Polycrystalline silicon production control method and device, electronic equipment and storage medium

Abstract

The invention provides a polycrystalline silicon production control method and device, electronic equipment and a storage medium, and relates to the technical field of polycrystalline silicon production. The method comprises the following steps: establishing a polycrystalline silicon production operation database; furnace batch production scoring is carried out according to the polycrystalline silicon production operation database to determine an optimal furnace batch set, and a production formula is determined according to the optimal furnace batch set; and controlling the production process according to the polycrystalline silicon production operation database and the production formula. According to the method, the polycrystalline silicon production operation database can be established based on the historical operation data of polycrystalline silicon production, then furnace batch production scoring is performed according to the polycrystalline silicon production operation database to determine the optimal furnace batch set, and then the production process curve can be formed through secondary fitting according to the optimal furnace batch set to determine the production formula. And finally, the production process is controlled in combination with a polycrystalline silicon production operation database and a production formula, efficient reproducibility of the production process is assisted to be achieved, and production deviation caused by experience deviation is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of polysilicon production, and in particular to a polysilicon production control method, device, electronic equipment and storage medium. Background Art

[0002] The polysilicon production process is a relatively complex chemical reduction reaction process, and its overall production time (single furnace) is approximately 90-110 hours. During the production process, it is greatly affected by related factors such as furnace conditions, environment, and the overall exhaust gas recovery system. Therefore, the entire current production process not only relies on the predetermined given stage input parameter quantities, but also highly relies on process engineers to make further compensation and revisions based on feedback from actual conditions.

[0003] Due to the large number of relevant process indicators involved in the entire process production (generally more than 30 parameters), long reaction process and adjustment lag, the overall process adjustment process is currently highly dependent on the process experience of relevant personnel and lacks stable reproduction capabilities. Summary of the invention

[0004] The problem solved by the present invention is how to realize a stable polysilicon production process.

[0005] In order to solve the above problems, the present invention provides a polysilicon production control method, device, electronic equipment and storage medium.

[0006] In a first aspect, the present invention provides a polysilicon production control method, comprising:

[0007] Establish a polysilicon production operation database;

[0008] Performing furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set, and determining a production formula according to the optimal furnace batch set;

[0009] The production process is controlled according to the polysilicon production operation database and the production recipe.

[0010] Optionally, the establishing of a polysilicon production operation database includes:

[0011] Acquire historical operation data of the polysilicon production process, wherein the historical operation data includes furnace batch out-of-furnace result data, production process operation data, abnormal records and production process inspection records;

[0012] An indexing rule for the historical operation data is established according to a preset dimension, so as to establish the polysilicon production operation database at different dimensional levels.

[0013] Optionally, the establishing of index rules for the historical operation data according to preset dimensions to establish the polysilicon production operation database at different dimensional levels includes:

[0014] The production batches are classified according to at least one of the furnace type dimension, the production process recipe version dimension, the furnace number dimension, and the production response time range dimension to establish the index rule.

[0015] Optionally, performing furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set includes:

[0016] Processing the polysilicon production operation database to obtain reduction furnace discharge data in the same dimension;

[0017] Scoring the reduction furnace discharge data using different sub-item scoring factors to obtain corresponding multiple scoring results;

[0018] The optimal furnace batch set under different division conditions is determined according to the multiple scoring results.

[0019] Optionally, the scoring of the reduction furnace discharge data by different sub-item scoring factors includes:

[0020] Corresponding scoring methods and scoring method parameters are used for the different sub-item scoring factors to achieve scoring.

[0021] Optionally, determining the production formula according to the optimal furnace batch set includes:

[0022] Generate a production process curve according to the optimal furnace batch set;

[0023] The production formula is determined according to the production process curve.

[0024] Optionally, controlling the production process according to the polysilicon production operation database and the production formula includes:

[0025] Dynamically score the current furnace batch operation status based on historical furnace batches to determine the dynamic scoring results;

[0026] Performing a production deviation prediction on the current furnace batch operation status according to the production formula to determine a deviation prediction result;

[0027] Performing anomaly similarity judgment based on the dynamic scoring result and the deviation prediction result, and determining the similarity between the current furnace batch and the historical furnace batch based on the anomaly records in the polysilicon production operation database;

[0028] An operation suggestion is generated according to the similarity to control the production process.

