Intelligent control method and system for semi-autogenous grinding-ball milling grading loop

Abstract

The invention provides an intelligent control method and system for a semi-autogenous grinding-ball milling grading loop. The method comprises the steps that input variables and output variables of the intelligent control system for semi-autogenous grinding-ball milling grading are obtained; on the basis of data in a pre-constructed historical data sample library and a typical working condition sample library of the semi-autogenous grinding-ball milling grading process system, fuzzification processing is conducted on the input variables and the output variables, and the subordination relation between the basic discourse domains of the input variables and the output variables and the corresponding states is obtained; determining the current related working condition based on the basic universe and membership of the input variable and the output variable; determining a target control rule in a plurality of control rules in a pre-constructed core control logic algorithm library in combination with expert knowledge and the working condition; and controlling an execution mechanism in the semi-autogenous grinding-ball milling grading loop to operate based on the target control rule.

Description

Intelligent Control Method and System for Semi-Autogenous Grinding-Ball Mill Classification Loop Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, and in particular to an intelligent control method and system for a semi-autogenous grinding-ball mill grading loop. Background Technology

[0002] The semi-autogenous grinding-ball mill classification process is characterized by numerous influencing factors, complex multi-factor coupling relationships, nonlinearity, large time delays, and difficulty in online detection of key information within the mill. Furthermore, rigorous conditional testing is challenging to obtain the optimal combination of process and control parameters. Current control strategies for the semi-autogenous grinding-ball mill classification process can be mainly divided into model control and rule-based control. Model control involves establishing a mathematical model between control variables (semi-autogenous grinding capacity, semi-autogenous grinding makeup water volume, semi-autogenous grinding mill speed, grinding pump pool level, hydrocyclone feed pump speed, hydrocyclone feed pressure, etc.) and controlled variables (semi-autogenous grinding power, axial pressure, hydrocyclone overflow concentration and particle size, etc.). When the controlled variables deviate from the setpoint, the values ​​of the control variables are calculated to ensure that the controlled variables remain within the setpoint, thus achieving process control. Model control relies on accurate mathematical models and requires a large amount of production data for model training. However, the actual production process has complex grinding mechanisms and numerous interfering factors, making it difficult to establish accurate mathematical models. Rule-based control, by analyzing and summarizing historical production data and manual operating experience, utilizes empirical and qualitative control rules formed during manual operation to ensure that major equipment operates within a stable and efficient range and that key process parameters are within their optimal range. Expert control, an important branch of rule-based control, starts from the control experience of operators and relevant experts, simulating the processes and methods of manual control. It does not require the establishment of precise mathematical models, and is highly robust, less susceptible to interference and parameter changes. It is particularly suitable for the control of nonlinear systems with large time delays, making it an effective control method for semi-autogenous grinding-ball mill processes.

[0003] In recent years, with the development of detection and analysis technologies, various new detection equipment have been applied to the detection and intelligent control of semi-autogenous grinding-ball mill classification loops. These mainly include ore size analysis systems and automatic steel ball adding systems. The ore size analysis system utilizes image segmentation technology to analyze the particle size distribution curve of the feed ore in real time, statistically analyze characteristic curves such as P80 / P50 / P20 of the ore, and can also separately calculate the proportions of coarse, fine, and medium ore. The analysis results can serve as a basis for adjusting the control system and regulating on-site production. The automatic steel ball adding system can automatically add steel balls as needed based on the mill's motion status and product particle size conditions. The automatic ball feeder can synchronously read the amount of ore processed by the system, calculate the cumulative amount of ore during the ball adding interval, and, combined with the unit steel consumption ratio, accurately calculate the required amount of balls to be added by the system, automatically starting and stopping the ball adding function module to achieve uniform ball adding. However, in practice, the results of ore size analysis systems are often only used as a reference indicator, allowing operators to manually adjust control parameters based on the analysis results. Automatic ball feeders can only add steel balls automatically at a fixed ratio according to the system's feed rate, unable to flexibly adjust the ball feeding strategy and quantity based on the semi-autogenous mill's operating status and product fineness, thus failing to stabilize the production process and reduce steel ball consumption. In actual production, mine sites may face challenges such as diverse ore sources and significant differences in ore properties between different ore bodies, making it impossible for the semi-autogenous mill-ball mill process to maintain a consistently stable operating condition. Significant changes in ore properties further increase the difficulty of optimization control.

[0004] Establishing fuzzy control rules in the semi-autogenous grinding-ball mill classification loop is crucial for control effectiveness. The rules for the entire loop can be broadly categorized into semi-autogenous grinding loop control rules, ball mill classification loop control rules, product particle size control rules, and automatic steel ball addition rules. The semi-autogenous grinding loop control is centered on the semi-autogenous grinding mill's operating conditions. Parameters such as the mill's operating power, bearing pressure, and their trends are used to determine the mill's operating conditions, and corresponding control rules are set according to different operating conditions to ensure stable operation and appropriate load conditions. The ball mill classification loop is a multi-objective, multi-variable control problem. Indicators such as the ball mill pump pool level, hydrocyclone feed pressure, hydrocyclone overflow concentration, and hydrocyclone overflow fineness are all control objectives. Therefore, it is necessary to rationally prioritize the satisfaction of multiple objectives and set the order of satisfying each control objective to avoid rule conflicts. The purpose of the automatic steel ball addition rules is to involve the automatic steel ball addition system in the loop control, adjusting the ball addition strategy and quantity in real time according to the operating conditions of the semi-autogenous grinding-ball mill classification loop, thereby stabilizing the loop process and reducing steel ball consumption.

[0005] The existing control rules for semi-autogenous grinding-ball mill classification loops are often based on the complete PID control of throughput, grinding concentration, and water supply. The results of rule control are output in the form of system throughput, semi-autogenous grinding concentration value, and hydrocyclone feed pressure value. It is impossible to directly establish a control relationship with specific equipment (such as feeder, water supply valve, slurry pump, etc.), and control is impossible when the on-site PID control conditions are not available.

