Training of data computing analysis model and start-stop control method of polycrystalline silicon reduction furnace

Abstract

The application provides a training method of a data calculation analysis model and a start-stop control method of a polycrystalline silicon reduction furnace, and the start-stop control method of the polycrystalline silicon reduction furnace comprises the following steps: acquiring the running parameters of the polycrystalline silicon reduction furnace in a polycrystalline silicon reduction furnace system in real time; inputting the running parameters into a data calculation analysis model to obtain a predicted current curve constituted by the running current values of the polycrystalline silicon reduction furnace; acquiring the current threshold of the power consumption of the polycrystalline silicon reduction furnace system; predicting the current curve of the current value and the total current value of the polycrystalline silicon reduction furnace according to the current threshold and the current curve; and obtaining the predicted running quantity of the polycrystalline silicon reduction furnace, the start time and the new predicted current curve of the total current value of the polycrystalline silicon reduction furnace system according to the predicted current value and the current curve. In the application, the prediction curve is generated by analyzing and calculating the total data of the reduction furnace, and the dynamic adjustment and optimization of the curve are realized along with the increase of the historical data, and the start-stop operation planning of the reduction furnace is carried out according to the above.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of polycrystalline silicon production, in particular to a training method of a data calculation and analysis model and a start-stop control method of a polycrystalline silicon reduction furnace. BACKGROUND

[0002] In the polycrystalline silicon production process, the control of the polycrystalline silicon reduction furnace and the reasonable production scheduling are particularly important. In the prior art, the mature application of the industrial internet platform is relied on for description, and the scheme design and description are not combined with the actual business requirements of the production process and the big data analysis application cannot solve the problem of guiding the specific operation of the start-stop of the reduction furnace in the polycrystalline silicon production process. SUMMARY

[0003] The embodiment of the present application provides a training method of a data calculation and analysis model and a start-stop control method of a polycrystalline silicon reduction furnace to solve the problem that the existing big data analysis application cannot solve the problem of guiding the specific operation of the start-stop of the reduction furnace in the polycrystalline silicon production process.

[0004] In order to solve the above technical problems, the present application is implemented as follows:

[0005] In a first aspect, the embodiment of the present application provides a training method of a data calculation and analysis model, comprising:

[0006] Obtaining sample running parameters and a sample current curve of each polycrystalline silicon reduction furnace in a polycrystalline silicon reduction furnace system, wherein the polycrystalline silicon reduction furnace system comprises at least one polycrystalline silicon reduction furnace production line, each polycrystalline silicon reduction furnace production line comprises at least one polycrystalline silicon reduction furnace, and the sample current curve is a current curve true value composed of running current values of the polycrystalline silicon reduction furnace in a polycrystalline silicon growth period;

[0007] Data preprocessing is performed on the sample running parameters to obtain processed sample running parameters that are non-abnormal and meet the storage structure;

[0008] The processed sample running parameters are input into a data calculation and analysis model to be trained for calculation and analysis, and a predicted sample current curve of the polycrystalline silicon reduction furnace in the polycrystalline silicon growth period is output;

[0009] The parameters of the data calculation and analysis model to be trained are adjusted according to the sample current curve and the predicted sample current curve, and a trained data calculation and analysis model is obtained.

[0010] Optionally, the sample running parameters are historical running parameters of the polycrystalline silicon reduction furnace system in the polycrystalline silicon growth period, and comprise at least one of the following: the number of polycrystalline silicon reduction furnaces, the running current value of each polycrystalline silicon reduction furnace, the running time, the material in and out, and the running power.

[0011] Optionally, the processed sample operation parameter is input into the data calculation and analysis model to be trained for calculation and analysis, and a predicted sample current curve of the polysilicon reduction furnace in a polysilicon growth period is output.

[0012] The sample current curve and the predicted sample current curve are input into a mechanism learning model, and the sample current curve and the predicted sample current curve are stored in a database, the mechanism learning model adjusts parameters of the data calculation and analysis model to be trained through data analysis and learning, and a trained data calculation and analysis model is obtained.

[0013] Optionally, the data preprocessing is performed on the sample operation parameter to obtain a processed sample operation parameter that is non-abnormal and meets a database structure.

[0014] The sample operation parameter is adjusted according to a storage structure meeting the database to obtain an adjusted sample operation parameter, and the adjusted sample operation parameter is stored in the database.

[0015] The adjusted sample operation parameter is calculated according to the running current value, the running time, the material in and out, and the running power of each polysilicon reduction furnace to obtain a calculated sample operation parameter, and the calculated sample operation parameter is filtered to obtain a processed sample operation parameter that is non-abnormal and meets a database structure.

[0016] In a second aspect, an embodiment of the present application provides a start-stop control method of a polysilicon reduction furnace, including:

[0017] Real-time acquisition of operation parameters of a polysilicon reduction furnace in a polysilicon reduction furnace system, the polysilicon reduction furnace system including at least one polysilicon reduction furnace production line, each polysilicon reduction furnace production line including at least one polysilicon reduction furnace;

[0018] Data preprocessing is performed on the operation parameters to obtain processed operation parameters that are non-abnormal and meet a storage structure.

