A rare earth production process virtual inspection and process simulation method and system

Abstract

This invention relates to a virtual inspection and process simulation method and system for rare earth production processes, belonging to the field of digital twin technology. A virtual workshop is established based on the geometric model of the production site, control scripts, and real-time data, simulating the entire process. By establishing the virtual rare earth workshop and creating a data connection between the actual workshop and the virtual factory, data visualization demonstrations of various production processes can be achieved, along with rapid inspection of production equipment. Furthermore, based on the extraction mechanism and historical data, component content is predicted, and automatic early warnings are issued, enabling real-time and accurate prediction of process indicators. Simultaneously, the system can quickly respond to changes in production conditions, providing system early warnings and optimization data, which helps reduce the burden on operators, optimize production process control, improve production management efficiency, and achieve effective utilization of on-site process data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital twinning, in particular to a rare earth production process virtual inspection and process simulation method and system based on digital twinning. BACKGROUND

[0002] Since the enrichment degree of rare earth in raw materials is low, it needs to go through dozens to hundreds of extraction processes to get high-purity target products, so the whole extraction process has large time lag, complex control, and time-consuming and laborious working condition detection characteristics, in addition, the data of each process is isolated and difficult to optimize and inspect, and since the extraction process has large time lag, real-time working condition prediction and timely optimization control are particularly important.

[0003] However, the existing rare earth process relies heavily on manual control and inspection, making process inspection difficult, and reacting slowly to abnormal working conditions, and relying heavily on manual process optimization, and having difficulty using on-site data.

[0004] Therefore, there is an urgent need in the art for a technical solution that can realize centralized monitoring of process indicators and simulate and predict process indicators using on-site data. SUMMARY

[0005] The purpose of the present application is to provide a rare earth production process virtual inspection and process simulation method and system based on digital twinning, first, a virtual workshop is established and is one-to-one corresponding to real-time data, the whole process simulation is completed, thereby realizing automatic inspection, and the component content is predicted according to the extraction mechanism and historical data, then automatically warning, thereby effectively solving the problems in the prior art that rely heavily on manual control and inspection, making process inspection difficult, and reacting slowly to abnormal working conditions, and relying heavily on manual process optimization, and having difficulty using on-site data.

[0006] To achieve the above purpose, the present application provides the following scheme:

[0007] A rare earth production process virtual inspection and process simulation method, the method comprising:

[0008] Obtaining real-time data of the production site; the real-time data includes process production index data;

[0009] Building a rare earth virtual workshop according to the geometric model and control script of the production site and the real-time data for user inspection;

[0010] Obtaining process production index historical data;

[0011] The parameter optimization algorithm is used to optimize an extraction mechanism model according to historical data of process production indexes, so as to obtain a data-driven model; the extraction mechanism model is a mathematical model representing an extraction mechanism of rare earth;

[0012] Obtain process information input by a user;

[0013] Use the data-driven model to predict component contents of each element in a process product according to the process information;

[0014] Determine whether to perform early warning according to the component contents of each element.

[0015] In some embodiments, the parameter optimization algorithm is used to optimize an extraction mechanism model according to historical data of process production indexes, so as to obtain a data-driven model, and specifically includes the following steps:

[0016] A particle swarm algorithm is used to optimize a separation coefficient in the extraction mechanism model according to the historical data of process production indexes, so as to obtain a data-driven model.

[0017] The application further provides a rare earth production process virtual inspection and process simulation system, which comprises a rare earth production control information system, a rare earth production virtual workshop and a digital twin service system.

[0018] The rare earth production control information system is used to obtain real-time data of a production site and control a production process.

[0019] The rare earth production virtual workshop is used to build a rare earth virtual workshop according to a geometric model and a control script of the production site and the real-time data, for user inspection.

[0020] The digital twin service system is used to:

[0021] Obtain historical data of process production indexes;

[0022] The parameter optimization algorithm is used to optimize an extraction mechanism model according to the historical data of process production indexes, so as to obtain a data-driven model; the extraction mechanism model is a mathematical model representing an extraction mechanism of rare earth;

[0023] Obtain process information input by a user;

[0024] Use the data-driven model to predict component contents of each element in a process product according to the process information;

[0025] Determine whether to perform early warning according to the component contents of each element.

[0026] In some embodiments, the rare earth production control information system comprises a basic control module and a process detection module.

[0027] The basic control module is configured to execute control instructions.

[0028] The basic control module comprises a motor frequency converter, a flow pump, a quantitative pump, a solenoid valve, and a PLC. The motor frequency converter is configured to adjust the rotating speed of the stirrer. The flow pump and the quantitative pump are configured to control the quantitative feeding of the extraction process. The solenoid valve is configured to control the feeding and discharging of the material liquid. The PLC is configured to obtain control instructions and transmit the control instructions to the motor frequency converter, the flow pump, the quantitative pump, and the solenoid valve.

[0029] The process detection module is configured to obtain real-time data.

[0030] The process detection module comprises a flow meter, a liquid level meter, a thermometer, a pH meter, and a component content detection device. The flow meter is configured to monitor the flow rate of the material liquid in and out of the extraction process. The liquid level meter is configured to monitor the liquid level in the extraction tank and the material storage tank. The thermometer and the pH meter are configured to configure the washing liquid and the extraction liquid to meet the temperature and pH requirements. The component content detection device is arranged at each detection stage of the rare earth extraction process.

[0031] In some embodiments, the rare earth production control information system further comprises:

[0032] A twin database configured to store historical data, real-time data, and early warning information.

[0033] In some embodiments, the rare earth production control information system further comprises a data transmission module configured to interact information between the twin database and the rare earth production control information system.