[0006] In actual production, mine sites may encounter problems such as diverse ore body sources and significant differences in ore properties between different ore bodies, making it impossible for the semi-autogenous grinding-ball mill classification system to maintain a consistently stable operating condition. When ore properties change significantly, the control rules, input and output variable domains usually need to be adjusted to adapt to the new ore properties and ensure the effectiveness of fuzzy control. In current fuzzy control practices, there is a lack of prediction of ore property changes based on the ore properties of each ore body, mine production scheduling, and ore blending plans. Furthermore, there is a lack of separate fuzzy control rules and related variable domains designed for various typical ore properties and typical operating conditions. Therefore, when ore properties change, the control effect is easily affected, requiring on-site operators to re-explore and extensively adjust the control rules and variable domains based on actual production conditions to ensure effective control. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides an intelligent control method for a semi-autogenous grinding-ball mill classification loop, applied in an intelligent control system for semi-autogenous grinding-ball mill classification. The method includes: obtaining input and output variables of the intelligent control system for semi-autogenous grinding-ball mill classification; the output variables include operating parameters of the semi-autogenous grinding mill and hydrocyclone, and material input parameters, which are related to throughput, ore particle size, water replenishment, and ball addition; and performing fuzzification processing on the input and output variables based on data from a pre-constructed historical data sample library and a typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the input variables... The basic domains of the input and output variables and their corresponding states are defined, with the basic domains and states used to describe the degree of change of the variables. The typical working condition sample library records working condition samples corresponding to different properties of mineral materials. Based on the basic domains and membership relationships of the input and output variables, the current working condition is determined. Combining expert knowledge and the working condition, a target control rule is determined from multiple control rules in the pre-built core control logic algorithm library. The control rule includes some or all of the adjustment information of the output variables and the adjustment order of the output variables. The actuators in the semi-autogenous grinding-ball mill classification loop are controlled to operate based on the target control rule.

[0008] In one embodiment, the input variables include at least one of the following: semi-autogenous mill operating power, rate of change of semi-autogenous mill operating power, semi-autogenous mill axial pressure, rate of change of semi-autogenous mill axial pressure, ball mill operating power, rate of change of ball mill operating power, ball mill axial pressure, rate of change of ball mill axial pressure, ball mill pump tank level, hydrocyclone feed pressure, overflow product concentration, and overflow product fineness.

[0009] In one embodiment, the output variables include at least one of the following: semi-autogenous mill throughput, semi-autogenous mill frequency, semi-autogenous mill makeup water volume, pump pool makeup water volume, slurry pump frequency, semi-autogenous mill feed particle size, semi-autogenous mill ball addition volume, and ball mill ball addition volume.

[0010] In one embodiment, constructing a historical data sample library for the semi-autogenous grinding-ball mill classification process system includes: reading multiple sets of key historical data from the semi-autogenous grinding-ball mill classification process system and performing data analysis processing on each set of key historical data; establishing the historical data sample library for the semi-autogenous grinding-ball mill classification process system based on the key historical data and the corresponding analysis results; wherein, the key historical data includes the detection results from the ore metering unit, the power detection unit of the semi-autogenous grinding mill and the ball mill, the axial pressure detection unit of the semi-autogenous grinding mill and the ball mill, the online analysis system for feed lump size, the liquid level detection unit, the slurry concentration detection unit, and the online analysis system for product fineness.

[0011] In one embodiment, constructing the typical operating condition sample library includes: analyzing historical data in the historical data sample library of the semi-autogenous grinding-ball mill classification process system, the ore properties of each ore body on site, mine production scheduling, and ore blending plans; predicting changes in the ore properties in the semi-autogenous grinding-ball mill classification loop based on the analysis results; dividing the operating status of the semi-autogenous grinding-ball mill classification loop when actually processing ores of different properties, and classifying them into different typical operating conditions based on the changes; configuring matching control rules and parameter affiliation ranges for each of the divided typical operating conditions according to the operating parameters and load ranges of the equipment in the semi-autogenous grinding-ball mill classification loop; and matching and storing the typical operating conditions and their corresponding control rules and parameter affiliation ranges to form the typical operating condition sample library.

[0012] In one embodiment, the input and output variables are fuzzified based on data from a pre-constructed historical data sample library and a typical operating condition sample library of a semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universes of discourse and the membership relationships of the corresponding states of the input and output variables. This includes: fuzzifying the input and output variables based on data from the pre-constructed historical data sample library of the semi-autogenous grinding-ball mill classification process system and combined with obtained field control experience data, to determine candidate basic universes of discourse and candidate membership relationships of the corresponding states of the input and output variables; determining matching target typical operating condition samples based on data from the typical operating condition sample library and the input and output variables, wherein the ore properties involved in the target typical operating condition samples match the ore properties in the current scenario; and determining the basic universe of discourse and membership relationships based on the target typical operating condition samples among the candidate basic universes of discourse and candidate membership relationships.

[0013] In one embodiment, determining the current operating condition based on the basic domain and membership of the input and output variables includes: independently identifying the operating conditions of the semi-autogenous grinding circuit and the ball milling classification circuit in the semi-autogenous grinding-ball milling classification circuit based on the parameter combination states involved in the input and output variables and the basic domain and membership, respectively; the operating conditions of the semi-autogenous grinding circuit include underload, underload trend, overload, overload trend, bloating, and normal state; the operating conditions of the ball milling classification circuit include high liquid level, low liquid level, high overflow concentration, low overflow concentration, coarse product particle size, fine product particle size, and normal state.