[0029] In a second aspect, the present invention provides a polysilicon production control device, comprising:

[0030] The first module is used to establish a polysilicon production operation database;

[0031] The second module is used to perform furnace batch production scoring according to the polysilicon production operation database to determine the optimal furnace batch set, and determine the production formula according to the optimal furnace batch set;

[0032] The third module is used to control the production process according to the polysilicon production operation database and the production formula.

[0033] In a third aspect, the present invention provides an electronic device, including a memory and a processor;

[0034] The memory is used to store computer programs;

[0035] The processor is used to implement the polysilicon production control method as described in the first aspect when executing the computer program.

[0036] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the polysilicon production control method as described in the first aspect is implemented.

[0037] The beneficial effects of the polysilicon production control method of the present invention are as follows: first, a polysilicon production operation database can be established based on the historical operation data of polysilicon production, and then furnace batch production scoring can be performed based on the polysilicon production operation database to determine the optimal furnace batch set, and then the production process curve can be formed by secondary fitting based on the optimal furnace batch set to determine the production formula, and finally the polysilicon production operation database and the production formula are combined to control the production process, for example, the real-time production status is compared with the historical operation data to obtain a scoring result, etc., to form a guiding result for the production process, to assist in achieving efficient reproducibility of the production process, and to reduce production deviations caused by experience deviations. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram of a process flow of a polysilicon production control method according to an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a process for establishing a polysilicon production operation database according to an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of a process for determining an optimal furnace batch set according to an embodiment of the present invention;

[0041] Figure 4 A schematic diagram of a process for determining a production formula according to an embodiment of the present invention;

[0042] Figure 5 A schematic diagram of a process flow of controlling a production process according to an embodiment of the present invention;

[0043] Figure 6 A system architecture diagram of a polysilicon production control device according to an embodiment of the present invention;

[0044] Figure 7 4 is a system architecture diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0046] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0047] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0048] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0049] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0050] like Figure 1As shown, a polysilicon production control method provided by an embodiment of the present invention includes:

[0051] S100: Establish a polysilicon production operation database.

[0052] Specifically, firstly, a polysilicon production operation database is established, which may include furnace batch result data, production process operation data, abnormal records, production process inspection records and other historical operation data.

[0053] S200: Performing furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set, and determining a production formula according to the optimal furnace batch set.

[0054] Specifically, batch production scoring is performed based on the polysilicon production operation database to determine the optimal batch set (or golden batch set). For example, relevant batches are scored using a scoring model, and the optimal batch set under different scoring conditions is output. The production formula is then determined based on the optimal batch set. For example, a production process curve is generated based on the optimal batch set, and then the production formula is determined.

[0055] Among them, polysilicon is produced in a single furnace. Each furnace production process needs to go through the process of installation-reaction production-cooling-disassembly-warehousing. The whole process is called a furnace batch; usually there are multiple reduction furnaces for production, and the production start time and end time of each reduction furnace are inconsistent.

[0056] S300: Controlling the production process according to the polysilicon production operation database and the production recipe.

[0057] Specifically, the production process is controlled according to the polysilicon production operation database and production formula to achieve guidance and control of the production process.

[0058] In this embodiment, first, a polysilicon production operation database can be established based on the historical operation data of polysilicon production, and then the furnace batch production is scored according to the polysilicon production operation database to determine the optimal furnace batch set. Then, the production process curve can be formed by secondary fitting based on the optimal furnace batch set to determine the production formula. Finally, the polysilicon production operation database and the production formula are combined to control the production process. For example, the real-time production status is compared with the historical operation data to obtain a scoring result, etc., to form a guiding result for the production process, to help achieve efficient reproducibility of the production process, and to reduce production deviations caused by experience bias.

[0059] Optionally, the establishing of a polysilicon production operation database includes:

[0060] S110: Acquire historical operation data of the polysilicon production process, wherein the historical operation data includes furnace batch out-of-furnace result data, production process operation data, abnormal records, and production process inspection records;

[0061] Specifically, combined Figure 2 As shown, the historical operation data includes the furnace batch result data (such as output, quality, cost, etc.), relevant operation data of the production process (such as H2, TCS input flow, pressure, exhaust gas outlet temperature, etc.), abnormal alarm records of the control system, operation records, inspection records of the reduction furnace production process and other additional record information related to the reduction furnace.

