to execute the program codes stored in the memory 1003 to implement the steps of the above-mentioned embodiments of the multi-crystalline silicon silicon rod production information tracing method.

[0214] The embodiments of the present disclosure further provide a computer readable storage medium, and the computer readable storage medium stores program codes. The program codes are executed by a processor to implement the steps and corresponding contents of the above-mentioned embodiments of the multi-crystalline silicon silicon rod production information tracing method.

[0215] The preferred embodiments of the present disclosure are described in detail above with reference to the drawings, but the present disclosure is not limited to the specific details of the above-described embodiments. Within the technical concept of the present disclosure, various changes, modifications, replacements and variations can be made to the embodiments, and all the changes, modifications, replacements and variations shall fall within the protection scope of the present disclosure.

[0216] In addition, it should be noted that each specific technical feature described in the above specific embodiments can be combined in any appropriate manner without contradiction, and shall be considered as the disclosed content of the present disclosure. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations. The technical scope of the present application shall be determined by the scope of the claims.

Claims

1. A method of tracing production information of a polysilicon silicon ingot, characterized by, The method comprises: Importing basic information of silicon powder for producing the polysilicon silicon rod into a label code generation system as a data field, and generating first traceability information according to the data field by the label code generation system based on a pre-set query condition; Determining second traceability information of the polysilicon silicon rod according to growth information of the polysilicon silicon rod in a production process and growth information of the polysilicon silicon rod generated under the same growth condition in history; Generating third traceability information of the polysilicon silicon rod according to event information of the polysilicon silicon rod in a production, transportation and packaging line flow process, and making a pop-up window prompt of event information corresponding to transportation and packaging of the polysilicon silicon rod in an inspection system according to the third traceability information; Generating and printing production information identification of the polysilicon silicon rod according to the first traceability information, the second traceability information and the third traceability information of the polysilicon silicon rod, wherein the production information identification is used for querying production information of the polysilicon silicon rod.

2. The polysilicon silicon ingot production information tracing method according to claim 1, characterized by, The determining of the second traceability information of the polysilicon silicon rod according to the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod generated under the same growth condition in history comprises: A growth Pearson correlation coefficient between the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod generated under the same growth condition in history; Determining whether the growth of the polysilicon silicon rod is at a key node according to the growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod generated under the same growth condition in history; If the growth of the polysilicon silicon rod is at the key node, a key node label is added to corresponding monitoring video; Generating the second traceability information according to the production information of the polysilicon silicon rod in the production process and the monitoring video corresponding to the key node label.

3. The polycrystalline silicon silicon rod production information tracing method according to claim 2, characterized by, The growth information comprises a growth diameter and a growth length, and the growth Pearson correlation coefficient between the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod generated under the same growth condition in history comprises: Discretizing the growth diameter and the growth length of the polysilicon silicon rod in the production process according to production time points to obtain growth diameters and growth lengths corresponding to different production time points; A first growth Pearson correlation coefficient between the growth diameters of the polysilicon silicon rod at different production time points in the production process and the growth diameters of the polysilicon silicon rod generated under the same growth condition in history; A second growth Pearson correlation coefficient between the growth lengths of the polysilicon silicon rod at different production time points in the production process and the growth lengths of the polysilicon silicon rod generated under the same growth condition in history.

4. The polycrystalline silicon silicon rod production information tracing method according to claim 3, characterized by, The method comprises the following steps: obtaining a first threshold set and a second threshold set, wherein the first threshold set and the second threshold set respectively include a plurality of first thresholds and a plurality of second thresholds; comparing the first growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod generated under the same growth condition in history with the plurality of first thresholds in the first threshold set to obtain a first comparison result; comparing the second growth Pearson correlation coefficient between the polysilicon silicon rod and the polysilicon silicon rod generated under the same growth condition in history with the plurality of second thresholds in the second threshold set to obtain a second comparison result; determining whether the growth of the polysilicon silicon rod is at a critical node according to the first comparison result and the second comparison result.

