Rare earth production control method and system

Abstract

The present application belongs to the technical field of rare earth production, and relates to a rare earth production control method and system. The method comprises the following steps: obtaining control parameter data of a rare earth production process and corresponding rare earth product quality data; screening the optimal rare earth product quality data, and obtaining the control parameter data corresponding to the group of rare earth product quality data for preprocessing to obtain first control parameters; taking each control parameter in the first control parameters as a control reference parameter of a current production cycle, and outputting multiple groups of feeding amounts at the next moment through a PID control algorithm; averaging the multiple groups of feeding amounts at the next moment obtained based on different characteristic quantities of the same control parameter to obtain an average feeding amount; and weighting and averaging the average feeding amounts corresponding to different control parameters to obtain a final feeding amount at the next moment. The method obtains feeding amount control parameters, can effectively improve the quality of rare earth production, and reduce the carbon content and iron content in the rare earth product.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rare earth production, and discloses a rare earth production control method and system. BACKGROUND

[0002] The rare earth elements are all lanthanides; however, the atomic radii and chemical properties of the 15 elements (atomic numbers 57-71) of the lanthanide series are extremely similar, and the ionic radii only differ by 0.01-0.02 nm, resulting in almost identical solubility, coordination ability, etc. in solvents, which are difficult to separate by traditional chemical precipitation or simple extraction. At present, the separation of rare earths mainly relies on solvent extraction to obtain rare earth compounds with high purity, such as rare earth oxides. Except for rare earth elements with high vapor pressure, most rare earth metals and alloys also need to be obtained by electrolysis of rare earth oxides, such as praseodymium and neodymium metals.

[0003] At present, the production parameters of traditional rare earth electrolysis plants, product process technical indicators and manual experience parameters cannot be organically integrated, and a large amount of production data has not been effectively utilized, and almost all the related parameters in the production process are still controlled based on manual experience; this method depends on manual experience, which is low in efficiency and accuracy.

[0004] How to make good use of existing production data to further improve the production quality of products is an important research direction of rare earth production at present. SUMMARY

[0005] The present application aims to provide a rare earth production control method and system, which solves the problem of low efficiency and unsatisfactory effect when the traditional method relies on manual experience to adjust and improve product quality; the specific scheme is as follows:

[0006] In a first aspect, a rare earth production control method is provided, comprising the following steps:

[0007] Obtaining control parameter data of a rare earth production process and corresponding rare earth product quality data;

[0008] Screening out the optimal rare earth product quality data, and obtaining the control parameter data corresponding to the group of rare earth product quality data to obtain the first control parameter after preprocessing;

[0009] Taking one of the first control parameters as the control reference parameter of the current production cycle, and outputting the next time feeding amount through the PID control algorithm, and repeating the execution until a group of next time feeding amount data is obtained based on each of the first control parameters;

[0010] The average feeding amount is obtained by averaging a plurality of groups of next time feeding amounts obtained based on different characteristic quantities of the same control parameter; and the final next time feeding amount is obtained by weighted averaging of the average feeding amounts corresponding to different control parameters.

[0011] Further, the control parameter data includes current data, temperature data, and feeding amount data; wherein the current data, the temperature data, and the feeding amount data are time series data representing parameter changes in the production process.

[0012] The rare earth product quality data includes, but is not limited to, element carbon content data and iron content data.

[0013] Further, the optimal rare earth product quality data is screened out, specifically:

[0014] The optimal rare earth product quality data is determined based on the element carbon content data and the iron content data of the rare earth product; including:

[0015] The rare earth product quality data satisfying the constraint condition is obtained, with the constraint condition being that the element carbon content data is less than a first threshold value and the iron content data is less than a second threshold value;

[0016] If there is only one group of rare earth product quality data satisfying the constraint condition, the group of rare earth product quality data is taken as the optimal rare earth product quality data.

[0017] If there is no rare earth product quality data satisfying the constraint condition or there are multiple groups of rare earth product quality data satisfying the constraint condition, a quality evaluation coefficient is obtained based on a preset weight, and the rare earth product quality data corresponding to the quality evaluation coefficient with the smallest value is taken as the optimal rare earth product quality data.