[0034] In some embodiments, the rare earth production virtual workshop comprises a data interaction module, a geometric model library, a user interaction module, and a scene switching module.

[0035] The data interaction module is configured to query relevant data in the twin database according to the correspondence between the process and the data.

[0036] The geometric model library is configured to establish a rare earth virtual workshop and visualize the rare earth virtual workshop. The rare earth virtual workshop is established by using modeling software.

[0037] The user interaction module is configured to control the scene view and the demonstration animation during user inspection.

[0038] The scene switching module is configured to switch the rare earth virtual workshop during user inspection.

[0039] In some embodiments, the digital twin service system further comprises a process optimization module configured to:

[0040] An optimization strategy is obtained by using an optimization control algorithm according to a target set by a user; the optimization control algorithm adopts an extraction process reagent amount optimization control method of static setting and dynamic compensation.

[0041] In some embodiments, the control production process specifically comprises:

[0042] The production process is controlled according to the optimization strategy.

[0043] In some embodiments, the digital twin service system further comprises a model updating module configured to:

[0044] An error content is obtained by calculating an error between the predicted component content and actual process index data.

[0045] The data-driven model is adjusted by using the error content by using a parameter optimization algorithm.

[0046] According to the specific embodiments of the present application, the following technical effects are provided:

[0047] The present application establishes a virtual workshop according to the geometric model and control script of the production site and real-time data of the site, completes the simulation of the entire process, and realizes the data visualization demonstration of each production process and the rapid inspection of the production equipment by establishing a rare earth virtual workshop and establishing the data connection between the actual workshop and the virtual factory. In addition, the component content is predicted according to the extraction mechanism and historical data, and then an automatic early warning is performed, realizing the real-time and accurate prediction of the process index. At the same time, the production condition change can be quickly responded to, and the system early warning and optimization data are provided, which is beneficial to reduce the burden of the operator, optimize the production process control, improve the production management efficiency, and realize the effective utilization of the on-site process data. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0049] Figure 1 The rare earth production process virtual inspection and process simulation method flowchart provided for the first embodiment of the present application.

[0050] Figure 2 The block diagram of the rare earth production process virtual inspection and process simulation system provided for the second embodiment of the present application.

[0051] Figure 3A system block diagram of rare earth production process virtual inspection and process simulation based on digital twinning is provided for the fourth embodiment of the present application.

[0052] Figure 4 A logic flow diagram of a rare earth production control information system is provided for the fourth embodiment of the present application.

[0053] Figure 5 A logic flow diagram of a twin database is provided for the fourth embodiment of the present application.

[0054] Figure 6 A logic flow diagram of a digital twinning service system is provided for the fourth embodiment of the present application.

[0055] Figure 7 A logic flow diagram of a process index prediction model is provided for the fourth embodiment of the present application. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0057] Current rare earth process relies heavily on manual control and inspection, which leads to an increasingly urgent need for a rare earth production process virtual inspection and process simulation system. Taking a multi-component cascade extraction separation process as an example, the current rare earth extraction process obtains process indicators, i.e., the component content of each element, by offline testing of the solution at the detection stage, prevents abnormal conditions, and further adjusts the set value of the controller to optimize the production process. In view of the problems of difficult process inspection in rare earth production process, slow response to abnormal conditions, heavy reliance on manual process optimization, and difficulty in utilizing on-site data, the present application designs a general framework that centrally monitors and processes on-site data, stores and exchanges data in a database, simulates the process in a virtual scene, monitors the state, freely inspects, and integrates each module in a rare earth twinning service system to realize human-computer interaction functions. Digital twinning can be regarded as a collection of models and algorithms, which can realize data visualization, prediction, and process optimization functions. Its main function is to reduce the burden of manual work and improve production efficiency.

[0058] The purpose of the present application is to provide a rare earth production process virtual inspection and process simulation method and system based on digital twinning. First, a virtual workshop is established and matched with real-time data one by one to complete the simulation of the entire process, thereby realizing automatic inspection, predicting component content according to extraction mechanism and historical data, and then automatically warning, thereby effectively solving the problems in the prior art that a large amount of manual control and inspection is relied on, making process inspection difficult, and reacting slowly to abnormal conditions, and that process optimization relies heavily on manual work, and it is difficult to utilize on-site data.

[0059] In order to make the above-mentioned purposes, characteristics and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0060] Embodiment one:

[0061] As shown in the figure, the present embodiment provides a rare earth production process virtual inspection and process simulation method, which comprises: Figure 1

[0062] S1, acquiring real-time data of a production site; the real-time data comprises process production index data.

[0063] The process production index data comprises the flow of the feed and bleed extraction process, the liquid level in the extraction tank and the storage tank, the temperature and PH value of the washing liquid and the extraction liquid, and the component content in the material at each detection stage.

[0064] S2, building a rare earth virtual workshop according to the geometric model and control script of the production site and the real-time data, for user inspection.

[0065] First, a virtual workshop is built according to the geometric model and control script of the production site, and then the real-time data in the above is configured one by one into the virtual workshop, so that the user can freely inspect. Among them, the virtual workshop can be established according to the actual production equipment using modeling software such as 3DMAX, MAYA, etc.

[0066] S3, acquiring process production index historical data.

[0067] The process production index historical data can be manually inputted or obtained after the real-time data described above is stored.

[0068] S4, optimizing the extraction mechanism model according to the process production index historical data by using a parameter optimization algorithm to obtain a data-driven model.

[0069] ​The extraction mechanism model is a mathematical model representing the extraction mechanism of rare earths, including separation factor models between extraction stages (such as formulas (1)-(3)) and specific formulas such as material conservation (such as formulas (4)-(7)), separation factor compensation parameter formulas (such as formulas (8)-(9)).