[0019] The processed operation parameters are input into a trained data calculation and analysis model for calculation and analysis, and a predicted current curve composed of running current values of the polysilicon reduction furnace in a polysilicon growth period is obtained.

[0020] Acquisition of a current threshold of power consumption of the polysilicon reduction furnace system;

[0021] According to the current threshold and the current curve, a current curve of a current value of each polysilicon reduction furnace and a total current value of the polysilicon reduction furnace system in a future preset time period is predicted.

[0022] According to the predicted current value of each of the polysilicon reduction furnaces and the total current value of the polysilicon reduction furnace system, a predicted number of operations of the polysilicon reduction furnaces, an opening time of each of the polysilicon reduction furnaces, and a new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of operations of the polysilicon reduction furnaces and the opening time of each of the polysilicon reduction furnaces are obtained.

[0023] Optionally, the method further comprises:

[0024] The predicted number of operations of the polysilicon reduction furnaces, the opening time of each of the polysilicon reduction furnaces, and the new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of operations of the polysilicon reduction furnaces and the opening time of each of the polysilicon reduction furnaces are dynamically displayed in real time.

[0025] Optionally, the operation parameters of the polysilicon reduction furnace system include at least one of the following: a number of operations of the polysilicon reduction furnaces, an operation current value of each of the polysilicon reduction furnaces, an operation time, an in-out material, and an operation power.

[0026] Optionally, the real-time acquisition of the operation parameters of the polysilicon reduction furnaces in the polysilicon reduction furnace system comprises:

[0027] The operation parameters of the polysilicon reduction furnace system are acquired in real time from a distributed control system and a data acquisition and monitoring control system.

[0028] In a third aspect, an embodiment of the present application provides a training device of a data calculation and analysis model, comprising:

[0029] A first acquisition module is configured to acquire sample operation parameters and a sample current curve of each of polysilicon reduction furnaces in a polysilicon reduction furnace system, the polysilicon reduction furnace system comprising at least one polysilicon reduction furnace production line, each polysilicon reduction furnace production line comprising at least one polysilicon reduction furnace, and the sample current curve being a current curve true value of operation current values of the polysilicon reduction furnaces in a polysilicon growth period.

[0030] A first data preprocessing module is configured to perform data preprocessing on the sample operation parameters to obtain processed sample operation parameters that are non-abnormal and meet a storage structure.

[0031] A calculation and analysis module is configured to input the processed sample operation parameters into a data calculation and analysis model to be trained for calculation and analysis, and output a predicted sample current curve of the polysilicon reduction furnaces in the polysilicon growth period.

[0032] A parameter adjusting module is configured to adjust parameters of the data calculation and analysis model to be trained according to the sample current curve and the predicted sample current curve, and obtain a trained data calculation and analysis model.

[0033] Optionally, the sample operation parameters are historical operation parameters of the polysilicon reduction furnace system in a polysilicon growth period, and include at least one of the following: the number of operations of the polysilicon reduction furnace, the operation current value of each polysilicon reduction furnace, the operation time, the in-out material, and the operation power.

[0034] Optionally, the parameter adjusting module includes:

[0035] A parameter adjusting submodule is configured to input the sample current curve and the predicted sample current curve into a mechanism learning model, and store the sample current curve and the predicted sample current curve in a database, wherein the mechanism learning model adjusts the parameters of the data calculation and analysis model to be trained through data analysis and learning, and obtains a trained data calculation and analysis model.

[0036] Optionally, the first data preprocessing module includes:

[0037] A structure adjusting submodule is configured to adjust the sample operation parameters according to a storage structure of the database, obtain adjusted sample operation parameters, and store the adjusted sample operation parameters in the database.

[0038] A screening submodule is configured to perform preliminary screening calculation on the adjusted sample operation parameters according to the operation current value of each polysilicon reduction furnace, the operation time, the in-out material, and the operation power, obtain calculated sample operation parameters, and perform abnormal result screening filtering on the calculated sample operation parameters, to obtain processed sample operation parameters that are non-abnormal and meet the database structure.

[0039] In a fourth aspect, an embodiment of the present application provides a start-stop control device for a polysilicon reduction furnace, including:

[0040] A second acquisition module is configured to acquire operation parameters of the polysilicon reduction furnace in a polysilicon reduction furnace system in real time, wherein the polysilicon reduction furnace system includes at least one polysilicon reduction furnace production line, and each polysilicon reduction furnace production line includes at least one polysilicon reduction furnace.

[0041] A second data preprocessing module is configured to perform data preprocessing on the operation parameters, to obtain processed operation parameters that are non-abnormal and meet a storage structure.

[0042] A first processing module is configured to input the processed operation parameters into the trained data calculation and analysis model for calculation and analysis, to obtain a predicted current curve composed of the operation current value of the polysilicon reduction furnace in a polysilicon growth period.