5. The polysilicon silicon ingot production information tracing method according to claim 1, wherein The method comprises the following steps: generating a third trace information of the polysilicon silicon rod according to event information of the polysilicon silicon rod in the production, transportation and packaging line flow process, comprising: generating a production event according to the event information of the polysilicon silicon rod in the production, transportation and packaging line flow process; determining whether the production event of the polysilicon silicon rod is an abnormal event; if the production event of the polysilicon silicon rod is an abnormal event, labeling the monitoring video of the polysilicon silicon rod in the production, transportation and packaging process with an abnormal label; 6. The polysilicon silicon ingot production information tracing method according to claim 5, wherein generating the third trace information of the polysilicon silicon rod according to the abnormal label of the polysilicon silicon rod and the monitoring video labeled with the abnormal label. The method comprises the following steps: configuring a historical operation curve model according to the temperature in the reduction furnace, the concentration of the feed gas, the material flow and impurity content of the polysilicon silicon rod in the previous process to predict the predicted impurity content of the polysilicon silicon rod in the subsequent process; determining whether there is an impurity content abnormality of the polysilicon silicon rod in each process according to the predicted impurity content of each process and the corresponding impurity content threshold set for each process; storing real-time pictures, historical personnel operation records, quality information retrieval records and equipment state monitoring information of the polysilicon silicon rod in the transportation and packaging line flow process into an event information database; identifying the real-time pictures, historical personnel operation records, quality information retrieval records and equipment state monitoring information to determine whether there is an information abnormality; 7. The polysilicon silicon ingot production information tracing method according to claim 6, characterized by, determining whether the production event of the polysilicon silicon rod is an abnormal event according to the first result of whether there is an impurity content abnormality of the polysilicon silicon rod in each process and the second result of whether there is an information abnormality. The historical operation curve model is obtained by training in the following way: constructing a regression equation: I=β0+β1T+β2C+β3*F+ε, wherein I is the impurity content, T is the temperature in the reduction furnace, C is the concentration of the feed gas, F is the material flow of the polysilicon silicon rod in the previous process, β0 is the intercept term, β1, β2, β3 are regression coefficients, and ε is the error term; According to the temperature in the reduction furnace corresponding to the polysilicon silicon rod produced historically, the concentration of the feed gas, the material flow of the polysilicon silicon rod in the previous process and the impurity content, the intercept term β0, the regression coefficient β1, the regression coefficient β2 and the regression coefficient β3 of the regression equation are estimated using the least squares method; After determining the intercept term β0, the regression coefficient β1, the regression coefficient β2 and the regression coefficient β3 of the regression equation, the model parameters are adjusted by cross-validation to obtain the historical operation curve model.

8. A polycrystalline silicon rod production information traceability device, characterized in that, The device comprises: A first generation module configured to import the basic information of the silicon powder for producing the polysilicon silicon rod as a data field into a label code generation system, and generate first traceability information based on the data field and a pre-set query condition through the label code generation system; A second generation module configured to determine second traceability information of the polysilicon silicon rod according to the growth information of the polysilicon silicon rod in the production process and the growth information of the polysilicon silicon rod produced historically under the same growth conditions; A third generation module configured to generate third traceability information of the polysilicon silicon rod according to event information in the transportation and packaging line circulation process of the polysilicon silicon rod, and make a pop-up window prompt in an inspection system for the event information corresponding to the transportation and packaging of the polysilicon silicon rod according to the third traceability information; A fourth generation module configured to generate and print production information identification of the polysilicon silicon rod according to the first traceability information, the second traceability information and the third traceability information of the polysilicon silicon rod, wherein the production information identification is used to query the production information of the polysilicon silicon rod.

9. An electronic device, comprising: It comprises: A memory having a computer program stored thereon; A processor for executing the computer program in the memory to implement the steps of the method of any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps of the method of any one of claims 1-7.

Images