[0018] Further, the quality evaluation coefficient is obtained based on the preset weight, specifically as follows:

[0019] q = acon C + bcon Fe

[0020] Wherein, q represents the quality evaluation coefficient, a represents the weight of element carbon, con C represents the element carbon content, b represents the weight of iron, con Fe represents the iron content; and a+b=1, a>b.

[0021] Further, the control parameter data corresponding to the group of rare earth product quality data is preprocessed, specifically:

[0022] Based on the current time series data corresponding to the optimal rare earth product quality data, the peak current, the valley current, the flat current, and the average current are obtained.

[0023] obtaining a peak temperature, a valley temperature, a flat temperature and a mean temperature based on the temperature time series data corresponding to the optimal rare earth product quality data;

[0024] obtaining a peak feeding amount, a valley feeding amount, a flat feeding amount and a mean feeding amount based on the feeding amount time series data corresponding to the optimal rare earth product quality data;

[0025] The first control parameters include a peak current, a valley current, a flat current, a mean current, a peak temperature, a valley temperature, a flat temperature, a mean temperature, a peak feeding amount, a valley feeding amount, a flat feeding amount and a mean feeding amount.

[0026] Further, one of the first control parameters is taken as a control reference parameter of the current production cycle, and a feeding amount at the next moment is output through a PID control algorithm, specifically:

[0027] When the control reference parameter is a current or temperature data, a control amount of the reference parameter is obtained through a PID control method; then the control amount of the reference parameter is converted into a feeding control amount through linear mapping conversion, and the feeding amount at the next moment is obtained based on the planned feeding amount at the next moment and the feeding control amount;

[0028] When the control reference parameter is a feeding amount data, a control amount of the reference parameter is obtained through a PID control method; then the feeding amount at the next moment is obtained based on the control reference parameter and the control amount of the reference parameter.

[0029] Further, a plurality of groups of the feeding amount at the next moment obtained based on different characteristic amounts of the same control parameter are averaged to obtain an average feeding amount, specifically:

[0030] A first average feeding amount is obtained based on four groups of the feeding amount at the next moment obtained when the peak current, the valley current, the flat current and the mean current are respectively taken as the control reference parameter;

[0031] A second average feeding amount is obtained based on four groups of the feeding amount at the next moment obtained when the peak temperature, the valley temperature, the flat temperature and the mean temperature are respectively taken as the control reference parameter;

[0032] A third average feeding amount is obtained based on four groups of the feeding amount at the next moment obtained when the peak feeding amount, the valley feeding amount, the flat feeding amount and the mean feeding amount are respectively taken as the control reference parameter.

[0033] Further, the average feeding amounts corresponding to different control parameters are weighted and averaged to obtain a final feeding amount at the next moment, specifically as follows:

[0034] Q t+1 = w1Q I + w2Q T+w3Q Q

[0035] wherein, Q t+1 represents the final next time feeding amount, w1 represents the weight of the first average feeding amount Q I , w2 represents the weight of the second average feeding amount Q T , and w3 represents the weight of the third average feeding amount Q Q ; and w1+w2+w3=1.

[0036] Further, when the purity of the rare earth product is taken as the control target, w1<w2; when the production efficiency or the morphology control of the rare earth product is taken as the control target, w1>w2.

[0037] In a second aspect, a rare earth production control system is provided, comprising a data storage server and a control center, wherein the data storage server is in communication connection with the control center.

[0038] The data storage server is configured to collect and store production data of each production unit in the rare earth electrolysis plant.

[0039] The control center is configured to obtain the production data in the data storage server and execute the rare earth production control method as described above.