[0043] a third obtaining module, configured to obtain a current threshold of power consumption of the polysilicon reduction furnace system;

[0044] a first predicting module, configured to predict, according to the current threshold and the current curve, a current curve of a current value of each of the polysilicon reduction furnaces and a total current value of the polysilicon reduction furnace system in a future preset time period;

[0045] a second predicting module, configured to obtain, according to the predicted current curve of the current value of each of the polysilicon reduction furnaces and the total current value of the polysilicon reduction furnace system, a predicted number of the polysilicon reduction furnaces in operation, an on-time of each of the polysilicon reduction furnaces, and a new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of the polysilicon reduction furnaces in operation and the on-time of each of the polysilicon reduction furnaces.

[0046] Optionally, the method further comprises:

[0047] a fourth obtaining module, configured to dynamically display, in real time, the predicted number of the polysilicon reduction furnaces in operation, the on-time of each of the polysilicon reduction furnaces, and the new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of the polysilicon reduction furnaces in operation and the on-time of each of the polysilicon reduction furnaces.

[0048] Optionally, the operation parameters of the polysilicon reduction furnace system include at least one of the following: a number of the polysilicon reduction furnaces in operation, a running current value of each of the polysilicon reduction furnaces, a running time, an in-out material, and a running power.

[0049] Optionally, the obtaining of the operation parameters of the polysilicon reduction furnace system comprises:

[0050] an obtaining sub-module, configured to obtain, in real time, the operation parameters of the polysilicon reduction furnace system from a distributed control system and a data acquisition and monitoring control system.

[0051] In a fifth aspect, an electronic device is provided, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, and the program or instruction, when executed by the processor, implements the steps of the method for training the data calculation analysis model according to any one of the first aspect, or the method for starting and stopping the polysilicon reduction furnace according to any one of the second aspect.

[0052] In a sixth aspect, an embodiment of the present application provides a readable storage medium, the readable storage medium storing a program or instructions, the program or instructions being executed by a processor to implement the method for training the data calculation and analysis model according to any one of the first aspect, or to implement the steps in the method for starting and stopping control of the polysilicon reduction furnace according to any one of the second aspect.

[0053] In the present application, by collecting, analyzing, calculating and learning the full data of the polysilicon reduction furnace, a prediction curve is generated and dynamically adjusted and optimized as the historical data increases, and the starting and stopping operation of the polysilicon reduction furnace is planned accordingly. BRIEF DESCRIPTION OF DRAWINGS

[0054] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the application. Moreover, like reference numerals designate like parts throughout the several views in the drawings. In the drawings:

[0055] Figure 1 is a flowchart of the method for training the data calculation and analysis model provided by an embodiment of the present application;

[0056] Figure 2 is a flowchart of the method for starting and stopping control of the polysilicon reduction furnace provided by an embodiment of the present application;

[0057] Figure 3 is a structural schematic diagram of the method for starting and stopping control of the polysilicon reduction furnace provided by an embodiment of the present application;

[0058] Figure 4 is a flowchart of another method for starting and stopping control of the polysilicon reduction furnace provided by an embodiment of the present application;

[0059] Figure 5 is a workshop schematic diagram of the method for starting and stopping control of the polysilicon reduction furnace provided by an embodiment of the present application;

[0060] Figure 6 is a current curve schematic diagram of the method for starting and stopping control of the polysilicon reduction furnace provided by an embodiment of the present application;

[0061] Figure 7 is a schematic diagram of the training device of the data calculation and analysis model provided by an embodiment of the present application;

[0062] Figure 8 is a schematic diagram of the starting and stopping control device of the polysilicon reduction furnace provided by an embodiment of the present application;

[0063] Figure 9 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0065] Please refer to Figure 1 The embodiment of the present application provides a data calculation analysis model training method, comprising:

[0066] Step 11: Obtain sample running parameters and sample current curves of each polysilicon reduction furnace in a polysilicon reduction furnace system, wherein the polysilicon reduction furnace system comprises at least one polysilicon reduction furnace production line, each polysilicon reduction furnace production line comprises at least one polysilicon reduction furnace, and the sample current curve is a current curve true value composed of running current values of the polysilicon reduction furnace in a polysilicon growth period.

[0067] In the embodiment of the present application, the sample running parameters are historical running parameters of the polysilicon reduction furnace system in the polysilicon growth period, and at least one of the following is included: running number of the polysilicon reduction furnace, running current value, running time, material in and out, and running power of each polysilicon reduction furnace.

[0068] In the embodiment of the present application, the sample running parameters are collected from an industrial internet platform, and the industrial internet platform is installed and deployed with a real-time database and a relational database for debugging. Then, data of a distributed control system (DCS), a programmable logic controller system (PLC), and a supervisory control and data acquisition system (SCADA) are integrated into the real-time database, so as to quickly and effectively improve the fine management level of enterprises.

[0069] Step 12: Data preprocessing is performed on the sample running parameters to obtain processed sample running parameters that are non-abnormal and meet a storage structure.

[0070] In the embodiment of the present application, the data preprocessing on the sample running parameters to obtain processed sample running parameters that are non-abnormal and meet a database structure comprises:

[0071] Adjust the sample running parameter according to the storage structure of the database to obtain an adjusted sample running parameter and store it in the database.