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] Formula (1) represents the separation factor of two adjacent component elements, the relationship between the organic phase component content, formula 2 represents the separation factor of the i-th element relative to the first element, formula (3) is equivalent to the separation factor of the i-th element relative to the last element; formulas (4)-(5) represent the relationship between the water phase compositions of adjacent two stages in the extraction section, formulas (6)-(7) represent the relationship between the organic phases of adjacent two stages in the washing section, and formulas (8)-(9) represent the relationship between the water phase and the organic phase containing the separation factor compensation.

[0080] Wherein, X represents the component content of the water phase, and Y represents the component content of the organic phase. The material conservation between extraction stages is used to describe the conservation of extraction element component content in the upper and lower extraction stages. In the material conservation model, f F is the component of the feed liquid, f′ B is the outlet mole fraction of the difficult extraction component, is the outlet mole fraction of the easy extraction component, and the component contents of A outlet and B outlet are divided by respectively The organic phase composition of A outlet and the water phase composition of B outlet can be obtained, represents the extraction amount, W represents the washing amount, i represents the number of extraction elements, i is from 1 to N, N represents the number of the last extraction element, Y i represents the organic phase component content of the i-th element, X i represents the water phase component content of the i-th element, and β(1+i) / i denotes the separation coefficient of the previous element relative to the following element, β 1 / i denotes the separation coefficient of the i-th element relative to the first element, β i / N- denotes the separation coefficient of the last element relative to the i-th element, X [k+1，i] denotes the water phase component content of the i-th element at the k+1 level, Y [k，i] denotes the component content of the i-th element in the organic phase at the k level, the outlet mole fraction of the difficult-to-extract component f' B The product of the water phase component content can be used to establish a difficult-to-extract material conservation relationship with the water phase, organic phase component content, extraction amount, and washing amount of the feed level, and similarly, the outlet of the easy-to-extract component The product of the organic phase component content can be used to establish an easy-to-extract component extraction amount and washing amount material conservation relationship with the water phase and organic phase of the feed level. Through these conservation relationships, the component content of each level can be inferred level by level. By adding a compensation coefficient K to the separation coefficient, the on-site extraction insufficiency can be simulated, and after transformation, the relationship between the water phase and the organic phase can be obtained, and the component content of the other one can be obtained.

[0081] Then, the separation coefficient in the extraction mechanism model is optimized according to the process index historical data using the particle swarm algorithm to obtain a data-driven model. The optimization method is to optimize the compensation parameter K of the separation coefficient by using the improved particle swarm algorithm, and calculate the optimal compensation parameter.

[0082] S5, obtaining process information input by a user.

[0083] S6, predicting the component content of each element in a process product according to the process information by using the data-driven model. In this embodiment, the component content of Ce, Pr and Nd is mainly predicted according to the process information input by the user.

[0084] S7, judging whether to perform early warning according to the component content of each element.

[0085] The process index real-time prediction model established in the digital twin system can realize real-time and accurate prediction of the process index through the process index real-time prediction module. At the same time, the production condition change can be quickly responded, system early warning and optimization data can be provided, which is beneficial to reduce the burden of the operator, optimize the production process control, improve the production management efficiency, and realize effective utilization of the on-site process data.

[0086] The rare earth virtual plant is established, and the data connection between the actual plant and the virtual plant is established, so that the data visualization demonstration of each production process can be realized, and the rapid inspection of the production equipment can be realized.

[0087] The application optimizes the control strategy according to the optimization target set by the user and the actual process index data through the process optimization module in the digital twin service system, and realizes the optimization of the production process.

[0088] Embodiment two:

[0089] As shown in Figure 2 The embodiment provides a rare earth production process virtual inspection and process simulation system, which comprises a rare earth production control information system M1, a rare earth production virtual workshop M2 and a digital twin service system M3.

[0090] The rare earth production control information system M1 is used for acquiring real-time data of a production site and controlling a production process.

[0091] The rare earth production control information system M1 comprises a basic control module and a process detection module.

[0092] The basic control module is used for executing control instructions.

[0093] The basic control module comprises a motor frequency converter, a flow pump, a quantitative pump, an electromagnetic valve and a PLC; the motor frequency converter is used for adjusting the rotating speed of a stirrer; the flow pump and the quantitative pump are used for controlling the quantitative feeding of an extraction process; the electromagnetic valve is used for controlling the feeding and discharging of a material liquid; and the PLC is used for acquiring the control instructions and transmitting the control instructions to the motor frequency converter, the flow pump, the quantitative pump and the electromagnetic valve.

[0094] The process detection module is used for acquiring real-time data.

[0095] The process detection module comprises a flow meter, a liquid level meter, a thermometer, a pH meter and a component content detection device; the flow meter is used for monitoring the flow of a material liquid in and out of an extraction process; the liquid level meter is used for monitoring the liquid level in an extraction tank and a material storage tank; the thermometer and the pH meter are used for configuring washing liquid and extraction liquid meeting the temperature and pH requirements; and the component content detection device is arranged at each detection stage of a rare earth extraction process.

[0096] The rare earth production virtual workshop M2 is used for building a rare earth virtual workshop according to a geometric model and a control script of a production site and the real-time data, for user inspection.

[0097] The rare earth production virtual workshop M2 comprises a data interaction module, a geometric model library, a user interaction module and a scene switching module.

[0098] The data interaction module is used for querying related data in a twin database according to the correspondence between a process and data at regular time intervals.

[0099] The geometric model library is used to establish a rare earth virtual workshop and visualize the rare earth virtual workshop; and the rare earth virtual workshop is established by using modeling software.