[0040] The present application has the following advantages:

[0041] Based on the existing production data records, the present application screens out the production process control data with the best rare earth quality, and then uses the screened production process control data as the reference parameters to output the next time parameter control by using the PID control method. All the next time parameter controls are converted into feeding amount control parameters through the mapping relationship, and the final next time feeding amount is obtained through the processing of multiple groups of feeding amount control parameters. The method for obtaining the feeding amount control parameters can effectively improve the quality of rare earth production and reduce the carbon content and iron content in the rare earth product. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The flowchart of the rare earth production control method provided by the present application is shown. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0044] At present, the traditional rare earth electrolysis factory production parameters, product process technology index and artificial experience parameters cannot be organically integrated, and a large amount of production data has not been effectively utilized, and almost still based on artificial experience to control the related parameters in the production process; this method depending on artificial experience, one is low efficiency, two is accuracy to be considered.

[0045] Based on the above problems existing in the traditional rare earth production process, the present application expects to provide the following scheme to solve some of the problems.

[0046] Embodiment 1

[0047] This embodiment describes a rare earth production control system, which includes a data storage server and a control center connected in communication. Among them, the data storage server is used to collect and store the production data of each production unit in the rare earth electrolysis factory; the control center is used to obtain the production data in the data storage server, and to realize the parameter control of the current rare earth production process based on the obtained production data, so as to improve the quality of rare earth production and reduce the carbon content and iron content in rare earth products.

[0048] Because the production data of each production unit in the current traditional rare earth electrolysis factory is independent, there is a data island in production and manufacturing, and the equipment monitoring means is backward; it is difficult to integrate and use the related production data to play its value. In this embodiment, by adopting the design of industrial internet of things, the sensors of each production unit are unified to realize the equipment internet of things and data acquisition of each production unit; specifically, the data collected by the sensor can be transmitted to the data storage server through Modbus-TCP or Modbus-ASCII communication protocol, and the data storage server can adopt OPC (OLE for Process Control, OLE for Process Control) server. OPC server is a standard interface software based on industrial automation field, which is used to realize the data interaction and sharing between different manufacturers' equipment and systems. It is the key component of realizing data interconnection in industrial automation field, which solves the problem of equipment integration through standardized interface.

[0049] The control center is integrated with a method for screening data in the data storage server, and the screened data is applied through PID algorithm to provide basis for the control parameter acquisition of the current production; and then improve the quality of rare earth production. For the specific data processing method and control method integrated in the control center, please refer to the scheme content of embodiment 2. It needs to be further explained that the control center is composed of one or more computer devices, and its work mainly depends on the software system which can run on computer hardware devices.

[0050] Embodiment 2

[0051] This embodiment provides a rare earth production control method, which comprises the following steps:Figure 1 The method comprises the following steps:

[0052] The control parameter data of the rare earth production process and the corresponding rare earth product quality data are acquired. In the embodiment, the control parameter data comprises current data, temperature data and charging amount data; wherein, the current data, the temperature data and the charging amount data are time series data representing the parameter changes in the production process. It should be noted that the control parameter data can also include voltage data. Since the current data and the voltage data have strong correlation in the steady state, in order to save the workload of subsequent processing, only the current data is included in the embodiment. If the voltage data is added, the processing flow is the same as the data processing process introduced later in the embodiment. As for the time series data, the related parameters in the embodiment are all composed of data continuously collected by the corresponding sensors in each production unit at a certain data collection period. The collection period can be 1 time per second, or different collection periods can be set according to different stages of rare earth electrolysis. For example, in the initial stage of electrolysis, the data can be collected once every 100-500 ms when the change is relatively fast, and in the stage when the electrolysis tends to be stable, the data can be collected once every 1-5 s. The specific setting method can be determined according to the actual situation of the electrolysis site.

[0053] As for the rare earth product quality data, it includes element carbon content data, iron content data, calcium content data and magnesium content data, etc. Since the element carbon impurities and iron impurities in the rare earth product are the two main indicators for measuring the product quality, in the embodiment, the rare earth product quality data is represented by the carbon content data and the iron content data. Taking praseodymium-neodymium metal product as an example, the national standard is that the element carbon content is less than or equal to 300 ppm, and the iron content is less than or equal to 3000 ppm, which is a qualified product of the first class. In production, under the premise that other impurity elements meet the national standard, the lower the element carbon content and the iron content, the higher the quality of the rare earth product.