[0072] According to the running current value, running time, incoming and outgoing material and running power of each polysilicon reduction furnace, the adjusted sample running parameter is calculated to obtain a calculated sample running parameter, and the calculated sample running parameter is filtered to obtain a non-anomalous and database-structure-satisfied processed sample running parameter, and the non-anomalous and database-structure-satisfied processed sample running parameter is classified according to the type of the polysilicon reduction furnace.

[0073] Step 13: The processed sample running parameter is input into a data calculation and analysis model to be trained for calculation and analysis, and a predicted sample current curve of the polysilicon reduction furnace in a polysilicon growth cycle is output.

[0074] Step 14: The parameters of the data calculation and analysis model to be trained are adjusted according to the sample current curve and the predicted sample current curve to obtain a trained data calculation and analysis model.

[0075] In the embodiment of the application, the sample current curve and the predicted sample current curve are input into a mechanism learning model, and the sample current curve and the predicted sample current curve are stored in a database, the mechanism learning model adjusts the parameters of the data calculation and analysis model to be trained through data analysis and learning to obtain a trained data calculation and analysis model.

[0076] In the embodiment of the application, the processed sample running parameter is input into a data calculation and analysis model for data calculation and analysis according to an embedded big data analysis model, the result is output to a database and then transmitted to a mechanism learning model, and after massive data analysis and learning, the mechanism learning model optimizes the big data analysis model in the data calculation and analysis model to further optimize the analysis model.

[0077] In the embodiment of the application, the full-amount data of the polysilicon reduction furnace is collected, analyzed, calculated and learned to generate a prediction curve and realize dynamic adjustment and optimization of the curve as the historical data increases, and the start and stop operation of the polysilicon reduction furnace is planned accordingly.

[0078] Please refer to Figure 2 The embodiment of the application provides a start and stop control method of a polysilicon reduction furnace.

[0079] Step 21: Real-time acquisition of the running parameter of the polysilicon reduction furnace in a polysilicon reduction furnace system, the polysilicon reduction furnace system comprising at least one polysilicon reduction furnace production line, and each polysilicon reduction furnace production line comprising at least one polysilicon reduction furnace.

[0080] The operation parameter is collected from the industrial internet platform to the running polycrystalline silicon reduction furnace in real time, and is installed and deployed with a real-time database and a relational database based on the industrial internet platform, and the data of a distributed control system (DCS), a programmable logic controller system (PLC) and a supervisory control and data acquisition system (SCADA) are integrated into the real-time database, so that the fine management level of the enterprise is quickly and effectively improved.

[0081] Step 22: data preprocessing is performed on the operation parameter to obtain processed operation parameters that are non-abnormal and meet a storage structure;

[0082] In the embodiment of the application, optionally, the data preprocessing on the operation parameter to obtain processed operation parameters that are non-abnormal and meet a database structure comprises:

[0083] The operation parameter is adjusted according to a storage structure meeting the database to obtain adjusted operation parameters, and the adjusted operation parameters are stored in the database;

[0084] According to the running current value, running time, in-out material and running power of each polycrystalline silicon reduction furnace, the adjusted operation parameters are calculated to obtain calculated operation parameters, the calculated operation parameters are filtered for abnormal results to obtain processed operation parameters that are non-abnormal and meet a database structure, and the processed operation parameters that are non-abnormal and meet the database structure are classified according to the polycrystalline silicon reduction furnace model.

[0085] Step 23: the processed operation parameters are input into a trained data calculation and analysis model for calculation and analysis to obtain a predicted current curve constituted by the running current value of the polycrystalline silicon reduction furnace in a polycrystalline silicon growth cycle;

[0086] In the embodiment of the application, the future value of the running current curve of each reduction furnace is predicted, and finally the current historical curve and the predicted curve of each production line are obtained.

[0087] Step 24: a current threshold of the power consumption of the polycrystalline silicon reduction furnace system is obtained;

[0088] Step 25: according to the current threshold and the current curve, the current value of each polycrystalline silicon reduction furnace and the current curve of the total current value of the polycrystalline silicon reduction furnace system in a future preset time period are predicted;

[0089] Step 26: obtaining a predicted number of polysilicon reduction furnaces in operation, an opening time of each of the polysilicon reduction furnaces, and a new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of polysilicon reduction furnaces in operation and the opening time of each of the polysilicon reduction furnaces based on the current curve of the predicted current value of each of the polysilicon reduction furnaces and the total current value of the polysilicon reduction furnace system.

[0090] In the embodiment of the present application, the number of reduction furnaces that can be started at which time under the premise that the current does not exceed the threshold value is calculated according to the current threshold value and the current curve of each production line, and then a new current curve is generated according to the predicted opening time and the number of reduction furnaces.

[0091] In the embodiment of the present application, the full amount of data of the polysilicon reduction furnace is collected, analyzed, calculated and learned to generate a prediction curve and realize dynamic adjustment and optimization of the curve as the historical data increases, and the planning of the start and stop operation of the polysilicon reduction furnace is performed accordingly.