[0100] The user interaction module is used to control the scene view angle during user inspection and demonstrate animation.

[0101] The scene switching module is used to switch the rare earth virtual workshop during user inspection.

[0102] The digital twin service system M3 is used to:

[0103] Obtain process production index historical data.

[0104] Optimize an extraction mechanism model according to the process production index historical data by using a parameter optimization algorithm to obtain a data-driven model; the extraction mechanism model is a mathematical model representing the extraction mechanism of rare earth.

[0105] Obtain user-input process information.

[0106] Predict the component content of each element in a process product according to the process information by using the data-driven model.

[0107] Determine whether to perform early warning according to the component content of each element.

[0108] In some embodiments, the digital twin service system further includes a process optimization module used to:

[0109] Obtain an optimization strategy by using an optimization control algorithm according to a user-set target; the optimization control algorithm adopts an extraction process reagent amount optimization control method with static setting and dynamic compensation.

[0110] In the above, the rare earth production control information system M1 is used to control a production process, which can be controlled by using the optimization strategy, and the optimization strategy is used as a control instruction and executed by the basic control module.

[0111] The digital twin service system in this embodiment can further include a model updating module used to:

[0112] Calculate the error between the predicted component content and actual process index data to obtain error content; and then adjust the data-driven model by using the error content by using a parameter optimization algorithm.

[0113] In addition, the rare earth production process virtual inspection and process simulation system provided in this embodiment further includes a twin database. The twin database is a bridge between virtual and reality, and is used to store various data, including but not limited to historical data, real-time data, and early warning information.

[0114] In order to interact with the twin database, the rare earth production control information system further comprises a data transmission module.

[0115] Embodiment three:

[0116] The embodiment provides a rare earth production process virtual inspection and process simulation system based on digital twinning. In view of the problem that equipment inspection, working condition judgment, process optimization in the rare earth production process are heavily dependent on manual work, a technical scheme capable of realizing centralized monitoring of process indexes and simulating and predicting process indexes is provided.

[0117] The rare earth production process virtual inspection and process simulation system based on digital twinning provided by the embodiment is composed of a rare earth production control information system, a twin database, a rare earth production virtual plant and a digital twinning service system.

[0118] Each module exchanges data through the twin database, the rare earth production control information system stores the collected real-time data into the twin database and reads the optimization strategy from the twin database; the rare earth production virtual plant reads the data in the twin database in real time, and reflects the actual equipment state, process indexes, component content prediction values to the virtual plant; the digital twinning service system can not only query the data in the database, but also store the predicted component content values and optimization strategies in the twin database. The rare earth production control information system mainly realizes the issuing of equipment control instructions and the collection of field detection data, forms real-time process index data, and stores the real-time data in the twin database,

[0119] The twin database is used for storing field-collected real-time data and process index historical data, and optimized control strategy data and user management information, and the data can be queried by a user and can be used for optimizing a prediction model;

[0120] The rare earth production virtual plant mainly realizes the inspection of production equipment and the monitoring of production indexes, and a good rare earth production virtual plant is integrated into the digital twinning service system.

[0121] The digital twinning service system mainly realizes process index prediction, process optimization, virtual plant inspection, equipment data and process data query, and finally integrates the above functions into a user interface for convenient user operation.

[0122] According to the digital twin rare earth production process virtual inspection and process simulation system, the rare earth production control information system comprises a basic control module, a process detection module, a data transmission module, an optimization control module and a centralized control module; the basic control module and the process detection module are mainly arranged in the production site, and the data transmission module, the optimization control module and the centralized control module are arranged in the central control room. The basic module mainly executes the control instructions issued by the centralized control module, the process detection module mainly detects the equipment state and the process index, and the real-time detection data collected are transmitted to the centralized control module through the data transmission module. The optimization control module and the centralized control module are installed in the computer in the form of software, and the centralized control module can upload the real-time acquisition data through the data transmission module. The optimization control module reads the optimization strategy from the twin database through the data transmission module, and transmits the optimization strategy to the centralized control module after a certain conversion, and the centralized control further issues the optimization controller setting to the basic control module.

[0123] The basic control module is used for realizing loop control and quantitative control of production equipment and execution of optimization strategies. The module mainly comprises a motor frequency converter, a flow pump, a quantitative pump, an electromagnetic valve and a PLC. The motor frequency converter is used for speed regulation of a stirrer, the flow pump and the quantitative pump realize quantitative feeding in an extraction process, the electromagnetic valve is used for controlling feeding and discharging of a liquid, and the PLC converts control instructions obtained from the centralized control module into electrical signals and transmits the electrical signals to various actuators. The module executes control instructions obtained from the centralized control module and the optimization control module to realize automatic control of a production process.

[0124] The process detection module is used for detecting production indexes. Collected data of the module are transmitted back to the centralized control module through the data transmission module to realize collection and management of on-site data. The collected data are real-time data. The module mainly comprises a flow meter, a liquid level meter, a thermometer, a pH meter and component content detection equipment. The flow meter is used for monitoring flow rates of liquid in and out of an extraction process. The liquid level meter is used for monitoring liquid levels in extraction tanks and storage tanks to keep the liquid levels normal. The thermometer and the pH meter are used for configuring washing liquid and extraction liquid meeting process requirements. The component content detection equipment can be an X-ray fluorescence analyzer. The equipment is arranged at various detection stages of a rare earth extraction process. Various detection equipment uploads detection data to the centralized control module through a transmitter and the data transmission module. The module transmits data from the detection equipment to the data transmission module in real time for processing.