[0014] In one embodiment, a core control logic algorithm library is constructed, including: establishing corresponding control rules for each working condition based on the obtained field control experience and expert knowledge, and storing the working conditions and corresponding control rules to form the core control logic algorithm library; different working conditions correspond to different control rules, and the output variables involved in the control rules for adjustment include the semi-autogenous grinding mill throughput, semi-autogenous grinding mill frequency, semi-autogenous grinding concentration, semi-autogenous grinding feed particle size, and semi-autogenous grinding ball addition in the semi-autogenous grinding mill classification loop, as well as one or more of the following in the ball mill classification loop: slurry pump frequency, pump pool water replenishment, and ball mill ball addition.

[0015] In one embodiment, controlling the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the target control rules includes: defuzzifying the data in the target control rules to obtain executable variables that the actuators can read and identify; and controlling the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the corresponding executable variables.

[0016] Another embodiment of the present invention also provides an intelligent control system for a semi-autogenous grinding-ball mill classification loop, comprising: an acquisition module, used to acquire the input variables and output variables of the intelligent control system for the semi-autogenous grinding-ball mill classification, wherein the output variables include operating parameters of the semi-autogenous grinding mill and hydrocyclone, and material input parameters, wherein the material input parameters are related to the throughput, ore particle size, water replenishment, and ball addition; and a fuzzification processing module, used to perform fuzzification processing on the input variables and output variables based on data in a pre-constructed historical data sample library and typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universe of discourse and the membership relation of the corresponding states of the input variables and output variables. The system comprises a basic domain and a state, both used to describe the degree of change of variables. The typical working condition sample library records working condition samples corresponding to different properties of ore. The working condition determination module is used to determine the current working condition based on the basic domain and membership relationship of the input and output variables. The control rule determination module is used to determine the target control rule from multiple control rules in the pre-built core control logic algorithm library by combining expert knowledge and the working condition. The control rule includes some or all of the adjustment information of the output variables and the adjustment order of the output variables. The control module is used to control the actuator in the semi-autogenous grinding-ball mill classification loop to operate based on the target control rule.

[0017] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0018] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 is a flowchart illustrating the intelligent control method for the semi-autogenous grinding-ball milling classification loop in an embodiment of the present invention.

[0021] Figure 2 is a schematic diagram of the application process of the intelligent control method for the semi-autogenous grinding-ball milling classification loop in an embodiment of the present invention.

[0022] Figure 3 is a schematic diagram of the process flow of the intelligent control method for the semi-autogenous grinding-ball milling classification loop in an embodiment of the present invention.

[0023] Figure 4 is a structural block diagram of the intelligent control system for the semi-autogenous grinding-ball milling classification loop in an embodiment of the present invention. Detailed Implementation

[0024] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but these are not intended to limit the scope of the invention.

[0025] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the following description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope of this disclosure will be apparent to those skilled in the art.

[0026] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present disclosure and, together with the general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.

[0027] These and other features of the invention will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0028] It should also be understood that although the invention has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of the invention, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0029] The above and other aspects, features and advantages of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0030] Specific embodiments of the present disclosure are described thereafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure and can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the present disclosure. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but merely to serve as the basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of substantially any suitable detailed structures.

[0031] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in still another embodiment,” all of which may refer to one or more of the same or different embodiments according to this disclosure.

[0032] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0033] The control method based on expert control strategy is to summarize historical production data and manual operation experience, and use fuzzy control algorithm to simplify the operation rules into empirical and qualitative descriptions, so as to ensure that the main equipment is in a stable and efficient operating range and that important process parameters are in the optimal range. Fuzzy rule control has the advantages of relatively simple method, no need for precise model, and strong fault tolerance. The existing semi-autogenous mill-ball mill classification loop control method based on expert control strategy has the following shortcomings: (1) Most of the current semi-autogenous mill-ball mill classification fuzzy control strategies only select the power of semi-autogenous mill, grinding concentration of semi-autogenous mill, shaft pressure of semi-autogenous mill, power of ball mill, shaft pressure of ball mill, liquid level of pump pool, and feed pressure of hydrocyclone as input variables, and select the frequency of feeder, frequency of semi-autogenous mill, water supply of semi-autogenous mill, water supply of pump pool, frequency of hydrocyclone feed pump, and number of hydrocyclone sub-openings as output variables. The scope is not comprehensive enough, the control accuracy is poor, and it is difficult to play the role of stabilizing product fineness and reducing grinding steel consumption.

[0034] (2) Current control rules for semi-autogenous grinding-ball mill classification loops are often based on the complete PID control of throughput, grinding concentration, and water supply. The results of fuzzy control are output in the form of system throughput, semi-autogenous grinding concentration, and hydrocyclone feed pressure, etc., and cannot be directly established with specific equipment (such as feeder, water valve, slurry pump, etc.). Control is not possible when the on-site PID control conditions are not met.

[0035] (3) In the current semi-autogenous grinding-ball mill classification loop fuzzy control system, the control strategy is simple and cannot adapt to the ever-changing field environment. For example, when the ore properties change, the control effect is easily affected. On-site operators need to re-explore and make large-scale adjustments to the control rules and variable domains based on the actual production situation and operating experience in order to ensure the control effect.

[0036] To address the aforementioned technical problems, as shown in Figure 1, this embodiment of the invention provides an intelligent control method for a semi-autogenous grinding-ball mill classification loop, applied in a semi-autogenous grinding-ball mill classification intelligent control system. The method includes: S1: obtaining the input and output variables of the semi-autogenous grinding-ball mill classification intelligent control system. The output variables include operating parameters of the semi-autogenous grinding mill and hydrocyclone, and material input parameters. The material input parameters are related to throughput, ore particle size, water replenishment, and ball addition. S2: performing fuzzification processing on the input and output variables based on data from a pre-constructed historical data sample library and a typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the output... The basic domains of input and output variables and their corresponding states are defined, with the basic domains and states used to describe the degree of change of the variables. The typical working condition sample library records working condition samples corresponding to different properties of ore. S3: Determine the current working condition based on the basic domains and membership relationships of the input and output variables. S4: Combine expert knowledge and the working condition to determine the target control rule from multiple control rules in the pre-built core control logic algorithm library. The control rule includes some or all of the adjustment information of the output variables and the adjustment order of the output variables. S5: Control the actuators in the semi-autogenous grinding-ball mill classification loop to operate based on the target control rule.