[0092] In the embodiment of the present application, optionally, it further comprises:

[0093] The predicted number of polysilicon reduction furnaces in operation, the opening time of each of the polysilicon reduction furnaces, and the new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of polysilicon reduction furnaces in operation and the opening time of each of the polysilicon reduction furnaces are dynamically displayed in real time.

[0094] In the embodiment of the present application, the current curve is displayed on the visual prediction display page, and the number of reduction furnaces that can be started and the time of each production line are dynamically displayed.

[0095] In the embodiment of the present application, optionally, the real-time acquisition of the operation parameters of the polysilicon reduction furnace in the polysilicon reduction furnace system comprises:

[0096] The operation parameters of the polysilicon reduction furnace system are acquired in real time from the distributed control system and the data acquisition and monitoring control system.

[0097] In the embodiment of the present application, based on the industrial internet platform, based on various types of main data foundation after management, based on various business systems as the main starting point, based on the whole process of polysilicon production, storage, sales, etc. As a business breakthrough point, an intelligent factory production control integration platform for the polysilicon industry is creatively established, a new mode of "data + platform + application" is adopted; the platform deeply fuses the bidding and procurement system, the weighing system, the material management system, the warehouse logistics system, the warehouse management system (WMS), the sales system, the factory visitor system, and breaks through the whole process of production and sales of polysilicon enterprises: procurement demand, procurement order, raw material storage, production, product storage, product sales, loading and weighing, product delivery, etc. The present application realizes the integration with the distributed control system (DCS), the programmable logic controller system (PLC), the data acquisition and monitoring control system (SCADA), the warehouse management system (WMS), the weighing system, the enterprise resource planning (ERP) system, the customer relationship management system (CRM), the main data management system (MDM) and the digital delivery system based on the platform, breaks through the business process from the process control layer to the production layer to the management layer, fully learns and absorbs the experience of the chemical process manufacturing industry, and introduces the flow single, a wisdom crystallization of the discrete manufacturing industry, breaks through the originally independent and segmented production process, optimizes the reduction furnace production scheduling problem in the polysilicon production process, on the one hand reduces the personnel and energy consumption cost, on the other hand realizes scientific production scheduling to achieve fine energy management and precise process management.

[0098] Please refer to Figure 3In the embodiment of the present application, the electric workshop sends data in the power system to the protocol conversion gateway through the 104 protocol, the gateway converts the data of the 104 protocol into the OPC protocol and then sends the data to each data acquisition machine for storage and upward forwarding, the real-time database receives the OPC protocol sent by the DCS system, the WMS system and the SCADA system, receives the ODBC protocol sent by the electric quantity acquisition system, the industrial internet platform receives the material master data and the personnel master data of the main data system, the price information of the ERP system and feeds back the production consumption data, provides the product quality information to the CRM system, and provides the approval process to the OA system, the real-time database stores the received data and provides the data to the big data analysis application platform, the big data analysis application platform calculates the feasible scheme and the optimal scheme of starting the furnace in real time through a calculation model and power data, the on-duty dispatcher guides the field starting of the furnace according to the scheme provided by the system, and the electric power consumption data of the furnace after starting is returned to the industrial internet platform through the above process, so that the process and data are closed.

[0099] In the embodiment of the present application, the real-time database system is used to acquire the data of the control system such as the DCS system and the SCADA system in real time, so that the production process data of each production unit, materials, energy, electricity, equipment and the like are collected and stored for a long period, the industrial internet platform is used to realize unified management of the production process and business data and fusion and intercommunication with the related systems such as the ERP system, the CRM system and the office automation system (OA), meanwhile, the data analysis and visual display function of the industrial internet platform is used to analyze and comprehensively display the production process and business data in the platform according to the business management requirements and the data correlation, so that the production process is safe and controllable, the production management is refined, and the production decision is visualized and intelligentized, thereby enhancing the profitability and core competitiveness of the enterprise, promoting the fusion of informatization and the future intelligent enterprise business of Inner Mongolia Xintai, and continuously improving the industrialization level of polysilicon.

[0100] Please refer to Figure 4 In the model learning stage, the data acquisition module collects the full amount of data of the polysilicon reduction furnace such as the number of operations, the operation time, the in-out material and the power from the industrial internet platform, that is, the operation parameters of the polysilicon reduction furnace system are acquired in real time;

[0101] The data preprocessing module adjusts the collected full amount of data according to the structure of the platform database and stores the data through the data storage module, that is, the operation parameters are preprocessed to obtain the processed operation parameters which are non-abnormal and meet the storage structure;

[0102] The data calculation module performs simple data calculation to obtain preliminary results;

[0103] Filter out abnormal data and data not meeting the storage conditions through data filtering and data screening;

[0104] Classify qualified data according to the type of the reduction furnace through data classification;

[0105] Input the classified data into a data calculation and analysis model, the data calculation and analysis model performs data calculation and analysis according to the built-in big data analysis model, and outputs the results to a database and then transmits them to a mechanism learning model, after analyzing and learning a large amount of data, the mechanism learning model optimizes the big data analysis model in the data calculation and analysis model to further optimize the analysis model.