[0125] The data transmission module mainly realizes data transmission and data cleaning between the process detection module and the centralized control module of on-site detection data, and data interaction between the twin database and the rare earth production control information system. The data transmission module provides various interfaces for on-site detection equipment and communicates with the twin database in the server in a TCP / IP mode.

[0126] The optimization control module is mainly to convert the data optimization strategy in the twin database into an XMAL file and send it to the centralized control module for optimization control effect. This module reads the optimization strategy data in the twin database, and the optimization strategy of this module is generated by the process optimization module of the digital twin service and stored in the twin database. The optimization control module mainly acts as middleware between the centralized control module and the twin database.

[0127] The centralized control module mainly realizes the centralized control of various devices and the change of controller set values, as well as the centralized monitoring of field collected data. This module can be realized through configuration software. This module can communicate with the twin database through the data transmission module, and can also communicate with the underlying basic control module and process detection module.

[0128] According to the above-mentioned digital twin rare earth production process virtual inspection and process simulation system, the twin database includes user management information database, optimization strategy database, process index historical database, real-time acquisition database, and early warning database. The twin database can be realized by using Mysql database on the server.

[0129] The user management information database is used to store user historical operation data, operator login information, and device maintenance information, and realizes the recording and backtracking of production operation.

[0130] The real-time acquisition database is used to store the real-time detection data and control information collected by the rare earth production control information system. The database data is the key to realize virtual-real interaction. The data in this database will be refreshed according to the set detection period, and the outdated data will be copied to the process index historical database.

[0131] The process index historical database is mainly used to store past real-time data, which can be used for further optimization of data-driven models and can also be used for user backtracking.

[0132] The early warning information database is used to store abnormal process index data. Early warning information includes real-time condition early warning and equipment state early warning. Early warning information is generated by the rare earth production control information system, and is also presented in the rare earth digital twin service system and the rare earth virtual plant.

[0133] According to the above-mentioned rare earth production process virtual inspection and process simulation system based on digital twin, the rare earth production virtual plant includes a data interaction module, a geometric model library, a user interaction module, and a scene switching module. The rare earth production virtual plant can be realized by using Unity, UE4, etc. Scene building software.

[0134] The data interaction module is used to realize the data interaction of the virtual workshop and the twin database and the digital twin service system, and facilitates the visualization of the workshop data. The module mainly queries the related data in the twin database according to the one-to-one correspondence between the process and the data.

[0135] The geometric model library mainly stores the three-dimensional model of the actual workshop, and is used to establish the rare earth production virtual workshop and realize the visualization of the process. The geometric model can be established by using modeling software such as 3DMAX and MAYA according to the actual production equipment.

[0136] The user interaction module is used to realize the free inspection of the user in the virtual scene, the viewing function of the process data and the equipment information. The module can use a script program to realize the control of the scene view angle, the user inspection panel and the process demonstration animation.

[0137] The scene switching module is used to realize the movement of the user in different virtual production workshops, and facilitates the rapid inspection of the user in the whole workshop. The module switches the scene and initializes the data of each scene through a script program.

[0138] According to the above-mentioned rare earth production process virtual inspection and process simulation system based on digital twinning, the digital twin service system mainly includes a production process index prediction module, a virtual workshop inspection module, an equipment information query module, a prediction algorithm and a process optimization algorithm, a process optimization module, a model updating module and a user operation interface.

[0139] The process index prediction module includes a process mechanism model, a data-driven model, and a real-time prediction model. The process mechanism model is mainly a mathematical model established according to the process principle, including separation coefficient models (such as formulas (1)-(3)) and specific formulas (such as formulas (4)-(7)) such as material conservation between extraction stages, and separation coefficient compensation parameter formulas (such as formulas (8)-(9)). The data-driven model is to further improve the accuracy of the model by adjusting the separation coefficient of the model according to the historical process data and the model updating module. After the user inputs the basic process information on site, the real-time prediction model predicts the process indexes of each production link in the next detection period combined with the real-time acquisition data in the twin database. The module will pass the prediction results to the twin database and the process optimization module; the separation coefficient model is as follows, X represents the component content of the water phase, and Y represents the component content of the organic phase. Formula (1) represents the separation coefficient between two adjacent rare earth elements, formula (2) represents the separation coefficient of the extraction element relative to the first element. Formula (3) represents the separation coefficient of the extraction element relative to the last extraction element. The material conservation between extraction stages is used to describe the conservation of the component content of the extraction element between the upper and lower extraction stages. In the material conservation model, f F is the component of the feed liquid, B is the outlet molar fraction of the difficult-to-extract component, is the outlet molar fraction of the easy-to-extract component, and the component content of the A outlet and the B outlet are divided, respectively, to obtain the organic phase composition of the A outlet and the water phase composition of the B outlet, represents the extraction amount, W represents the washing amount, i represents the number of the extraction element, i is from 1 to N, N represents the number of the last extraction element, Y i represents the organic phase component content of the i-th element, i represents the water phase component content of the i-th element, (1+i) / i represents the separation coefficient of the previous element relative to the next element, 1 / i represents the separation coefficient of the i-th element relative to the first element, i / N represents the separation coefficient of the last element relative to the i-th element, [k+1，i] represents the water phase component content of the i-th element at the k+1 stage, [k，i] represents the component content of the i-th element at the k stage of the organic phase, B the product of the difficult-to-extract component outlet molar fraction f′ The product of the content of the organic phase component and the content of the aqueous phase component can establish a material conservation relationship between the extraction amount of the easy-to-extract component and the washing amount of the organic phase. Through the conservation relationship, the content of each component can be inferred step by step. By adding a compensation coefficient K to the separation coefficient, the insufficient extraction in the field can be simulated, and after transformation, the relationship between the aqueous phase and the organic phase can be obtained, and the content of the other component can be obtained.