[0062] S120: Establishing index rules for the historical operation data according to preset dimensions, so as to establish the polysilicon production operation database at different dimensional levels.

[0063] Specifically, combined Figure 2 As shown, by indexing data based on multiple dimensions such as the characteristics of the reduction furnace (such as furnace type, production commissioning time, production process relevance, etc.) and the production operation time corresponding to the reduction furnace batch, a polysilicon production operation database is established at different dimensional levels.

[0064] In this optional embodiment, indexing rules for historical operation data are established according to preset dimensions, so that a polysilicon production operation database is established at different dimensional levels, which is conducive to achieving a stable polysilicon process production process.

[0065] Optionally, the establishing of index rules for the historical operation data according to preset dimensions to establish the polysilicon production operation database at different dimensional levels includes:

[0066] The production batches are classified according to at least one of the furnace type dimension, the production process recipe version dimension, the furnace number dimension, and the production response time range dimension to establish the index rule.

[0067] Specifically, different dimensional levels include levels similar to the following: 1. Classifying production batches based on furnace type; 2. Classifying production batches based on production process formula version; 3. Classifying production batches based on furnace number (i.e. production equipment); 4. Classifying production batches based on production reaction time range (the reaction time of the polysilicon production process is not constant, and may be divided into 90-95 hours, 100-105 hours, 110 hours and similar situations according to actual process and production conditions). In other words, numerous batches can be divided according to production conditions or production results.

[0068] In this optional embodiment, the production batches are classified according to different dimensional levels to establish a polysilicon production operation database, which is conducive to achieving a stable polysilicon process production process.

[0069] Optionally, performing furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set includes:

[0070] S210: Processing the polysilicon production operation database to obtain reduction furnace output data in the same dimension.

[0071] Specifically, combined Figure 3 As shown, firstly, the polysilicon production operation database is cleaned and uniformly converted, and the relevant operation data, working conditions, etc. can be further compared and analyzed under the dimension of the production operation process (running time).

[0072] S220: Scoring the reduction furnace discharge data using different sub-item scoring factors to obtain corresponding multiple scoring results.

[0073] Specifically, combined Figure 3 As shown, if the evaluation criteria for the quality of production cannot be unified, it is impossible to effectively classify and analyze the production. Therefore, this embodiment provides a set of freely configurable and easy-to-use scoring standard models as the basis for further production analysis. Since the polysilicon production process is subject to market conditions, it is affected by three major factors such as output, quality, and cost. Under different conditions, it will directly affect the judgment of the quality of the furnace results. Therefore, in addition to having custom configurations, the scoring model also needs to allow the ability to retain multiple scoring standards at the same time. The reduction furnace out-of-the-furnace data calculates the score of each sub-item factor under the scoring strategy through the scoring strategy, and then weighted calculations are performed to form the score of the furnace out-of-the-furnace result; the furnace out-of-the-furnace result score includes the corresponding sub-item scoring factor score in addition to the overall score. By defining different numbers of scoring factors in the scoring strategy, the furnace out-of-the-furnace result score can be obtained under different scoring strategies.

[0074] S230: Determine the optimal batch set under different classification conditions according to the multiple scoring results.

[0075] Specifically, combined Figure 3 As shown, the golden batch set is to score the relevant furnace batches with a scoring model under given conditions, and output the optimal furnace batch set under different divisions, where different divisions refer to free division according to furnace number, furnace type, production date range, etc., to ensure that the golden batch furnace batches are under a given approximate operating condition (for example, under the same furnace type, the optimal furnace batch produced in the fourth quarter of 2023), and the approximate operating conditions of the production process are controlled by shrinkage, thereby improving the correlation between batches and facilitating further analysis and fitting.

[0076] In this optional embodiment, different sub-item scoring factors are used to determine the optimal furnace batch set under different classification conditions, thereby ensuring the accuracy of the furnace result scoring.

[0077] Optionally, the scoring of the reduction furnace discharge data by different sub-item scoring factors includes:

[0078] Corresponding scoring methods and scoring method parameters are used for the different sub-item scoring factors to achieve scoring.