[0106] In the model prediction stage, data measurement performs data calculation and analysis by collecting parameters of the running reduction furnace, predicts the future value of the running current curve of each reduction furnace, and finally obtains the current historical curve and the predicted curve of each production line, calculates when to start several reduction furnaces under the premise that the current does not exceed the threshold value according to the current threshold value and the current curve of each production line, and then generates a new current curve according to the start time and the number of the predicted reduction furnaces;

[0107] The final current curve is displayed on the visual prediction display page, and the number and time of the reduction furnaces that can be started on each production line are dynamically displayed.

[0108] In the embodiment of the application, the experience of people is solidified into the system in the form of an algorithm by using information technology to realize the system recommended start-up scheme, so as to realize scientific, standardized and refined management, which is the professional subject of the project.

[0109] Please refer to Figure 5 and Figure 6For the specific application of the embodiment of the present application, for example: there are 4 production lines in the reduction workshop, that is, four polysilicon reduction furnace systems, each production line has 38 polysilicon reduction furnaces, the growth time of polysilicon in each polysilicon reduction furnace is about 100 hours, the current value of each furnace from start to end presents low first and then high, after reaching the highest value (such as: 120A), it gently decreases to a certain current value (such as: 100A-80A), after the polysilicon product in the reduction furnace is grown (about 100 hours), the current value will be 0A instantaneously, and after the production is stable, the current value of each time point of the reduction furnace is calculated by using big data analysis, and the current / time curve of the reduction furnace is accurately drawn; if the maximum current of each production line is 2000A, the rated current of each reduction furnace is 100A, and the maximum current can reach 120A, the current consumption of each production line during production cannot exceed 2000A, according to the number of polysilicon reduction furnaces currently running in each production line and the current consumption, the current use of each production line is calculated, and the current consumption and curve within 100 hours after the current and one / two / three reduction furnaces are put into production are calculated, thereby guiding which production line one or more reduction furnaces to be put into production next without exceeding the current threshold of 2000A, effectively guiding production, that is, according to the predicted current value of each polysilicon reduction furnace and the current curve of the total current value of the polysilicon reduction furnace system, the predicted running number of the polysilicon reduction furnace, the opening time of each polysilicon reduction furnace, and the new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted running number of the polysilicon reduction furnace and the opening time of each polysilicon reduction furnace are obtained.

[0110] Please refer to Figure 7 The embodiment of the present application provides a data calculation and analysis model training device, which comprises:

[0111] The first acquisition module 71 is used for acquiring sample running parameters and a sample current curve of each polysilicon reduction furnace in a polysilicon reduction furnace system, the polysilicon reduction furnace system comprises at least one polysilicon reduction furnace production line, each polysilicon reduction furnace production line comprises at least one polysilicon reduction furnace, and the sample current curve is a current curve true value composed of running current values of the polysilicon reduction furnace in a polysilicon growth period;

[0112] The first data preprocessing module 72 is used for data preprocessing of the sample running parameters to obtain processed sample running parameters that are non-abnormal and meet a storage structure;

[0113] The calculation and analysis module 73 is used for inputting the processed sample running parameters into a data calculation and analysis model to be trained for calculation and analysis, and outputting a predicted sample current curve of the polysilicon reduction furnace in a polysilicon growth period;

[0114] The parameter adjusting module 74 is configured to adjust parameters of the data calculation and analysis model to be trained according to the sample current curve and the predicted sample current curve, and obtain a trained data calculation and analysis model.

[0115] In the embodiment of the present application, the sample operation parameter is a historical operation parameter of a polysilicon reduction furnace system in a polysilicon growth cycle, and includes at least one of the following: the number of operations of the polysilicon reduction furnace, the operation current value of each polysilicon reduction furnace, the operation time, the material in and out, and the operation power.

[0116] In the embodiment of the present application, the parameter adjusting module includes:

[0117] The parameter adjusting submodule is configured to input the sample current curve and the predicted sample current curve into a mechanism learning model, and store the sample current curve and the predicted sample current curve into a database, wherein the mechanism learning model adjusts the parameters of the data calculation and analysis model to be trained through data analysis and learning, and obtains a trained data calculation and analysis model.

[0118] In the embodiment of the present application, the first data preprocessing module includes:

[0119] The structure adjusting submodule is configured to adjust the sample operation parameter according to a storage structure of the database, obtain an adjusted sample operation parameter, and store the adjusted sample operation parameter in the database.

[0120] The screening submodule is configured to perform preliminary screening calculation on the adjusted sample operation parameter according to the operation current value of each polysilicon reduction furnace, the operation time, the material in and out, and the operation power, obtain a calculated sample operation parameter, perform abnormal result screening and filtering on the calculated sample operation parameter, and obtain a processed sample operation parameter that is non-abnormal and meets the database structure.

[0121] The training device of the data calculation and analysis model provided by the embodiment of the present application can realize each process of the method embodiment and achieve the same technical effect, and thus repeated description is omitted here. Figure 1 The method embodiment realizes each process and achieves the same technical effect, and thus repeated description is omitted here.