[0140] Formula (1) represents the separation coefficient of the adjacent two component elements, the relationship between the content of the organic phase component, formula (2) represents the separation coefficient of the i-th element relative to the first element, formula (3) is equivalent to the separation coefficient of the i-th element relative to the last element; Formula (4)-(5) represents the relationship between the aqueous phase components of the adjacent two levels of the extraction section, formula (6)-(7) represents the relationship between the adjacent two levels of the organic phase in the washing section, and formula (8)-(9) represents the relationship between the aqueous phase and the organic phase containing the separation coefficient compensation.

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150] The virtual plant inspection module is a sub-interface of the rare earth virtual plant integrated user operation interface, which facilitates user to view the virtual plant. The implementation of the module is based on the rare earth virtual plant. The module needs to receive user operations in the virtual plant and realize the connection with other modules.

[0151] The equipment information query module mainly reads real-time process indicators, equipment states and early warning information in the twin database, and then users can quickly query information according to needs. The module can use SQL statements to realize the query function of the twin database. In order to facilitate user query, a fuzzy query method can be designed. The input of the module is the process link, equipment number, process indicator and timestamp specified by the user, and the output information is the process indicator, equipment state and process indicator prediction result.

[0152] The prediction algorithm is mainly to first establish an extraction mechanism-based mathematical model offline, then obtain a data-driven model according to historical data of process indicators and a parameter optimization algorithm, and finally predict the component content of Ce, Pr and Nd according to the process information input by the user and provide an out-of-limit warning information. The data-driven model is to optimize the separation coefficient of the mechanism model through the parameter optimization algorithm, and the optimal separation coefficient compensation coefficient is calculated through the historical data of the component content at all levels. The data-driven model mainly includes historical data reading, compensation parameter optimization and prediction result output. According to the set feeding and discharging mode, the initial component content of the feeding level and the optimized mechanism model, the data-driven model calculates the component content at all levels. The parameter optimization algorithm is to optimize the separation coefficient in the mechanism model according to the mechanism model and the component content data. The optimization method is to optimize the compensation parameter K of the separation coefficient by using the improved particle swarm algorithm, and the optimal compensation parameter is calculated.

[0153] The process optimization module is to set the optimization target by the user, and the best optimization strategy is obtained through the optimization algorithm. The optimization algorithm is mainly an optimization control algorithm. Here, a static setting and dynamic compensation extraction process reagent amount optimization control method is adopted. The method mainly sets the feeding flow rate of the reagent amount according to the process, adjusts the feeding flow rate of the reagent amount according to the detected component content value, generates an optimized feeding flow rate setting and forms a control strategy feedback to the twin database.

[0154] The model updating module adjusts the compensation parameters of the model through the model optimization algorithm and the actual process indicators, so that the model is more close to the actual process. The main implementation principle is to take the error between the model prediction result and the actual process indicator data as input, adjust the model compensation coefficient through the optimization algorithm, and improve the prediction accuracy of the model.

[0155] The user operation interface integrates other modules of the digital twin service system, and is convenient for user operation. The user operation interface mainly realizes communication between each module and the twin database, a rare earth virtual workshop and the construction of a user interface.

[0156] Compared with the existing technology, the positive effects obtained by the present application are:

[0157] Through the process indicator real-time prediction model established in the digital twin system, the process indicator real-time prediction module can realize real-time and accurate prediction of the process indicators. At the same time, the production condition changes can be quickly responded, the system warning and optimization data can be provided, the burden of the operator can be reduced, the production process control can be optimized, the production management efficiency can be improved, and the effective utilization of the on-site process data can be realized.

[0158] This invention establishes a virtual workshop for rare earths and establishes a data connection between the actual workshop and the virtual factory, enabling data visualization demonstrations of various production processes and rapid inspection of production equipment.

[0159] This invention optimizes the production process by using a process optimization module in a digital twin service system to optimize the control strategy based on user-defined optimization goals and actual process indicator data.

[0160] Example 4:

[0161] This embodiment provides a virtual inspection and process simulation system for rare earth production based on digital twins, such as... Figure 3 As shown, the system consists of a rare earth production control information system, a digital twin database, and a rare earth digital twin service system. The rare earth production control information service system is used to monitor and control on-site processes, generate real-time process indicator data, and execute optimized control strategies. The digital twin database stores real-time process indicator data from the rare earth production control information system, as well as optimized strategies and process indicator prediction data from the rare earth digital twin service system. The data is categorized and stored according to the business needs of each module. The rare earth digital twin service system can use actual and historical process data to optimize real-time process indicator prediction models, conduct virtual factory inspections, quickly query production equipment data, and generate optimization strategies.

[0162] The rare earth production control information system mentioned above is as follows: Figure 4 As shown, it includes a basic control module, a process detection module, a data transmission module, an optimization control module, and a centralized control module;

[0163] The basic control module is used to implement the basic control functions of the process, execute initial control commands and optimization strategies. This module mainly consists of a controller and corresponding actuators, including a PLC, a fixed displacement pump, a flow pump, and a motor frequency converter.

[0164] The process monitoring module acquires real-time data on process parameters and stores this data in a twin database via a data transmission module. Specific monitoring devices include flow meters, pH meters, thermometers, level gauges, and component content analyzers. The data transmission module connects the monitoring and control devices to the centralized control module; the centralized control module connects the optimization control module to the twin database. This can be achieved using industrial Ethernet and PROFIBUS, transmitting data to the twin database via OPCUA and TCP / IP protocols. The optimization control module converts optimization strategy data into control commands. The centralized control module primarily sets field control parameters and monitors monitoring information; this part can be implemented using configuration software.