[0079] Specifically, the scoring factors include: single furnace output, primary conversion rate, growth rate, whether the furnace is turned over, whether it is fish scales, coral material ratio, dense material ratio, electricity consumption and TCS consumption, etc. Taking single furnace output as an example, the scoring method adopts interval adaptation [positive], for example, the interval is [8000-13000], and the specific scoring content is: interval value: (actual value-lower limit value) / (upper limit value-lower limit value)*weight, exceeding the upper limit: full score + (actual value-upper limit value) / (upper limit value-lower limit value)*additional coefficient; for interval adaptation [positive], the higher the value, the better. When the value is in a certain interval, it meets the requirements. The quality is determined based on the deviation from the lower limit of the interval. When the deviation is lower than the lower limit of the interval, points will be deducted.

[0080] In this optional embodiment, the accuracy of the scoring results is ensured by using corresponding scoring methods and scoring method parameters for different sub-item scoring factors to achieve scoring.

[0081] Optionally, determining the production formula according to the optimal furnace batch set includes:

[0082] S240: Generate a production process curve according to the optimal furnace batch set.

[0083] S250: Determine the production formula according to the production process curve.

[0084] Specifically, combined Figure 4 As shown in the figure, the production formula is the initially given key production control parameter table. By fitting, the approximation of its production curve under the same scoring condition can be explored. By comparing the deviation with its original input formula table, the impact of deviation and correction can be intuitively provided to process personnel, which is convenient for process personnel to adjust and optimize the material list.

[0085] Among them, the manipulated variables include TCS (Total Control Systems) flow, H2 flow and current, the controlled variables include voltage, power, exhaust gas temperature, bell water flow, chassis water flow, bell water return temperature, chassis water return temperature and exhaust gas pressure, and the simulated variables include TCS flow, H2 flow, current, voltage, power, exhaust gas temperature, bell water flow and chassis water flow.

[0086] In this optional embodiment, a production process curve is generated based on the optimal furnace batch set to determine the production formula, and the process personnel are intuitively provided with the impact of deviations and corrections, which facilitates the process personnel to adjust and optimize the material list.

[0087] Optionally, controlling the production process according to the polysilicon production operation database and the production formula includes:

[0088] S310: Dynamically score the current furnace batch operation status based on the historical furnace batches to determine the dynamic scoring result.

[0089] Specifically, combined Figure 5 As shown in the figure, during the real-time operation of the reduction furnace, the historical furnace batches with similar operation conditions to the current furnace batch are searched in real time, and the operation conditions of the current furnace batch are judged based on the historical furnace batches. Among them, the similarity of two furnace batches means that the silicon rod diameters are basically equal and the furnace conditions are basically the same under the same operation time.

[0090] The curve similarity of the operating curves of each influencing factor of the two furnace batches is calculated, and the similarity of each curve is weighted to obtain the similarity between the two furnace batches, and the weight is the weight obtained by the influencing factor analysis.

[0091] The curve similarity calculation method is: calculate the determination coefficient R of the two curves 2 , coefficient of determination R 2 Between 0 and 100%, the coefficient R 2 as the similarity of the two curves.

[0092] Among them, the coefficient of determination R 2 It is expressed as:

[0093]

[0094] Among them, RSS is the residual sum of squares, that is, the sum of squares of the difference between current data and historical data, and TSS is the variance of historical operating data.

[0095] Assuming there are M historical batches, the score of the current batch is the weighted average of the scores of the top 5% batches in similarity (assuming there are m batches) expressed as:

[0096] P=α1p1+α2p2+…+α m p m ;

[0097] Considering that the more distant the historical batch is from the current time, the weaker its guiding significance for the current production is, the weight α i Time weight T needs to be added i ,Right now:

[0098]

[0099] Among them, Δt i Indicates the time interval from a historical batch to the present, with Δt i As increases, the weight decreases linearly until the weight of the last batch is zero.

[0100] S320: Predicting the production deviation of the current furnace batch operation status according to the production formula to determine the deviation prediction result.

[0101] Specifically, combined Figure 5 As shown, the production deviation prediction model can be expressed as:

[0102] y i，t：t+Δt =f(y i，t：t-k , x i，i-k：t+Δt , S i );

[0103] The input includes: the observed value y of the controlled variable of the current batch i，t:t-k and the observed value of the manipulated variable x i，t-k:t+Δt And static parameter = S i , such as the weight of each influencing factor, furnace number and team information, etc.