[0054] The above-mentioned control parameter data is collected by the corresponding sensor and stored in the data storage server, and the rare earth product quality data is obtained by quality detection of a batch of finished products in a production cycle. That is, a set of rare earth product quality data corresponds to continuous control parameter data of a period. It can be understood that the rare earth product quality data obtained by detection is also stored in the data storage server. In the method disclosed in the embodiment, the control center acquires the above-mentioned data in the data storage server.

[0055] The control center acquires the control parameter data and the rare earth product quality data of multiple production cycles, and screens out the optimal rare earth product quality data, which is specifically determined based on the element carbon content data and the iron content data of the rare earth product to determine the optimal rare earth product quality data, which comprises:

[0056] The quality data of multiple groups of rare earth products are evaluated; and the quality data of the rare earth products satisfying the constraint condition are obtained, with the constraint condition being that the element carbon content data is less than a first threshold value and the iron content data is less than a second threshold value. As mentioned above, in the national standard, the element carbon content is less than or equal to 300 ppm and the iron content is less than or equal to 3000 ppm, which is a first-class qualified product. Based on the production requirements, the first threshold value is generally set to 200 ppm and the second threshold value is generally set to 1500 ppm. It should be noted that, according to the requirements for the quality of rare earth products, the first threshold value and the second threshold value can be set to be lower.

[0057] If there is only one group of rare earth product quality data (i.e., only one production cycle of rare earth product quality data satisfies the constraint condition), the rare earth product quality data of this group is taken as the optimal quality data of the rare earth product.

[0058] If there is no rare earth product quality data satisfying the constraint condition or there are multiple groups of rare earth product quality data satisfying the constraint condition, the quality evaluation coefficient is obtained based on a preset weight. Since the higher the quality of the rare earth product is, the less the element carbon content and the iron content are, the rare earth product quality data corresponding to the quality evaluation coefficient with the smallest value is taken as the optimal quality data of the rare earth product.

[0059] The quality evaluation coefficient is obtained according to the following formula:

[0060] q = acon + bcon C Fe

[0061] wherein q represents the quality evaluation coefficient, a represents the weight of the element carbon, con C represents the element carbon content, b represents the weight of the iron, con Fe represents the iron content; and a + b = 1, a > b.

[0062] Generally, the value range of a is 0.51-0.99 and the value range of b is 0.01-0.49. The weights of the element carbon and the iron are set mainly because the hard and brittle phase formed by carbon and praseodymium-neodymium metal is difficult to eliminate through subsequent processes, and the damage to the magnetic properties and processing properties is irreversible. In addition, iron as a metal impurity can be further effectively controlled through a purification process, and the influence on the material properties is relatively mild. Therefore, a > b.

[0063] After the optimal quality data of the rare earth product is selected through the above steps, the control parameter data corresponding to the group of rare earth product quality data is obtained and preprocessed to obtain the first control parameter.

[0064] ​Based on the optimal rare earth product quality data corresponding to the current time sequence data, the peak current, the valley current, the flat current and the average current are obtained; based on the optimal rare earth product quality data corresponding to the temperature time sequence data, the peak temperature, the valley temperature, the flat temperature and the average temperature are obtained; based on the optimal rare earth product quality data corresponding to the feeding amount time sequence data, the peak feeding amount, the valley feeding amount, the flat feeding amount and the average feeding amount are obtained. It can be understood that the peak data is the data with the maximum value in the corresponding time sequence data; the valley data is the data with the minimum value in the corresponding time sequence data; the average data is the average value of all data values in the corresponding time sequence data; and the flat data is the median value of all data values in the corresponding time sequence data. It should be noted that the feeding amount refers to the feeding amount per unit time, such as the mass of feeding per second.

[0065] The first control parameter includes the peak current, the valley current, the flat current, the average current, the peak temperature, the valley temperature, the flat temperature, the average temperature, the peak feeding amount, the valley feeding amount, the flat feeding amount and the average feeding amount.