[0037] The control method in this embodiment pre-constructs a historical data sample library and a typical sample database of the semi-autogenous grinding-ball mill classification process system. Based on the data in these two data sample libraries, the input and output variables are fuzzified to determine the basic domain characterizing the degree and range of variable change and the corresponding state membership relationships. These states include extremely high, high, normal, low, and extremely low states corresponding to the current parameter, and rising, slowly rising, stable, slowly decreasing, and decreasing states corresponding to parameter changes. Of course, the specific states are not limited to the above descriptions and can include other state parameters. After obtaining the basic domain and the corresponding state membership relationships, the system further determines the operating conditions of various loops, such as the operating conditions of the semi-autogenous grinding mill loop and the ball mill loop. After determining the operating conditions, the system combines the operating conditions with pre-stored expert knowledge, such as expert knowledge obtained through a dynamic expert knowledge base, and can also combine on-site construction experience, to match target control rules in the pre-constructed core control logic algorithm library. There can be one or more target control rules, each matching at least one operating condition, used to optimize the corresponding operating condition by adjusting the output variables. The target control rules include the output variables that need to be adjusted, how these output variables should be adjusted, and the adjustment order of each output variable, i.e., the priority of adjustment. These target control rules are the fuzzy inference results generated by the system. The system can then process these target control rules to obtain data that the actuators in the semi-autogenous grinding-ball milling stage loop can recognize. Afterward, the system controls each actuator to operate according to the corresponding rules, ultimately optimizing the operating conditions.

[0038] The above scheme fully considers all variables and parameters that may affect the operating conditions, as well as environmental factors, ore properties, and differences between different operating conditions in actual mining scenarios. This ensures that the final target control rules are more closely matched to the state of the semi-autogenous grinding-ball mill classification loop. By controlling each actuator to execute the corresponding rules, the optimization of the operating conditions can be effectively ensured. In other words, the scheme proposed in this application not only retains the advantages of previous schemes in practical implementation, but also demonstrates strong adaptability to actual mining scenarios. For example, it is highly adaptable to ore properties, and the overall control effect is more stable and effective. It can significantly improve the system throughput of the semi-autogenous grinding-ball mill classification loop, reduce the steel and electricity consumption per unit ore, and stabilize the concentration of the overflow product.

[0039] In one embodiment, the input variables include at least one of the following: semi-autogenous mill operating power, rate of change of semi-autogenous mill operating power, semi-autogenous mill axial pressure, rate of change of semi-autogenous mill axial pressure, ball mill operating power, rate of change of ball mill operating power, ball mill axial pressure, rate of change of ball mill axial pressure, ball mill pump tank level, hydrocyclone feed pressure, overflow product concentration, and overflow product fineness.

[0040] The output variables include at least one of the following: semi-autogenous mill throughput, semi-autogenous mill frequency, semi-autogenous mill water replenishment, pump pool water replenishment, slurry pump frequency, semi-autogenous mill feed particle size, semi-autogenous mill ball addition, and ball mill ball addition.

[0041] Based on the above, it can be seen that the solution of this embodiment simultaneously incorporates the ore lump size of the semi-autogenous mill, the ball quantity added to the semi-autogenous mill, and the ball quantity added to the ball mill into the control optimization consideration. For example, in this embodiment, by incorporating the lump size analysis system and the automatic ball adding equipment into the control system, the ore lump size of the semi-autogenous mill, the ball quantity added to the semi-autogenous mill, and the ball quantity added to the ball mill are considered as output variables of the system. Therefore, it effectively overcomes the problem of not being able to effectively utilize the ore lump size analysis system and the automatic ball adding equipment to participate in loop control.

[0042] Furthermore, before implementing the scheme of this embodiment, it is necessary to first construct a historical data sample library of the semi-autogenous grinding-ball mill classification process system, including: S6: reading multiple sets of key historical data in the semi-autogenous grinding-ball mill classification process system, and performing data analysis and processing on each set of key historical data; S7: establishing the historical data sample library of the semi-autogenous grinding-ball mill classification process system based on the key historical data and the corresponding analysis results; wherein, the key historical data includes the detection results of the ore metering unit, the power detection unit of the semi-autogenous grinding mill and the ball mill, the axial pressure detection unit of the semi-autogenous grinding mill and the ball mill, the online analysis system for feed lump size, the liquid level detection unit, the slurry concentration detection unit, and the online analysis system for product fineness.

[0043] The ore metering unit, the power detection unit for the semi-autogenous mill and the ball mill, the axial pressure detection unit for the semi-autogenous mill and the ball mill, the online analysis system for feed lump size, the liquid level detection unit, the slurry concentration detection unit, and the online analysis system for product fineness are all detection devices used in actual scenarios to detect and collect different parameters.

[0044] Furthermore, the typical operating condition sample library is constructed, including: S8: analyzing historical data in the historical data sample library of the semi-autogenous grinding-ball mill classification process system, ore properties of various ore bodies on site, mine production scheduling and ore blending plans; S9: predicting changes in feed properties in the semi-autogenous grinding-ball mill classification loop based on the analysis results; S10: dividing the operating status of the semi-autogenous grinding-ball mill classification loop when actually processing ores of different properties into different typical operating conditions based on the changes; S11: configuring matching control rules and parameter affiliation ranges for each of the divided typical operating conditions according to the working parameters and load ranges of the equipment in the semi-autogenous grinding-ball mill classification loop; S12: matching and storing the typical operating conditions and corresponding control rules and parameter affiliation ranges to form the typical operating condition sample library.