[0165] The twin database mentioned above, such as Figure 5 As shown, it includes a user management information database, an optimization strategy database, a real-time process indicator database, a historical process indicator database, and an early warning database.

[0166] User management information is mainly used to record user operations and user login information; the optimization strategy library is used to store optimization strategies generated by the rare earth twin database; the real-time acquisition database is used to store real-time process indicator data collected by the rare earth production control information system; the process indicator historical database is real-time acquisition data transferred according to a set time; the early warning database stores over-limit early warning data issued by the rare earth production control information system. The twin database is implemented using a MySQL database.

[0167] The aforementioned rare earth digital twin service system, such as Figure 6 As shown, it includes an equipment information query module, a process indicator prediction module, a process optimization module, a rare earth production virtual workshop module, and a user interface.

[0168] The equipment information query module primarily enables rapid data query and retrieval from the twin database, mainly using SQL statements to query process data. The process optimization module optimizes control strategies based on optimization algorithms, set optimization objectives, and actual process data. The real-time process indicator data prediction module first establishes a real-time prediction model and then uses real-time collected data to predict process indicator data. The rare earth production virtual workshop visualizes the production process and enables virtual inspection of workshop equipment by building a virtual workshop scene. This module can be implemented using scene building software Unity and modeling software 3ds Max, achieving consistency between virtual and real factory data by querying process data from the twin database. The user interface integrates all modules.

[0169] The real-time process index prediction model, such as Figure 7 As shown, the model mainly consists of a mechanistic model, a data-driven model, and a real-time prediction model. First, a mechanistic model is established based on the rare earth extraction separation funnel method. Then, historical data and predicted values ​​of process parameters from the twin database are compared to identify errors. Next, the separation coefficient is corrected using a differential evolution algorithm to obtain the optimal separation coefficient, thus establishing a data-driven model for the rare earth extraction process. To further ensure the model's accuracy, the real-time data from the twin database is compared with the predicted values ​​from the data-driven model, and the differential evolution algorithm is used again to further adjust the separation coefficient, achieving a more accurate prediction model.

[0170] According to the aforementioned virtual inspection and process simulation system for rare earth production based on digital twins, the construction process of the process index prediction module includes the following steps:

[0171] (1) A process mechanism model was established based on the rare earth extraction separation funnel method.

[0172] (2) Data-driven model is based on the process mechanism model, according to the historical process index data and model optimization algorithm to adjust the separation coefficient of the model, so that the accuracy of the model is improved.

[0173] (3) Real-time prediction model is based on the process mechanism model and data-driven model offline model, and the latest real-time data is dynamically adjusted to reduce the error of the model as much as possible.

[0174] (4) Finally, the real-time prediction model is packaged and integrated into the user operation interface.

[0175] According to the rare earth production process virtual inspection and process simulation system based on digital twinning, the construction process of the rare earth virtual plant comprises the following steps:

[0176] (1) According to the production process, the three-dimensional model of each device is established and stored in the geometric model library.

[0177] (2) Realize the visualization of the production process in the scene building software.

[0178] (3) Write scripts to realize user perspective control, data communication, and human-computer interaction interface.

[0179] (4) The virtual scene is packaged and integrated into the user operation interface.

[0180] The process index data includes detection level component content, stirring motor speed, temperature in the stirring tank, liquid level, PH value, detergent flow, extractant flow, and raw material flow.

[0181] The operation method of the rare earth production process virtual inspection and process simulation system based on digital twinning is as follows:

[0182] (1) Data acquisition and device control: when the production process is running normally, the device will be automatically controlled according to the initial control setting, and the field rare earth production control information system will collect the process production index data of the production site into the twin database.

[0183] (2) Data storage and classification: the twin database is realized through the database file in the server, which mainly stores the real-time process index data, process warning data, and rare earth twin service system optimization strategy data collected in the field. At the same time, according to the needs of different modules and the classification of process categories, more convenient and query data tables can be obtained.

[0184] (3) Virtual scene inspection: The rare earth virtual workshop is built by using virtual scene building software to build device geometric model and control script. The virtual scene realizes one-to-one correspondence with the real scene by reading the data in the twin database. Users can freely view the production process state by operation.

[0185] (4) Rare earth process simulation and process optimization: The rare earth digital twin service platform reads the real-time process index data in the twin database and uses it for real-time process index prediction, realizes the simulation of the rare earth process, and then stores the obtained process index prediction results in the twin database for use by other modules. Process optimization is to optimize the controller set value using historical process data, process optimization algorithm and user set optimization target. Then store the optimized control strategy in the twin database for the rare earth production control information system to execute.

[0186] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts between the various embodiments can be mutually referred to.