[0066] Again, one of the first control parameters is used as the control reference parameter of the current production cycle, and the feeding amount at the next time is output through the PID control algorithm. Specifically:

[0067] When the control reference parameter is current or temperature data, the control amount of the reference parameter is obtained through the PID control method; then the control amount of the reference parameter is converted into the feeding control amount through linear mapping conversion, such as: Q(t) = Q0 + k u(t), where Q0 is the reference feeding amount, k is the conversion coefficient (which needs to be adjusted according to the device range); u(t) is the control amount output by the PID algorithm when the current or temperature data is used as the reference parameter. Then the feeding amount at the next time is obtained based on the planned feeding amount at the next time and the feeding control amount. Here, the control is converted to the control of the feeding amount, which is convenient for subsequent data processing.

[0068] When the control reference parameter is feeding amount data, the control amount of the reference parameter is obtained through the PID control method; then the feeding amount at the next time is obtained based on the control reference parameter and the control amount of the reference parameter.

[0069] It should be noted that the above-mentioned PID control algorithm is mature, and the specific process can be referred to the description of the PID control algorithm in Chinese invention patent ZL202110968477.9. The difference here is mainly the difference in the reference parameter.

[0070] Repeat the above method for obtaining the feeding amount at the next moment until a set of feeding amount data at the next moment is obtained based on each of the first control parameters; that is, the peak current, valley current, average current, average current, peak temperature, valley temperature, average temperature, average temperature, peak feeding amount, valley feeding amount, average feeding amount, and average feeding amount in the first control parameters have all obtained a set of feeding amount data at the next moment through the above method.

[0071] The average feed rate is obtained by averaging multiple sets of feed rates for the next time step obtained based on different characteristic quantities of the same control parameter. Specifically:

[0072] The first average feeding amount is obtained based on four sets of feeding amounts at the next time step, obtained by using peak current, valley current, average current and mean current as control reference parameters respectively; that is, the first average feeding amount is obtained based on current-related data.

[0073] The second average feeding amount is obtained based on four sets of feeding amounts at the next time step, with peak temperature, valley temperature, average temperature and mean temperature as control reference parameters respectively; that is, the second average feeding amount is obtained based on temperature-related data.

[0074] The third average feed rate is obtained based on four sets of feed rates for the next time step, using peak feed rate, valley feed rate, average feed rate, and average feed rate as control benchmark parameters respectively. In other words, the third average feed rate is obtained based on feed rate-related data.

[0075] Finally, a weighted average is calculated for the average feed rate corresponding to different control parameters to obtain the final feed rate for the next time step. The specific formula is as follows:

[0076] Q t+1 =w1Q I +w2Q T +w3Q Q

[0077] Among them, Q t+1 w1 represents the final feed amount at the next moment, and w1 represents the first average feed amount Q. I The weight, w2 represents the second average feed amount Q. t The weight, w3 represents the third average feed amount Q. Q The weights are: w1 + w2 + w3 = 1.

[0078] When the purity of the rare earth product is the control target, the temperature control is more critical, because the volatilization loss and impurity introduction caused by high temperature are difficult to eliminate through subsequent processing, so in this case w1w2; in this case, w1 is 0.3, w2 is 0.4, and w3 is 0.3. When the production efficiency or morphology control of the rare earth product is the control target, the precise regulation of the current is more important, which can effectively avoid the crystallization defects or side reactions caused by improper current, so in this case w1>w2; in this case, w1 is 0.4, w2 is 0.3, and w3 is 0.3.

[0079] Based on the above embodiment, the production process control data with the optimal rare earth quality is screened out based on the existing production data record; the screened production process control data is used as a reference parameter to output the next time parameter control by using the PID control method; and all next time parameter controls are converted into feeding amount control parameters through a mapping relationship, and the final next time feeding amount is obtained through processing of multiple groups of feeding amount control parameters; the method for obtaining the feeding amount control parameter can effectively provide the quality of the rare earth production and reduce the carbon content and iron content in the rare earth product.