[0045] For example, a large amount of key historical data from a historical data sample library is analyzed, combined with ore property analysis of various ore bodies on-site, mine production scheduling, and ore blending plans, to comprehensively predict changes in the feed properties of the semi-autogenous grinding-ball mill classification loop. Then, based on these changes, the operating status of the semi-autogenous grinding-ball mill classification loop in actual production, when processing ores of different properties, is divided into several typical operating conditions. For each typical operating condition, the system analyzes the operating parameters and reasonable load ranges of equipment such as the semi-autogenous grinding mill, ball mill, and hydrocyclone, and customizes corresponding control rules and parameter ranges for each typical operating condition as needed. This allows the system or manual intervention to match the corresponding typical operating conditions based on the real-time data sample library, select appropriate control rules and parameter ranges, improve the versatility and specificity of the intelligent control system when processing ores of different properties, and ensure stable control performance.

[0046] After the sample database is constructed, the system can perform fuzzification processing on the input and output variables based on the data in the two sample databases. As shown in Figure 2, the input and output variables are fuzzified based on the data in the pre-constructed historical data sample database and typical operating condition sample database of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universe of discourse and the membership relationship of the corresponding state of the input and output variables. This includes: S201: Based on the data in the pre-constructed historical data sample database of the semi-autogenous grinding-ball mill classification process system, and combined with the obtained field control experience data, the input and output variables are fuzzified to determine the candidate basic universe of discourse and the candidate membership relationship of the corresponding state of the input and output variables, respectively; S202: Based on the data in the typical operating condition sample database and the input and output variables, a matching target typical operating condition sample is determined, and the ore properties involved in the target typical operating condition sample match the ore properties in the current scenario; S203: Based on the target typical operating condition sample, the basic universe of discourse and the membership relationship are determined from the candidate basic universe of discourse and the candidate membership relationship.

[0047] In this embodiment, the fuzzification effect is that parameters and variables are transformed based on fuzzy language to form fuzzy language variables and fuzzy language parameters. For example, linguistic variables such as "high," "very high," "normal," and "low" are used to describe the state of each variable in the system; and linguistic variables such as "rising," "slowly rising," "stable," "slowly falling," and "falling" are used to describe the rate of change of the corresponding variables. During analysis and processing, all input variables in the semi-autogenous grinding loop and the ball mill classification loop together form a set of input variables. Each different set of input variables corresponds to a working condition of the semi-autogenous grinding loop or the ball mill classification loop. After the system determines the input and output variables, it will use these variables to search and match in the historical information sample database to determine the relevant sample data, and then analyze it. It can also combine the obtained field control experience data to determine the candidate basic domains and the candidate membership relationships of the corresponding states. Next, the system calls the typical working condition sample library. Based on the sample data in the typical working condition sample library, it matches the input and output variables to determine the matching target typical working condition sample. The variable information recorded in this sample matches the input and output variables, so the ore properties in this sample match the ore properties in the current mining scenario. At this point, the system can combine the data in this sample to select the basic domain of discourse and the corresponding state's membership relationship from multiple candidate basic domains of discourse and corresponding state's candidate membership relationships. In this embodiment, the basic domain of discourse of each variable is divided into different linguistic variable intervals by multiple precise values. For example, the state corresponding to "very high" has a corresponding specific numerical interval, and the state corresponding to "low" also has a corresponding specific numerical interval.

[0048] The above embodiments enable the control system to adjust the domains and membership relationships of fuzzy control rules and related variables according to different production conditions. This allows the control system to automatically match or quickly adjust preset dynamic control rules and variable domains and membership relationships based on differences in operating conditions, ensuring the accuracy of the final control rules. For example, based on production data in the historical information sample database and on-site control experience, there are two typical operating conditions on-site: "flotation" and "leaching," depending on the ore body being processed and the ore blending method. Under these two conditions, the basic domains and membership relationships of several key variables in the semi-autogenous grinding-grinding classification loop differ significantly. These mainly include the basic domains of semi-autogenous grinding capacity, semi-autogenous grinding mill power, semi-autogenous grinding mill shaft pressure, number of hydrocyclones open, hydrocyclone feed pressure, overflow product concentration, and overflow product particle size, as well as their membership relationships for different states (extremely high, high, normal, low, extremely low). When the production state read from the real-time information sample database approaches a typical operating condition, the system can automatically or manually adjust the above variables to adapt to different ore properties and production conditions, demonstrating high flexibility.

[0049] In another embodiment, determining the current operating condition based on the basic domain and membership of the input and output variables includes: S301: independently identifying the operating conditions of the semi-autogenous grinding circuit and the ball milling classification circuit in the semi-autogenous grinding-ball milling classification circuit based on the parameter combination states involved in the input and output variables and the basic domain and membership of the input and output variables; the operating conditions of the semi-autogenous grinding circuit include underload, underload trend, overload, overload trend, bloating, and normal state; the operating conditions of the ball milling classification circuit include high liquid level, low liquid level, high overflow concentration, low overflow concentration, coarse product particle size, fine product particle size, and normal state.

[0050] In application, a working condition reasoning and identification system can be built, and then this system can be called to identify and determine the working conditions. For example, the system identifies the actual obtained variable parameters, determines the state combinations of key parameters, and uses the obtained field control experience and data from the dynamic expert knowledge base to summarize and reason about the current operating conditions of the semi-autogenous grinding-ball mill classification loop. This includes independently identifying the states of the semi-autogenous grinding loop and the ball mill classification loop, determining whether the semi-autogenous grinding loop is in an underload, underload trend, overload, overload trend, bloating, or normal state; and determining whether the ball mill classification loop is in a high liquid level, low liquid level, high overflow concentration, low overflow concentration, coarse product particle size, fine product particle size, or normal state. Alternatively, the above reasoning process can also be executed by the control system itself, depending on the specific circumstances.