[0104] The output includes: the predicted data y of the controlled variable in the next period of time Δt i，t:t+Δt .

[0105] The training set includes all batch historical operation data, such as manipulated variable operation data, controlled variable operation data, warning records, and operation records.

[0106] Among them, the prediction model adopts: (1) time series prediction algorithm; (2) statistical algorithm: multivariate linear regression algorithm, advantages: clear physical meaning, fast calculation speed, suitable for stable linear time series; (3) deep learning algorithm: convolutional neural network, transformer (attention mechanism), advantages: able to handle multi-parameter and multi-modal input, large amount of calculation, the larger the training set, the more accurate it is.

[0107] Among them, the objective function includes the standard deviation of the controlled variables such as voltage, power, exhaust temperature, bell water flow and chassis water flow and the predicted value. The smaller the objective function, the higher the model prediction accuracy.

[0108] By predicting the input process indicators, the difference between the predicted results and the input material list at the same operating time is the actual deviation. According to the deviation, further retrieval can be carried out in combination with historical data to determine how to make corrections.

[0109] S330: Perform an abnormal similarity judgment based on the dynamic scoring result and the deviation prediction result, and determine the similarity between the current furnace batch and the historical furnace batch based on the abnormal records in the polysilicon production operation database.

[0110] Specifically, combined Figure 5 As shown, combined with the corresponding operation records in the historical batch database, the currently predicted deviation working condition is judged on the basis of the combination factor coefficient k (the influence of relevant indicators in the production process on the actual process and results, which needs to be obtained through repeated iterations and corrections in combination with the actual production operation conditions and production experience on site), and the corresponding approximate operation record set is obtained by combining the similarity and the target result (input material list). The operation recommendation value can be obtained by classification and division based on the approximate operation set. By reviewing the operation recommendation value and comparing the actual operation value, after accumulating a sufficient amount of operation records and comparison records, the operation recommendation of the corresponding working condition can be easily converged to a given interval by clustering. The given interval is the operation result generated by the accumulation of actual experience.

[0111] In addition to deviation identification, pattern recognition can also be introduced to pre-identify abnormal working conditions, that is, adjustments in the actual production process are all for abnormal working conditions. The time parameter t can be given: [abnormal identification time period] = [operation start time-t, operation start time], and the operation record time can be divided into operation sets. The corresponding abnormal sample set can be formed by manually screening and merging the operation set and entering it. The abnormal sample set can judge the abnormal similarity based on pattern recognition and introduce deviation identification.

[0112] Among them, if the prediction of working condition A deteriorates and deteriorates to working condition A1, the deviation is ΔA; then according to this benchmark, the current working condition A, the deteriorated working condition A1 and the deviation ΔA are compared with the anomalies in the historical batch records, that is, it can be determined from the historical anomaly records which one is similar to.

[0113] S340: Generate operation suggestions according to the similarity to control the production process.

[0114] Specifically, combined Figure 5 As shown, operation suggestions are generated based on the similarity to control the production process, and the operation records are archived in the historical batch database.

[0115] Among them, since the production process is the production process of polysilicon ingots, relevant raw materials and temperature need to be controlled throughout the process. The so-called input recipe parameters are the relevant indicators to be achieved at the corresponding running time, and the stepping feed amount is controlled by the control system in the way of the change amount of the input value. The input recipe parameters can be in the form of a table like <running time>-<H2 flow value><TCS flow value><current value>. Since the production operation parameters cannot be directly modified when abnormal conditions occur during the production process (for example, the current flow rate of 1400 cannot be directly adjusted to 1100), the given correction value is the compensation for the input recipe parameters, such as -200. The system accepts this compensation value and calculates the target value to be controlled by combining the input recipe parameters and the running time.

[0116] In this alternative embodiment, the production process is controlled according to the polysilicon production operation database and the production recipe, so as to realize the guidance and control of the production process.

[0117] As Figure 6 shown, a polysilicon production control device 600 provided by an embodiment of the present invention includes:

[0118] A first module 610, configured to establish a polysilicon production operation database;

[0119] A second module 620, configured to perform furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set, and determine a production recipe according to the optimal furnace batch set;

[0120] A third module 630, configured to control the production process according to the polysilicon production operation database and the production recipe.