[0080] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for controlling rare earth production, characterized in that, Includes the following steps: The process involves acquiring control parameter data for the rare earth production process and corresponding rare earth product quality data. The control parameter data includes current data, temperature data, and feed rate data. The current data, temperature data, and feed rate data are all time-series data representing parameter changes during the production process. The rare earth product quality data includes, but is not limited to, elemental carbon content data and iron content data. The optimal rare earth product quality data is selected, and the corresponding control parameter data is preprocessed to obtain the first control parameter. Specifically, based on the current time series data corresponding to the optimal rare earth product quality data, the peak current, valley current, average current, and average current are obtained; based on the temperature time series data corresponding to the optimal rare earth product quality data, the peak temperature, valley temperature, average temperature, and average temperature are obtained; based on the feeding amount time series data corresponding to the optimal rare earth product quality data, the peak feeding amount, valley feeding amount, average feeding amount, and average feeding amount are obtained; the first control parameter includes peak current, valley current, average current, average current, peak temperature, valley temperature, average temperature, average temperature, peak feeding amount, valley feeding amount, average feeding amount, and average feeding amount. Using one of the first control parameters as the control reference parameter for the current production cycle, the feeding amount for the next moment is output through a PID control algorithm. Specifically: when the control reference parameter is current or temperature data, the control quantity of the reference parameter is obtained through PID control; then, the control quantity of the reference parameter is converted into a feeding control quantity through linear mapping transformation, and the feeding amount for the next moment is obtained based on the planned feeding amount and the feeding control quantity; when the control reference parameter is feeding amount data, the control quantity of the reference parameter is obtained through PID control; then, the feeding amount for the next moment is obtained based on the control reference parameter and the control quantity of the reference parameter; this process is repeated until a set of feeding amount data for the next moment is obtained based on each of the first control parameters. The average feeding amount is obtained by averaging multiple sets of feeding amounts at the next time step based on different characteristic quantities of the same control parameter. Specifically, the first average feeding amount is obtained based on four sets of feeding amounts at the next time step obtained when peak current, valley current, average current, and average current are used as control reference parameters, respectively. The second average feeding amount is obtained based on four sets of feeding amounts at the next time step obtained when peak temperature, valley temperature, average temperature, and average temperature are used as control reference parameters, respectively. The third average feeding amount is obtained based on four sets of feeding amounts at the next time step obtained when peak feeding amount, valley feeding amount, average feeding amount, and average feeding amount are used as control reference parameters, respectively. The final feeding amount at the next time step is obtained by weighted averaging of the average feeding amounts corresponding to different control parameters, as shown in the following formula: in, This indicates the final amount of material to be added at the next moment. Indicates the first average feed amount The weight, Indicates the second average feed amount The weight, Indicates the third average feed amount The weights; and .

2. The rare earth production control method as described in claim 1, characterized in that, The process of selecting the optimal rare earth product quality data is as follows: The optimal quality data for rare earth products is determined based on their elemental carbon and iron content data; including: Using the constraint that the carbon content data is less than the first threshold and the iron content data is less than the second threshold, the quality data of rare earth products that meet the constraint conditions are obtained. If there is only one set of rare earth product quality data that satisfies this constraint, then that set of rare earth product quality data shall be taken as the optimal rare earth product quality data. If there is no rare earth product quality data that meets the constraint, or if there are multiple sets of rare earth product quality data that meet the constraint, then a quality evaluation coefficient is obtained based on a preset weight, and the rare earth product quality data corresponding to the quality evaluation coefficient with the smallest value is taken as the optimal rare earth product quality data.

3. The rare earth production control method as described in claim 2, characterized in that, The quality evaluation coefficient is obtained based on preset weights, as shown in the following formula: in, q This represents the quality evaluation coefficient. α This indicates the weight of element carbon. Indicates the carbon content of the element. β Indicates the weight of iron. Indicates iron content; and α + β =1, α > β .

4. The rare earth production control method as described in claim 1, characterized in that, When the purity of rare earth products is the control target, < When the production efficiency or morphology control of rare earth products is the control objective, > .

5. A rare earth production control system, characterized in that, It includes a data storage server and a control center, wherein the data storage server is communicatively connected to the control center; The data storage server is used to collect and store production data from each production unit in the rare earth electrolysis plant. The control center is used to acquire production data from the data storage server and execute the rare earth production control method as described in any one of claims 1-4.

Images