[0051] In one embodiment, a core control logic algorithm library is also pre-built. The construction process includes: S13: Based on the obtained field control experience and expert knowledge, corresponding control rules are established for each working condition, and the working conditions and corresponding control rules are stored to form the core control logic algorithm library; different working conditions correspond to different control rules. The output variables involved in the control rules for adjustment include the semi-autogenous mill processing capacity, semi-autogenous mill frequency, semi-autogenous mill grinding concentration, semi-autogenous mill feed particle size and semi-autogenous mill ball addition amount of the semi-autogenous mill in the semi-autogenous mill-ball mill classification loop, as well as one or more of the following in the ball mill classification loop: slurry pump frequency, pump pool water addition amount and ball mill ball addition amount.

[0052] For example, the control system in this embodiment, by combining acquired field control experience and expert knowledge, establishes corresponding control rules for each of the deduced semi-autogenous grinding loop operating conditions or ball mill classification loop operating conditions, thereby forming a semi-autogenous grinding-ball mill classification loop control logic algorithm library. The dynamic core control logic algorithm library includes semi-autogenous grinding loop control logic algorithms, ball mill classification loop control logic algorithms, automatic steel ball addition control logic algorithms, and product particle size adjustment control logic algorithms, each corresponding to different process flows of the semi-autogenous grinding-ball mill classification loop, to obtain control rules for different process flows, i.e., control rules for different operating conditions. This logic algorithm library can be dynamic, meaning it can be adjusted in real time by adding expert knowledge and changing the content of the typical operating condition sample library, ensuring the applicability of the core control logic algorithm when the ore properties change. In this embodiment, the control rules often involve multiple input variables and multiple output variables. When establishing the rules, it is necessary to consider the impact of adjusting multiple output variables on the semi-autogenous grinding-ball milling classification process, reasonably determine the priority order of the execution of multiple output variables, and set a reasonable observation time for the execution of non-lowest priority output variables to avoid the adverse effects of the lag of the semi-autogenous grinding-ball milling classification process system on the control effect.

[0053] Furthermore, after the target control rule is determined, when the actuator is controlled in conjunction with the target control rule, the process includes: S501: defuzzifying the data in the target control rule to obtain executable variables that the actuator can read and identify; S502: controlling the actuator in the semi-autogenous grinding-ball milling classification loop to operate based on the corresponding executable variables.

[0054] For example, the target control rules, which are the result of fuzzy inference, are transformed into precise quantities that the actuator can execute. Fuzzy variables in the target control rules, such as the semi-autogenous mill throughput and grinding concentration, are variables that the actuator cannot directly execute. They need to be converted into executable variables such as feeder frequency and semi-autogenous mill water replenishment using methods such as formula transformation. These are then sent to the corresponding actuators in the actuator mechanism shown in Figure 3 for adjustment. In other words, this embodiment provides specific adjustment information and the adjustment sequence for each actuator, allowing for precise control and adjustment of each actuator to ensure the optimization of the final operating condition.

[0055] As shown in Figure 4, another embodiment of the present invention also provides an intelligent control system for a semi-autogenous grinding-ball mill classification loop, comprising: an acquisition module, used to acquire the input variables and output variables of the semi-autogenous grinding-ball mill classification intelligent control system, wherein the output variables include operating parameters of the semi-autogenous grinding mill and hydrocyclone, and material input parameters, wherein the material input parameters are related to the throughput, ore particle size, water replenishment, and ball addition; and a fuzzification processing module, used to perform fuzzification processing on the input variables and output variables based on data in a pre-constructed historical data sample library and typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universe of discourse and corresponding states of the input variables and output variables. Membership relationships, the basic domains of discourse, and states are used to describe the degree of change of variables. The typical working condition sample library records working condition samples corresponding to different properties of ore. The working condition determination module is used to determine the current working condition based on the basic domains of discourse and membership relationships of the input and output variables. The control rule determination module is used to determine the target control rule from multiple control rules in the pre-built core control logic algorithm library by combining expert knowledge and the working condition. The control rule includes some or all of the adjustment information of the output variables and the adjustment order of the output variables. The control module is used to control the actuators in the semi-autogenous grinding-ball mill classification loop to operate based on the target control rule.

[0056] In one embodiment, the input variables include at least one of the following: semi-autogenous mill operating power, rate of change of semi-autogenous mill operating power, semi-autogenous mill axial pressure, rate of change of semi-autogenous mill axial pressure, ball mill operating power, rate of change of ball mill operating power, ball mill axial pressure, rate of change of ball mill axial pressure, ball mill pump tank level, hydrocyclone feed pressure, overflow product concentration, and overflow product fineness.

[0057] In one embodiment, the output variables include at least one of the following: semi-autogenous mill throughput, semi-autogenous mill frequency, semi-autogenous mill makeup water volume, pump pool makeup water volume, slurry pump frequency, semi-autogenous mill feed particle size, semi-autogenous mill ball addition volume, and ball mill ball addition volume.

[0058] In one embodiment, constructing a historical data sample library for the semi-autogenous grinding-ball mill classification process system includes: reading multiple sets of key historical data from the semi-autogenous grinding-ball mill classification process system and performing data analysis processing on each set of key historical data; establishing the historical data sample library for the semi-autogenous grinding-ball mill classification process system based on the key historical data and the corresponding analysis results; wherein, the key historical data includes the detection results from the ore metering unit, the power detection unit of the semi-autogenous grinding mill and the ball mill, the axial pressure detection unit of the semi-autogenous grinding mill and the ball mill, the online analysis system for feed lump size, the liquid level detection unit, the slurry concentration detection unit, and the online analysis system for product fineness.