[0122] Please refer to Figure 8 The embodiment of the present application provides a start-stop control device of a polysilicon reduction furnace, which includes:

[0123] The second acquisition module 81 is configured to acquire operation parameters of the polysilicon reduction furnace in a polysilicon reduction furnace system in real time, wherein the polysilicon reduction furnace system includes at least one polysilicon reduction furnace production line, and each polysilicon reduction furnace production line includes at least one polysilicon reduction furnace.

[0124] The second data preprocessing module 82 is configured to perform data preprocessing on the operation parameters to obtain processed operation parameters which are non-abnormal and meet a storage structure.

[0125] The first processing module 83 is configured to input the processed operation parameters into a trained data calculation and analysis model to perform calculation and analysis, so as to obtain a predicted current curve constituted by operation current values of the polysilicon reduction furnace in a polysilicon growth cycle.

[0126] The third acquisition module 84 is configured to acquire a current threshold of power consumption of the polysilicon reduction furnace system.

[0127] The first prediction module 85 is configured to predict, according to the current threshold and the current curve, current values of each of the polysilicon reduction furnaces and a current curve of a total current value of the polysilicon reduction furnace system in a future preset time period.

[0128] The second prediction module 86 is configured to obtain, according to the predicted current values of each of the polysilicon reduction furnaces and the total current value of the polysilicon reduction furnace system, a predicted number of the polysilicon reduction furnaces, an opening time of each of the polysilicon reduction furnaces, and a new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of the polysilicon reduction furnaces and the opening time of each of the polysilicon reduction furnaces.

[0129] In the embodiment of the present application, optionally, the device further comprises:

[0130] The fourth acquisition module is configured to dynamically display, in real time, the predicted number of the polysilicon reduction furnaces, the opening time of each of the polysilicon reduction furnaces, and the new predicted current curve of the total current value of the polysilicon reduction furnace system according to the predicted number of the polysilicon reduction furnaces and the opening time of each of the polysilicon reduction furnaces.

[0131] In the embodiment of the present application, optionally, the operation parameters of the polysilicon reduction furnace system include at least one of the following: a number of the polysilicon reduction furnaces, an operation current value of each of the polysilicon reduction furnaces, an operation time, an in-out material, and an operation power.

[0132] In the embodiment of the present application, optionally, the operation parameters of the polysilicon reduction furnace system are acquired by:

[0133] The acquisition sub-module is configured to acquire, in real time, the operation parameters of the polysilicon reduction furnace system from a distributed control system and a data acquisition and monitoring control system.

[0134] The polysilicon reduction furnace start-stop control device provided by the embodiment of the present application can realize Figure 2The method embodiments of the present application implement the various processes and achieve the same technical effects, and thus, details are not repeated here.

[0135] The present application provides an electronic device 90, as shown in Figure 9 Figure 9 The present application provides an electronic device 90, as shown in

[0136] The present application provides an electronic device 90, as shown in

[0137] The computer readable medium includes permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition in this paper, the computer readable medium does not include transitory computer readable media such as modulated data signals and carriers.

[0138] It should be noted that in this paper, the term "includes", "contains" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method, article or device. Without more limitations, the element defined by the statement "includes a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0139] ​The above-mentioned embodiment numbers of the application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0140] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and the necessary general hardware platform, of course, they can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for causing a service classification device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0141] The above-mentioned is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should be regarded as the protection scope of the present application.

Claims

1. A method for starting and stopping a polysilicon reduction furnace, characterized in that: include: Acquiring in real time operating parameters of a polysilicon reduction furnace in a polysilicon reduction furnace system, wherein the polysilicon reduction furnace system includes at least one polysilicon reduction furnace production line, and each polysilicon reduction furnace production line includes at least one polysilicon reduction furnace; Performing data preprocessing on the operating parameters to obtain processed operating parameters that are not abnormal and meet the storage structure; Inputting the processed operating parameters into the trained data calculation and analysis model for calculation and analysis to obtain a predicted current curve consisting of operating current values ​​of the polysilicon reduction furnace during the polysilicon growth period; Obtaining a current threshold value of power consumption of the polysilicon reduction furnace system; Predicting, based on the current threshold and the current curve, the current value of each of the polysilicon reduction furnaces and the current curve of the total current value of the polysilicon reduction furnace system within a preset time period in the future; According to the current curve of the predicted current value of each of the polysilicon reduction furnaces and the total current value of the polysilicon reduction furnace system, the predicted number of running polysilicon reduction furnaces, the start time of each of the polysilicon reduction furnaces and the new predicted current curve of the total current value of the polysilicon reduction furnace system obtained based on the predicted number of running polysilicon reduction furnaces and the start time of each of the polysilicon reduction furnaces are obtained.

2. The method for starting and stopping a polysilicon reduction furnace according to claim 1, wherein: Also includes: The predicted number of running polysilicon reduction furnaces, the start time of each polysilicon reduction furnace and the new predicted current curve of the total current value of the polysilicon reduction furnace system obtained based on the predicted number of running polysilicon reduction furnaces and the start time of each polysilicon reduction furnace are dynamically displayed in real time.