[0121] As Figure 7 shown, an electronic device 700 provided by an embodiment of the present invention includes a memory 720 and a processor 710; the memory 720 is used to store a computer program; the processor 710 is used to implement the polysilicon production control method as described above when executing the computer program.

[0122] Or, an electronic device 700 includes a memory 720 and a processor 710 coupled to the memory 720; the memory 720 is configured to store a computer program; the processor 710 is configured to perform the following operations when executing the computer program:

[0123] Establish a polysilicon production operation database;

[0124] Perform furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set, and determine a production recipe according to the optimal furnace batch set;

[0125] The production process is controlled according to the polysilicon production operation database and the production recipe.

[0126] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the polysilicon production control method as described above is implemented.

[0127] In other words, a non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following operations:

[0128] Establish a polysilicon production operation database;

[0129] Performing furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set, and determining a production formula according to the optimal furnace batch set;

[0130] The production process is controlled according to the polysilicon production operation database and the production recipe.

[0131] An electronic device 700 that can be used as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 700 is intended to represent various forms of digital electronic computer equipment, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 700 can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present invention described herein and / or required.

[0132] The electronic device 700 includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The computing unit, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0133] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present invention. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0134] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A polysilicon production control method, characterized in that: include: Establish a polysilicon production operation database; Performing furnace batch production scoring according to the polysilicon production operation database to determine an optimal furnace batch set, and determining a production formula according to the optimal furnace batch set; The production process is controlled according to the polysilicon production operation database and the production recipe.

2. The polysilicon production control method according to claim 1, characterized in that: The establishment of a polysilicon production operation database comprises: Acquire historical operation data of the polysilicon production process, wherein the historical operation data includes furnace batch out-of-furnace result data, production process operation data, abnormal records and production process inspection records; An indexing rule for the historical operation data is established according to a preset dimension, so as to establish the polysilicon production operation database at different dimensional levels.

3. The polysilicon production control method according to claim 2, characterized in that: The step of establishing the indexing rules of the historical operation data according to the preset dimensions to establish the polysilicon production operation database at different dimensional levels includes: The production batches are classified according to at least one of the furnace type dimension, the production process recipe version dimension, the furnace number dimension, and the production response time range dimension to establish the index rule.

4. The polysilicon production control method according to claim 1, characterized in that: The performing furnace batch production scoring according to the polysilicon production operation database to determine the optimal furnace batch set comprises: Processing the polysilicon production operation database to obtain reduction furnace discharge data in the same dimension; Scoring the reduction furnace discharge data using different sub-item scoring factors to obtain corresponding multiple scoring results; The optimal furnace batch set under different division conditions is determined according to the multiple scoring results.

5. The polysilicon production control method according to claim 4, characterized in that: The scoring of the reduction furnace discharge data by different sub-item scoring factors includes: Corresponding scoring methods and scoring method parameters are used for the different sub-item scoring factors to achieve scoring.

6. The polysilicon production control method according to claim 1, characterized in that: Determining the production formula according to the optimal furnace batch set includes: Generate a production process curve according to the optimal furnace batch set; The production formula is determined according to the production process curve.

7. The polysilicon production control method according to claim 6, characterized in that: The controlling of the production process according to the polysilicon production operation database and the production formula comprises: Dynamically score the current furnace batch operation status based on historical furnace batches to determine the dynamic scoring results; Performing a production deviation prediction on the current furnace batch operation status according to the production formula to determine a deviation prediction result; Performing anomaly similarity judgment based on the dynamic scoring result and the deviation prediction result, and determining the similarity between the current furnace batch and the historical furnace batch based on the anomaly records in the polysilicon production operation database; An operation suggestion is generated according to the similarity to control the production process.

8. A polysilicon production control device, characterized in that: include: The first module is used to establish a polysilicon production operation database; The second module is used to perform furnace batch production scoring according to the polysilicon production operation database to determine the optimal furnace batch set, and determine the production formula according to the optimal furnace batch set; The third module is used to control the production process according to the polysilicon production operation database and the production formula.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the polysilicon production control method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the polysilicon production control method according to any one of claims 1 to 7 is implemented.

Images