[0187] The principles and implementation modes of the present application are described by applying specific examples in this paper. The above description of the embodiments is only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A rare earth production process virtual inspection and process simulation method, characterized in that, The method comprises: acquiring real-time data of a production site; the real-time data comprises process production index data; building a rare earth virtual plant according to a geometric model and a control script of the production site and the real-time data for user inspection; acquiring process production index historical data; optimizing an extraction mechanism model according to the process production index historical data by using a parameter optimization algorithm to obtain a data-driven model; acquiring process information input by a user; predicting component contents of each element in a process finished product according to the process information by using the data-driven model; judging whether to perform early warning according to the component contents of each element; the optimization of the extraction mechanism model according to the process production index historical data by using the parameter optimization algorithm to obtain the data-driven model specifically comprises: optimizing separation coefficients in the extraction mechanism model according to the process production index historical data by using a particle swarm algorithm to obtain the data-driven model; the extraction mechanism model is a mathematical model representing an extraction mechanism of rare earth, comprising separation coefficient models between extraction stages, material conservation formulas and separation coefficient compensation parameter formulas; a product of a difficult-to-extract component outlet molar fraction and a water phase component content and a product of an easy-to-extract component outlet molar fraction and an organic phase component content establish a difficult-to-extract material conservation relationship with a feed stage water phase and organic phase, and an easy-to-extract material conservation relationship with an extraction amount and a washing amount; through these conservation relationships, component contents of each stage are deduced, a compensation coefficient K is added to the separation coefficients to simulate an insufficient extraction condition in the field, and a relationship between the water phase and the organic phase is obtained through deformation to calculate the component content of the other component; then, the particle swarm algorithm is used to optimize the separation coefficients in the extraction mechanism model according to the process production index historical data to obtain the data-driven model; the optimization mode is to optimize the compensation parameter K of the separation coefficients by using the improved particle swarm algorithm to calculate the optimal compensation parameter.

2. A rare earth production process virtual inspection and process simulation system, characterized in that, The system comprises a rare earth production control information system, a rare earth production virtual plant and a digital twin service system; the rare earth production control information system is configured to acquire real-time data of a production site and control a production process; the rare earth production virtual plant is configured to build a rare earth virtual plant according to a geometric model and a control script of the production site and the real-time data for user inspection; the digital twin service system is configured to: acquire process production index historical data; optimize an extraction mechanism model according to the process production index historical data by using a parameter optimization algorithm to obtain a data-driven model; acquire process information input by a user; predict component contents of each element in a process finished product according to the process information by using the data-driven model; judge whether to perform early warning according to the component contents of each element; The extraction mechanism model is a mathematical model representing the extraction mechanism of rare earth, including a separation factor model between extraction stages and a material conservation formula, and a separation factor compensation parameter formula; the product of the outlet molar fraction of a difficult-to-extract component and the content of the aqueous phase component is related to the content of the aqueous phase and the organic phase component of the feed stage, the extraction amount, and the washing amount to establish a difficult-to-extract material conservation relationship; similarly, the product of the outlet of the easily extracted component and the content of the organic phase component is related to the content of the aqueous phase and the organic phase of the feed stage to establish an easily extracted component extraction amount and washing amount material conservation relationship; through these conservation relationships, the component content of each stage is inferred step by step, a compensation coefficient K is added to the separation factor to simulate the insufficient extraction on site, and then the relationship between the aqueous phase and the organic phase is obtained by transformation to obtain the content of the other component; then, the particle swarm algorithm is used to optimize the separation factor in the extraction mechanism model according to the historical data of the process production index to obtain a data-driven model; the optimization method is to optimize the compensation parameter K of the separation factor by using the improved particle swarm algorithm to calculate the optimal compensation parameter.

3. The rare earth production flow virtual inspection and process simulation system of claim 2, wherein, The rare earth production control information system comprises a basic control module and a process detection module. The basic control module is configured to execute control instructions. The basic control module comprises a motor frequency converter, a flow pump, a quantitative pump, an electromagnetic valve, and a PLC; the motor frequency converter is configured to adjust the rotating speed of a stirrer; the flow pump and the quantitative pump are configured to control the quantitative feeding of an extraction process; the electromagnetic valve is configured to control the feeding and discharging of a material liquid; and the PLC is configured to acquire control instructions and transmit the control instructions to the motor frequency converter, the flow pump, the quantitative pump, and the electromagnetic valve. The process detection module is configured to acquire real-time data. The process detection module comprises a flow meter, a liquid level meter, a thermometer, a pH meter, and a component content detection device; the flow meter is configured to monitor the flow of a material liquid in and out of an extraction process; the liquid level meter is configured to monitor the liquid level in an extraction tank and a storage tank; the thermometer and the pH meter are configured to configure a washing liquid and an extraction liquid that meet the temperature and pH requirements; and the component content detection device is arranged at each detection stage of a rare earth extraction process.

4. The rare earth production flow virtual inspection and process simulation system of claim 2, wherein, Further comprising: a twin database configured to store historical data, real-time data, and early warning information.

5. The rare earth production flow virtual inspection and process simulation system of claim 4, wherein, The rare earth production control information system further comprises a data transmission module configured to interact information between the twin database and the rare earth production control information system.

6. The rare earth production flow virtual inspection and process simulation system of claim 4, wherein, The rare earth production virtual workshop comprises a data interaction module, a geometric model library, a user interaction module, and a scene switching module. The data interaction module is configured to query related data in the twin database according to the correspondence between a process and data at regular time intervals. The geometric model library is configured to establish a rare earth virtual workshop and visualize the rare earth virtual workshop; the rare earth virtual workshop is established by using modeling software. The user interaction module is configured to control the scene view angle and demonstrate animation when a user inspects. The scene switching module is configured to switch the rare earth virtual workshop when a user inspects.

7. The rare earth production flow virtual inspection and process simulation system of claim 2, wherein, The digital twin service system further comprises a process optimization module configured to: An optimization strategy is obtained by using an optimization control algorithm according to a target set by a user; the optimization control algorithm adopts an extraction process reagent amount optimization control method of static setting and dynamic compensation.

8. The rare earth production flow virtual inspection and process simulation system of claim 7, wherein, The control production process specifically includes: The production process is controlled according to the optimization strategy.

9. The rare earth production flow virtual tour and process simulation system of claim 2, wherein, The digital twin service system further includes a model updating module configured to: An error content is obtained by calculating an error between the predicted component content and actual process index data; The data-driven model is adjusted by using the error content and a parameter optimization algorithm.

Images