3. The method for starting and stopping a polysilicon reduction furnace according to claim 1, wherein: The operating parameters of the polysilicon reduction furnace system include at least one of the following: the number of operating polysilicon reduction furnaces, the operating current value of each polysilicon reduction furnace, the operating time, the input and output materials, and the operating power.

4. The method for starting and stopping a polysilicon reduction furnace according to claim 1, wherein: The real-time acquisition of the operating parameters of the polysilicon reduction furnace in the polysilicon reduction furnace system includes: The operating parameters of the polysilicon reduction furnace system are obtained in real time from the distributed control system and the data acquisition and monitoring control system.

5. The method for starting and stopping a polysilicon reduction furnace according to claim 1, wherein: Before inputting the processed operating parameters into the trained data calculation and analysis model for calculation and analysis to obtain a predicted current curve consisting of the operating current value of the polysilicon reduction furnace during the polysilicon growth period, the method further includes: Obtaining sample operating parameters and a sample current curve of each polysilicon reduction furnace in a polysilicon reduction furnace system, wherein the polysilicon reduction furnace system includes at least one polysilicon reduction furnace production line, and each polysilicon reduction furnace production line includes at least one polysilicon reduction furnace, wherein the sample current curve is a true value of a current curve formed by operating current values ​​of the polysilicon reduction furnace during a polysilicon growth cycle; Performing data preprocessing on the sample operating parameters to obtain processed sample operating parameters that are non-abnormal and meet the storage structure; Inputting the processed sample operating parameters into the data calculation and analysis model to be trained for calculation and analysis, and outputting a predicted sample current curve of the polysilicon reduction furnace during the polysilicon growth cycle; The parameters of the data calculation and analysis model to be trained are adjusted according to the sample current curve and the predicted sample current curve to obtain a trained data calculation and analysis model.

6. The method for starting and stopping a polysilicon reduction furnace according to claim 5, wherein: The sample operating parameters are historical operating parameters of the polysilicon reduction furnace system during the polysilicon growth cycle, including at least one of the following: the number of operating polysilicon reduction furnaces, the operating current value of each polysilicon reduction furnace, the operating time, the input and output materials, and the operating power.

7. The method for starting and stopping a polysilicon reduction furnace according to claim 6, wherein: The processing of the sample operation parameters into the data calculation and analysis model to be trained for calculation and analysis, and outputting a predicted sample current curve of the polysilicon reduction furnace during the polysilicon growth cycle, includes: The sample current curve and the predicted sample current curve are input into a mechanism learning model, and the sample current curve and the predicted sample current curve are stored in a database. The mechanism learning model adjusts the parameters of the data calculation and analysis model to be trained through data analysis and learning to obtain a trained data calculation and analysis model.

8. The method for starting and stopping a polysilicon reduction furnace according to claim 6, wherein: The data preprocessing of the sample operating parameters to obtain processed sample operating parameters that are not abnormal and meet the database structure includes: Adjusting the sample operating parameters according to the storage structure of the database to obtain the adjusted sample operating parameters and storing them in the database; The adjusted sample operating parameters are preliminarily screened and calculated based on the operating current value, operating time, input and output materials, and operating power of each polysilicon reduction furnace to obtain the calculated sample operating parameters, and the calculated sample operating parameters are screened and filtered for abnormal results to obtain processed sample operating parameters that are non-abnormal and meet the database structure.

9. A start-stop control device for a polysilicon reduction furnace, characterized in that: include: A second acquisition module is configured to acquire in real time operating parameters of a polysilicon reduction furnace in a polysilicon reduction furnace system, wherein the polysilicon reduction furnace system includes at least one polysilicon reduction furnace production line, and each polysilicon reduction furnace production line includes at least one polysilicon reduction furnace; A second data preprocessing module is used to perform data preprocessing on the operating parameters to obtain processed operating parameters that are not abnormal and meet the storage structure; A first processing module is configured to input the processed operating parameters into a trained data calculation and analysis model for calculation and analysis, thereby obtaining a predicted current curve consisting of operating current values ​​of the polysilicon reduction furnace during a polysilicon growth period; A third acquisition module is used to obtain a current threshold of power consumption of the polysilicon reduction furnace system; a first prediction module, configured to predict, based on the current threshold and the current curve, a current value of each of the polysilicon reduction furnaces and a current curve of a total current value of the polysilicon reduction furnace system within a preset time period in the future; The second prediction module is used to obtain the predicted number of operating polysilicon reduction furnaces, the start time of each polysilicon reduction furnace and the new predicted current curve of the total current value of the polysilicon reduction furnace system based on the predicted current value of each polysilicon reduction furnace and the start time of each polysilicon reduction furnace.

10. An electronic device, characterized in that: It includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps in the start-stop control method of the polysilicon reduction furnace as described in any one of claims 1 to 7 are implemented.

11. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps in the start-stop control method of the polysilicon reduction furnace according to any one of claims 1 to 7 are implemented.

Images