[0059] In one embodiment, constructing the typical operating condition sample library includes: analyzing historical data in the historical data sample library of the semi-autogenous grinding-ball mill classification process system, the ore properties of each ore body on site, mine production scheduling, and ore blending plans; predicting changes in the ore properties in the semi-autogenous grinding-ball mill classification loop based on the analysis results; dividing the operating status of the semi-autogenous grinding-ball mill classification loop when actually processing ores of different properties, and classifying them into different typical operating conditions based on the changes; configuring matching control rules and parameter affiliation ranges for each of the divided typical operating conditions according to the operating parameters and load ranges of the equipment in the semi-autogenous grinding-ball mill classification loop; and matching and storing the typical operating conditions and their corresponding control rules and parameter affiliation ranges to form the typical operating condition sample library.

[0060] In one embodiment, the input and output variables are fuzzified based on data from a pre-constructed historical data sample library and a typical operating condition sample library of a semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universes of discourse and the membership relationships of the corresponding states of the input and output variables. This includes: fuzzifying the input and output variables based on data from the pre-constructed historical data sample library of the semi-autogenous grinding-ball mill classification process system and combined with obtained field control experience data, to determine candidate basic universes of discourse and candidate membership relationships of the corresponding states of the input and output variables; determining matching target typical operating condition samples based on data from the typical operating condition sample library and the input and output variables, wherein the ore properties involved in the target typical operating condition samples match the ore properties in the current scenario; and determining the basic universe of discourse and membership relationships based on the target typical operating condition samples among the candidate basic universes of discourse and candidate membership relationships.

[0061] In one embodiment, determining the current operating condition based on the basic domain and membership of the input and output variables includes: independently identifying the operating conditions of the semi-autogenous grinding circuit and the ball milling classification circuit in the semi-autogenous grinding-ball milling classification circuit based on the parameter combination states involved in the input and output variables and the basic domain and membership, respectively; the operating conditions of the semi-autogenous grinding circuit include underload, underload trend, overload, overload trend, bloating, and normal state; the operating conditions of the ball milling classification circuit include high liquid level, low liquid level, high overflow concentration, low overflow concentration, coarse product particle size, fine product particle size, and normal state.

[0062] In one embodiment, a core control logic algorithm library is constructed, including: establishing corresponding control rules for each working condition based on the obtained field control experience and expert knowledge, and storing the working conditions and corresponding control rules to form the core control logic algorithm library; different working conditions correspond to different control rules, and the output variables involved in the control rules for adjustment include the semi-autogenous grinding mill throughput, semi-autogenous grinding mill frequency, semi-autogenous grinding concentration, semi-autogenous grinding feed particle size, and semi-autogenous grinding ball addition in the semi-autogenous grinding mill classification loop, as well as one or more of the following in the ball mill classification loop: slurry pump frequency, pump pool water replenishment, and ball mill ball addition.

[0063] In one embodiment, controlling the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the target control rules includes: defuzzifying the data in the target control rules to obtain executable variables that the actuators can read and identify; and controlling the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the corresponding executable variables.

[0064] Another embodiment of the present invention provides an electronic device, comprising: one or more processors; a memory configured to store one or more programs; and when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the intelligent control method for the semi-autogenous grinding-ball milling classification loop as described above.

[0065] Furthermore, one embodiment of the present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the intelligent control method for the semi-autogenous grinding-ball milling classification loop as described above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above-described method embodiments, and will not be repeated here.

[0066] Furthermore, embodiments of the present invention also provide a computer program product, which is tangibly stored on a computer-readable medium and includes computer-readable instructions, which, when executed, cause at least one processor to perform a semi-autogenous grinding-ball milling classification loop intelligent control method as described in the embodiments above.

[0067] It should be noted that the computer storage medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access storage medium (RAM), a read-only storage medium (ROM), an erasable programmable read-only storage medium (EPROM or flash memory), an optical fiber, a portable compact disk read-only storage medium (CD-ROM), an optical storage medium, a magnetic storage medium, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program configured for use by or in connection with an instruction execution system, system, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, antenna, optical fiber, RF, etc., or any suitable combination thereof.

[0068] Furthermore, those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0069] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0070] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction system that implements the functions specified in one or more flowcharts and / or one or more block diagrams.

[0071] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of protection of this application is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this application as described above, which are not provided in detail for the sake of brevity.

Claims

1. A smart control method for a semi-autogenous grinding-ball mill classification loop, characterized in that, The method, applied to a semi-autogenous grinding-ball mill classification intelligent control system, includes: obtaining the input and output variables of the semi-autogenous grinding-ball mill classification intelligent control system; the output variables include operating parameters of the semi-autogenous grinding mill and hydrocyclone, and material input parameters, which are related to throughput, ore particle size, water replenishment, and ball addition; and performing fuzzification processing on the input and output variables based on data from a pre-constructed historical data sample library and typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universe of discourse and the membership of the corresponding states of the input and output variables. The relationships, including the basic domain and state, are used to describe the degree of change of variables. The typical working condition sample library records working condition samples corresponding to different properties of ore. Based on the basic domain and membership relationship of the input and output variables, the current working condition is determined. Combining expert knowledge and the working condition, a target control rule is determined from multiple control rules in the pre-built core control logic algorithm library. The control rule includes some or all of the adjustment information of the output variables and the adjustment order of the output variables. The actuators in the semi-autogenous grinding-ball mill classification loop are controlled to operate based on the target control rule.

2. The intelligent control method for the semi-autogenous grinding-ball mill classification loop according to claim 1, characterized in that, The input variables include at least one of the following: semi-autogenous mill operating power, rate of change of semi-autogenous mill operating power, semi-autogenous mill axial pressure, rate of change of semi-autogenous mill axial pressure, ball mill operating power, rate of change of ball mill operating power, ball mill axial pressure, rate of change of ball mill axial pressure, ball mill pump pool level, hydrocyclone feed pressure, overflow product concentration, and overflow product fineness.

3. The intelligent control method for the semi-autogenous grinding-ball mill classification loop according to claim 1, characterized in that, The output variables include at least one of the following: semi-autogenous mill throughput, semi-autogenous mill frequency, semi-autogenous mill water replenishment, pump pool water replenishment, slurry pump frequency, semi-autogenous mill feed particle size, semi-autogenous mill ball addition, and ball mill ball addition.

4. The intelligent control method for the semi-autogenous grinding-ball milling classification loop according to claim 1, characterized in that, Constructing a historical data sample library for the semi-autogenous grinding-ball mill classification process system includes: reading multiple sets of key historical data from the semi-autogenous grinding-ball mill classification process system and performing data analysis and processing on each set of key historical data; establishing a historical data sample library for the semi-autogenous grinding-ball mill classification process system based on the key historical data and the corresponding analysis results; wherein, the key historical data includes the detection results from the ore metering unit, the power detection unit of the semi-autogenous grinding mill and the ball mill, the axial pressure detection unit of the semi-autogenous grinding mill and the ball mill, the online analysis system for feed lump size, the liquid level detection unit, the slurry concentration detection unit, and the online analysis system for product fineness.

5. The intelligent control method for the semi-autogenous grinding-ball mill classification loop according to claim 1, characterized in that, The construction of the typical operating condition sample library includes: analyzing historical data in the historical data sample library of the semi-autogenous grinding-ball mill classification process system, ore properties of various ore bodies on site, mine production scheduling and ore blending plans; predicting changes in feed properties in the semi-autogenous grinding-ball mill classification loop based on the analysis results; dividing the operating status of the semi-autogenous grinding-ball mill classification loop when actually processing ores of different properties, and classifying them into different typical operating conditions based on the changes; configuring matching control rules and parameter affiliation ranges for each of the divided typical operating conditions according to the operating parameters and load ranges of the equipment in the semi-autogenous grinding-ball mill classification loop; and matching and storing the typical operating conditions and their corresponding control rules and parameter affiliation ranges to form the typical operating condition sample library.

6. The intelligent control method for the semi-autogenous grinding-ball milling classification loop according to claim 1, characterized in that, The input and output variables are fuzzified based on data from a pre-constructed historical data sample library and a typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universe of discourse and the membership relationship of the corresponding states of the input and output variables. This includes: fuzzifying the input and output variables based on data from the pre-constructed historical data sample library of the semi-autogenous grinding-ball mill classification process system and combined with obtained field control experience data to determine candidate basic universes of discourse and candidate membership relationships of the corresponding states of the input and output variables; determining matching target typical operating condition samples based on data from the typical operating condition sample library and the input and output variables, wherein the ore properties involved in the target typical operating condition samples match the ore properties in the current scenario; and determining the basic universe of discourse and membership relationship based on the target typical operating condition samples among the candidate basic universes of discourse and candidate membership relationships.

7. The intelligent control method for the semi-autogenous grinding-ball milling classification loop according to claim 1, characterized in that, The determination of the current operating conditions based on the basic domains and membership relationships of the input and output variables includes: independently identifying the operating conditions of the semi-autogenous grinding circuit and the ball milling classification circuit in the semi-autogenous grinding-ball milling classification circuit based on the parameter combination states involved in the input and output variables and the basic domains and membership relationships; the operating conditions of the semi-autogenous grinding circuit include underload, underload trend, overload, overload trend, bloating, and normal state; the operating conditions of the ball milling classification circuit include high liquid level, low liquid level, high overflow concentration, low overflow concentration, coarse product particle size, fine product particle size, and normal state.

8. The intelligent control method for the semi-autogenous grinding-ball mill classification loop according to claim 1, characterized in that, Constructing a core control logic algorithm library includes: establishing corresponding control rules for each working condition based on acquired field control experience and expert knowledge, and storing the working conditions and corresponding control rules to form the core control logic algorithm library; different working conditions correspond to different control rules, and the output variables involved in the control rules for adjustment include the semi-autogenous grinding mill throughput, semi-autogenous grinding mill frequency, semi-autogenous grinding concentration, semi-autogenous grinding feed particle size, and semi-autogenous grinding ball addition in the semi-autogenous grinding mill classification loop, as well as one or more of the following in the ball mill classification loop: slurry pump frequency, pump pool water replenishment, and ball mill ball addition.

9. The intelligent control method for the semi-autogenous grinding-ball mill classification loop according to claim 1, characterized in that, The control of the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the target control rules includes: defuzzifying the data in the target control rules to obtain executable variables that the actuators can read and identify; and controlling the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the corresponding executable variables.

10. A semi-autogenous grinding-ball mill classification loop intelligent control system, characterized in that, include: The acquisition module is used to acquire the input and output variables of the semi-autogenous grinding-ball mill classification intelligent control system. The output variables include the operating parameters of the semi-autogenous grinding mill and hydrocyclone, and the material input parameters. The material input parameters are related to the throughput, ore particle size, water replenishment, and ball addition. The fuzzification processing module is used to perform fuzzification processing on the input and output variables based on the data in the pre-constructed historical data sample library and typical operating condition sample library of the semi-autogenous grinding-ball mill classification process system, respectively, to obtain the basic universe of discourse and the membership relationship of the corresponding states of the input and output variables. The basic universe of discourse and the states are used to describe the degree of change of the variables. The typical operating condition sample library records operating condition samples corresponding to different properties of ore. The operating condition determination module is used to determine the current operating condition based on the basic domains and membership relationships of the input and output variables; the control rule determination module is used to determine the target control rule from multiple control rules in a pre-built core control logic algorithm library by combining expert knowledge and the operating condition, wherein the control rule includes some or all of the adjustment information of the output variables and the adjustment order of the output variables; the control module is used to control the actuators in the semi-autogenous grinding-ball milling classification loop to operate based on the target control rule.

Images